arXiv Daily Digest - 2026-01-22

We study positional encodings for multi-view transformers that process tokens from a set of posed input images, and seek a mechanism that encodes patches uniquely, allows SE(3)-invariant attention with multi-frequency similarity, and can be adaptive to the geometry of the underlying scene. We find that prior (absolute or relative) encoding schemes for multi-view attention do not meet the above desiderata, and present RayRoPE to address this gap. RayRoPE represents patch positions based on associated rays but leverages a predicted point along the ray instead of the direction for a geometry-aware encoding. To achieve SE(3) invariance, RayRoPE computes query-frame projective coordinates for computing multi-frequency similarity. Lastly, as the 'predicted' 3D point along a ray may not be precise, RayRoPE presents a mechanism to analytically compute the expected position encoding under uncertainty. We validate RayRoPE on the tasks of novel-view synthesis and stereo depth estimation and show that it consistently improves over alternate position encoding schemes (e.g. 15% relative improvement on LPIPS in CO3D). We also show that RayRoPE can seamlessly incorporate RGB-D input, resulting in even larger gains over alternatives that cannot positionally encode this information.

Large language models (LLMs) are being increasingly integrated into legal applications, including judicial decision support, legal practice assistance, and public-facing legal services. While LLMs show strong potential in handling legal knowledge and tasks, their deployment in real-world legal settings raises critical concerns beyond surface-level accuracy, involving the soundness of legal reasoning processes and trustworthy issues such as fairness and reliability. Systematic evaluation of LLM performance in legal tasks has therefore become essential for their responsible adoption. This survey identifies key challenges in evaluating LLMs for legal tasks grounded in real-world legal practice. We analyze the major difficulties involved in assessing LLM performance in the legal domain, including outcome correctness, reasoning reliability, and trustworthiness. Building on these challenges, we review and categorize existing evaluation methods and benchmarks according to their task design, datasets, and evaluation metrics. We further discuss the extent to which current approaches address these challenges, highlight their limitations, and outline future research directions toward more realistic, reliable, and legally grounded evaluation frameworks for LLMs in legal domains.

We study the problem of estimating causal effects under hidden confounding in the following unpaired data setting: we observe some covariates $X$ and an outcome $Y$ under different experimental conditions (environments) but do not observe them jointly; we either observe $X$ or $Y$. Under appropriate regularity conditions, the problem can be cast as an instrumental variable (IV) regression with the environment acting as a (possibly high-dimensional) instrument. When there are many environments but only a few observations per environment, standard two-sample IV estimators fail to be consistent. We propose a GMM-type estimator based on cross-fold sample splitting of the instrument-covariate sample and prove that it is consistent as the number of environments grows but the sample size per environment remains constant. We further extend the method to sparse causal effects via $\ell_1$-regularized estimation and post-selection refitting.

Large language models (LLMs) have achieved impressive proficiency in basic arithmetic, rivaling human-level performance on standard numerical tasks. However, little attention has been given to how these models perform when numerical expressions deviate from the prevailing conventions present in their training corpora. In this work, we investigate numerical reasoning across a wide range of numeral scripts and formats. We show that LLM accuracy drops substantially when numerical inputs are rendered in underrepresented scripts or formats, despite the underlying mathematical reasoning being identical. We further demonstrate that targeted prompting strategies, such as few-shot prompting and explicit numeral mapping, can greatly narrow this gap. Our findings highlight an overlooked challenge in multilingual numerical reasoning and provide actionable insights for working with LLMs to reliably interpret, manipulate, and generate numbers across diverse numeral scripts and formatting styles.

Machine learning and artificial intelligence conferences such as NeurIPS and ICML now regularly receive tens of thousands of submissions, posing significant challenges to maintaining the quality and consistency of the peer review process. This challenge is particularly acute for best paper awards, which are an important part of the peer review process, yet whose selection has increasingly become a subject of debate in recent years. In this paper, we introduce an author-assisted mechanism to facilitate the selection of best paper awards. Our method employs the Isotonic Mechanism for eliciting authors' assessments of their own submissions in the form of a ranking, which is subsequently utilized to adjust the raw review scores for optimal estimation of the submissions' ground-truth quality. We demonstrate that authors are incentivized to report truthfully when their utility is a convex additive function of the adjusted scores, and we validate this convexity assumption for best paper awards using publicly accessible review data of ICLR from 2019 to 2023 and NeurIPS from 2021 to 2023. Crucially, in the special case where an author has a single quota -- that is, may nominate only one paper -- we prove that truthfulness holds even when the utility function is merely nondecreasing and additive. This finding represents a substantial relaxation of the assumptions required in prior work. For practical implementation, we extend our mechanism to accommodate the common scenario of overlapping authorship. Finally, simulation results demonstrate that our mechanism significantly improves the quality of papers selected for awards.

We present a methodology for extracting structured risk factors from corporate 10-K filings while maintaining adherence to a predefined hierarchical taxonomy. Our three-stage pipeline combines LLM extraction with supporting quotes, embedding-based semantic mapping to taxonomy categories, and LLM-as-a-judge validation that filters spurious assignments. To evaluate our approach, we extract 10,688 risk factors from S&P 500 companies and examine risk profile similarity across industry clusters. Beyond extraction, we introduce autonomous taxonomy maintenance where an AI agent analyzes evaluation feedback to identify problematic categories, diagnose failure patterns, and propose refinements, achieving 104.7% improvement in embedding separation in a case study. External validation confirms the taxonomy captures economically meaningful structure: same-industry companies exhibit 63% higher risk profile similarity than cross-industry pairs (Cohen's d=1.06, AUC 0.82, p<0.001). The methodology generalizes to any domain requiring taxonomy-aligned extraction from unstructured text, with autonomous improvement enabling continuous quality maintenance and enhancement as systems process more documents.

Retrieval-based systems approximate access to a corpus by exposing only a truncated subset of available evidence. Even when relevant information exists in the corpus, truncation can prevent compatible evidence from co-occurring, leading to failures that are not captured by relevance-based evaluation. This paper studies retrieval from a structural perspective, modeling query answering as a feasibility problem under truncation. We formalize retrieval as a sequence of candidate evidence sets and characterize conditions under which feasibility in the limit implies feasibility at finite retrieval depth. We show that monotone truncation suffices to guarantee finite witnessability for individual queries. For classes of queries, we identify finite generation of witness certificates as the additional condition required to obtain a uniform retrieval bound, and we show that this condition is necessary. We further exhibit sharp counterexamples demonstrating failure under non-monotone truncation, non-finitely-generated query classes, and purely slotwise coverage. Together, these results isolate feasibility preservation as a correctness criterion for retrieval independent of relevance scoring or optimization, and clarify structural limitations inherent to truncation-based retrieval.

Principal component analysis (PCA) is a tool to capture factors that explain variation in data. Across domains, data are now collected across multiple contexts (for example, individuals with different diseases, cells of different types, or words across texts). While the factors explaining variation in data are undoubtedly shared across subsets of contexts, no tools currently exist to systematically recover such factors. We develop multi-context principal component analysis (MCPCA), a theoretical and algorithmic framework that decomposes data into factors shared across subsets of contexts. Applied to gene expression, MCPCA reveals axes of variation shared across subsets of cancer types and an axis whose variability in tumor cells, but not mean, is associated with lung cancer progression. Applied to contextualized word embeddings from language models, MCPCA maps stages of a debate on human nature, revealing a discussion between science and fiction over decades. These axes are not found by combining data across contexts or by restricting to individual contexts. MCPCA is a principled generalization of PCA to address the challenge of understanding factors underlying data across contexts.

Large language models are typically trained by treating text as a single global distribution, often resulting in geographically homogenized behavior. We study metadata conditioning as a lightweight approach for localization, pre-training 31 models (at 0.5B and 1B parameter scales) from scratch on large-scale English news data annotated with verified URLs, country tags, and continent tags, covering 4 continents and 17 countries. Across four controlled experiments, we show that metadata conditioning consistently improves in-region performance without sacrificing cross-region generalization, enables global models to recover localization comparable to region-specific models, and improves learning efficiency. Our ablation studies demonstrate that URL-level metadata alone captures much of the geographic signal, while balanced regional data coverage remains essential, as metadata cannot fully compensate for missing regions. Finally, we introduce a downstream benchmark of 800 localized news MCQs and show that after instruction tuning, metadata conditioned global models achieve accuracy comparable to LLaMA-3.2-1B-Instruct, despite being trained on substantially less data. Together, these results establish metadata conditioning as a practical and compute-efficient approach for localization of language models.

Cervical spine fractures are critical medical conditions requiring precise and efficient detection for effective clinical management. This study explores the viability of 2D projection-based vertebra segmentation for vertebra-level fracture detection in 3D CT volumes, presenting an end-to-end pipeline for automated analysis of cervical vertebrae (C1-C7). By approximating a 3D volume through optimized 2D axial, sagittal, and coronal projections, regions of interest are identified using the YOLOv8 model from all views and combined to approximate the 3D cervical spine area, achieving a 3D mIoU of 94.45 percent. This projection-based localization strategy reduces computational complexity compared to traditional 3D segmentation methods while maintaining high performance. It is followed by a DenseNet121-Unet-based multi-label segmentation leveraging variance- and energy-based projections, achieving a Dice score of 87.86 percent. Strategic approximation of 3D vertebral masks from these 2D segmentation masks enables the extraction of individual vertebra volumes. The volumes are analyzed for fractures using an ensemble of 2.5D Spatio-Sequential models incorporating both raw slices and projections per vertebra for complementary evaluation. This ensemble achieves vertebra-level and patient-level F1 scores of 68.15 and 82.26, and ROC-AUC scores of 91.62 and 83.04, respectively. We further validate our approach through an explainability study that provides saliency map visualizations highlighting anatomical regions relevant for diagnosis, and an interobserver variability analysis comparing our model's performance with expert radiologists, demonstrating competitive results.

Large Language Models (LLMs) have revolutionized intelligent application development. While standalone LLMs cannot perform any actions, LLM agents address the limitation by integrating tools. However, debugging LLM agents is difficult and costly as the field is still in it's early stage and the community is underdeveloped. To understand the bugs encountered during agent development, we present the first comprehensive study of bug types, root causes, and effects in LLM agent-based software. We collected and analyzed 1,187 bug-related posts and code snippets from Stack Overflow, GitHub, and Hugging Face forums, focused on LLM agents built with seven widely used LLM frameworks as well as custom implementations. For a deeper analysis, we have also studied the component where the bug occurred, along with the programming language and framework. This study also investigates the feasibility of automating bug identification. For that, we have built a ReAct agent named BugReAct, equipped with adequate external tools to determine whether it can detect and annotate the bugs in our dataset. According to our study, we found that BugReAct equipped with Gemini 2.5 Flash achieved a remarkable performance in annotating bug characteristics with an average cost of 0.01 USD per post/code snippet.

Estimating task progress requires reasoning over long-horizon dynamics rather than recognizing static visual content. While modern Vision-Language Models (VLMs) excel at describing what is visible, it remains unclear whether they can infer how far a task has progressed from partial observations. To this end, we introduce Progress-Bench, a benchmark for systematically evaluating progress reasoning in VLMs. Beyond benchmarking, we further explore a human-inspired two-stage progress reasoning paradigm through both training-free prompting and training-based approach based on curated dataset ProgressLM-45K. Experiments on 14 VLMs show that most models are not yet ready for task progress estimation, exhibiting sensitivity to demonstration modality and viewpoint changes, as well as poor handling of unanswerable cases. While training-free prompting that enforces structured progress reasoning yields limited and model-dependent gains, the training-based ProgressLM-3B achieves consistent improvements even at a small model scale, despite being trained on a task set fully disjoint from the evaluation tasks. Further analyses reveal characteristic error patterns and clarify when and why progress reasoning succeeds or fails.

We identify a novel phenomenon in language models: benign fine-tuning of frontier models can lead to privacy collapse. We find that diverse, subtle patterns in training data can degrade contextual privacy, including optimisation for helpfulness, exposure to user information, emotional and subjective dialogue, and debugging code printing internal variables, among others. Fine-tuned models lose their ability to reason about contextual privacy norms, share information inappropriately with tools, and violate memory boundaries across contexts. Privacy collapse is a ``silent failure'' because models maintain high performance on standard safety and utility benchmarks whilst exhibiting severe privacy vulnerabilities. Our experiments show evidence of privacy collapse across six models (closed and open weight), five fine-tuning datasets (real-world and controlled data), and two task categories (agentic and memory-based). Our mechanistic analysis reveals that privacy representations are uniquely fragile to fine-tuning, compared to task-relevant features which are preserved. Our results reveal a critical gap in current safety evaluations, in particular for the deployment of specialised agents.

Training deep computer vision models requires manual oversight or hyperparameter tuning of the learning rate (LR) schedule. While existing adaptive optimizers schedule the LR automatically, they suffer from computational and memory overhead, incompatibility with regularization, and suboptimal LR choices. In this work, we introduce the ZENITH (Zero-overhead Evolution using Norm-Informed Training History) optimizer, which adapts the LR using the temporal evolution of the gradient norm. Image classification experiments spanning 6 CNN architectures and 6 benchmarks demonstrate that ZENITH achieves higher test accuracy in lower wall-clock time than baselines. It also yielded superior mAP in object detection, keypoint detection, and instance segmentation on MS COCO using the R-CNN family of models. Furthermore, its compatibility with regularization enables even better generalization.

We investigate intelligent personal assistants (IPAs) accessibility for deaf and hard of hearing (DHH) people who can use their voice in everyday communication. The inability of IPAs to understand diverse accents including deaf speech renders them largely inaccessible to non-signing and speaking DHH individuals. Using an Echo Show, we compare the usability of natural language input via spoken English; with Alexa's automatic speech recognition and a Wizard-of-Oz setting with a trained facilitator re-speaking commands against that of a large language model (LLM)-assisted touch interface in a mixed-methods study. The touch method was navigated through an LLM-powered "task prompter," which integrated the user's history and smart environment to suggest contextually-appropriate commands. Quantitative results showed no significant differences across both spoken English conditions vs LLM-assisted touch. Qualitative results showed variability in opinions on the usability of each method. Ultimately, it will be necessary to have robust deaf-accented speech recognized natively by IPAs.

Vision-Language-Action (VLA) models have shown promise in robot manipulation but often struggle to generalize to new instructions or complex multi-task scenarios. We identify a critical pathology in current training paradigms where goal-driven data collection creates a dataset bias. In such datasets, language instructions are highly predictable from visual observations alone, causing the conditional mutual information between instructions and actions to vanish, a phenomenon we term Information Collapse. Consequently, models degenerate into vision-only policies that ignore language constraints and fail in out-of-distribution (OOD) settings. To address this, we propose BayesianVLA, a novel framework that enforces instruction following via Bayesian decomposition. By introducing learnable Latent Action Queries, we construct a dual-branch architecture to estimate both a vision-only prior $p(a \mid v)$ and a language-conditioned posterior $π(a \mid v, \ell)$. We then optimize the policy to maximize the conditional Pointwise Mutual Information (PMI) between actions and instructions. This objective effectively penalizes the vision shortcut and rewards actions that explicitly explain the language command. Without requiring new data, BayesianVLA significantly improves generalization. Extensive experiments across on SimplerEnv and RoboCasa demonstrate substantial gains, including an 11.3% improvement on the challenging OOD SimplerEnv benchmark, validating the ability of our approach to robustly ground language in action.

AI coding agents are now submitting pull requests (PRs) to software projects, acting not just as assistants but as autonomous contributors. As these agentic contributions are rapidly increasing across real repositories, little is known about how they behave in practice and why many of them fail to be merged. In this paper, we conduct a large-scale study of 33k agent-authored PRs made by five coding agents across GitHub. (RQ1) We first quantitatively characterize merged and not-merged PRs along four broad dimensions: 1) merge outcomes across task types, 2) code changes, 3) CI build results, and 4) review dynamics. We observe that tasks related to documentation, CI, and build update achieve the highest merge success, whereas performance and bug-fix tasks perform the worst. Not-merged PRs tend to involve larger code changes, touch more files, and often do not pass the project's CI/CD pipeline validation. (RQ2) To further investigate why some agentic PRs are not merged, we qualitatively analyze 600 PRs to derive a hierarchical taxonomy of rejection patterns. This analysis complements the quantitative findings in RQ1 by uncovering rejection reasons not captured by quantitative metrics, including lack of meaningful reviewer engagement, duplicate PRs, unwanted feature implementations, and agent misalignment. Together, our findings highlight key socio-technical and human-AI collaboration factors that are critical to improving the success of future agentic workflows.

This work investigates the performance of Large Language Models (LLMs) in generating ABAP code. Despite successful applications of generative AI in many programming languages, there are hardly any systematic analyses of ABAP code generation to date. The aim of the study is to empirically analyze to what extent various LLMs can generate syntactically correct and functional ABAP code, how effectively they use compiler feedback for iterative improvement, and which task types pose special challenges. For this purpose, a benchmark with 180 tasks is conducted, consisting of adapted HumanEval tasks and practical SAP scenarios. The results show significant performance differences between the models: more powerful LLMs achieve success rates of around 75% after several iterations and benefit greatly from compiler feedback, while smaller models perform significantly weaker. Overall, the study highlights the high potential of powerful LLMs for ABAP development processes, especially in iterative error correction.

While large language models (LLMs) are increasingly used to summarize long documents, this trend poses significant challenges in the legal domain, where the factual accuracy of deposition summaries is crucial. Nugget-based methods have been shown to be extremely helpful for the automated evaluation of summarization approaches. In this work, we translate these methods to the user side and explore how nuggets could directly assist end users. Although prior systems have demonstrated the promise of nugget-based evaluation, its potential to support end users remains underexplored. Focusing on the legal domain, we present a prototype that leverages a factual nugget-based approach to support legal professionals in two concrete scenarios: (1) determining which of two summaries is better, and (2) manually improving an automatically generated summary.

Fog and mobile edge computing (MEC) will play a key role in the upcoming fifth generation (5G) mobile networks to support decentralized applications, data analytics and management into the network itself by using a highly distributed compute model. Furthermore, increasing attention is paid to providing user-centric cybersecurity solutions, which particularly require collecting, processing and analyzing significantly large amount of data traffic and huge number of network connections in 5G networks. In this regard, this paper proposes a MEC-oriented solution in 5G mobile networks to detect network anomalies in real-time and in autonomic way. Our proposal uses deep learning techniques to analyze network flows and to detect network anomalies. Moreover, it uses policies in order to provide an efficient and dynamic management system of the computing resources used in the anomaly detection process. The paper presents relevant aspects of the deployment of the proposal and experimental results to show its performance.

Peer review is at the heart of modern science. As submission numbers rise and research communities grow, the decline in review quality is a popular narrative and a common concern. Yet, is it true? Review quality is difficult to measure, and the ongoing evolution of reviewing practices makes it hard to compare reviews across venues and time. To address this, we introduce a new framework for evidence-based comparative study of review quality and apply it to major AI and machine learning conferences: ICLR, NeurIPS and *ACL. We document the diversity of review formats and introduce a new approach to review standardization. We propose a multi-dimensional schema for quantifying review quality as utility to editors and authors, coupled with both LLM-based and lightweight measurements. We study the relationships between measurements of review quality, and its evolution over time. Contradicting the popular narrative, our cross-temporal analysis reveals no consistent decline in median review quality across venues and years. We propose alternative explanations, and outline recommendations to facilitate future empirical studies of review quality.

Diffusion Large Language Models (dLLMs) break the rigid left-to-right constraint of traditional LLMs, enabling token generation in arbitrary orders. Intuitively, this flexibility implies a solution space that strictly supersets the fixed autoregressive trajectory, theoretically unlocking superior reasoning potential for general tasks like mathematics and coding. Consequently, numerous works have leveraged reinforcement learning (RL) to elicit the reasoning capability of dLLMs. In this paper, we reveal a counter-intuitive reality: arbitrary order generation, in its current form, narrows rather than expands the reasoning boundary of dLLMs. We find that dLLMs tend to exploit this order flexibility to bypass high-uncertainty tokens that are crucial for exploration, leading to a premature collapse of the solution space. This observation challenges the premise of existing RL approaches for dLLMs, where considerable complexities, such as handling combinatorial trajectories and intractable likelihoods, are often devoted to preserving this flexibility. We demonstrate that effective reasoning is better elicited by intentionally forgoing arbitrary order and applying standard Group Relative Policy Optimization (GRPO) instead. Our approach, JustGRPO, is minimalist yet surprisingly effective (e.g., 89.1% accuracy on GSM8K) while fully retaining the parallel decoding ability of dLLMs. Project page: https://nzl-thu.github.io/the-flexibility-trap

Learning long-horizon embodied behaviors from synthetic data remains challenging because generated scenes are often physically implausible, language-driven programs frequently "succeed" without satisfying task semantics, and high-level instructions require grounding into executable action sequences. To address these limitations, we introduce V-CAGE, a closed-loop framework for generating robust, semantically aligned manipulation datasets at scale. First, we propose a context-aware instantiation mechanism that enforces geometric consistency during scene synthesis. By dynamically maintaining a map of prohibited spatial areas as objects are placed, our system prevents interpenetration and ensures reachable, conflict-free configurations in cluttered environments. Second, to bridge the gap between abstract intent and low-level control, we employ a hierarchical instruction decomposition module. This decomposes high-level goals (e.g., "get ready for work") into compositional action primitives, facilitating coherent long-horizon planning. Crucially, we enforce semantic correctness through a VLM-based verification loop. Acting as a visual critic, the VLM performs rigorous rejection sampling after each subtask, filtering out "silent failures" where code executes but fails to achieve the visual goal. Experiments demonstrate that V-CAGE yields datasets with superior physical and semantic fidelity, significantly boosting the success rate and generalization of downstream policies compared to non-verified baselines.

Large Language Models (LLMs) are increasingly used for clinical decision support, where hallucinations and unsafe suggestions may pose direct risks to patient safety. These risks are particularly challenging as they often manifest as subtle clinical errors that evade detection by generic metrics, while expert-authored fine-grained rubrics remain costly to construct and difficult to scale. In this paper, we propose a retrieval-augmented multi-agent framework designed to automate the generation of instance-specific evaluation rubrics. Our approach grounds evaluation in authoritative medical evidence by decomposing retrieved content into atomic facts and synthesizing them with user interaction constraints to form verifiable, fine-grained evaluation criteria. Evaluated on HealthBench, our framework achieves a Clinical Intent Alignment (CIA) score of 60.12%, a statistically significant improvement over the GPT-4o baseline (55.16%). In discriminative tests, our rubrics yield a mean score delta ($μ_Δ = 8.658$) and an AUROC of 0.977, nearly doubling the quality separation achieved by GPT-4o baseline (4.972). Beyond evaluation, our rubrics effectively guide response refinement, improving quality by 9.2% (from 59.0% to 68.2%). This provides a scalable and transparent foundation for both evaluating and improving medical LLMs. The code is available at https://anonymous.4open.science/r/Automated-Rubric-Generation-AF3C/.

Large language models have achieved near-expert performance in structured reasoning domains like mathematics and programming, yet their ability to perform compositional multi-hop reasoning in specialized scientific fields remains limited. We propose a bottom-up learning paradigm in which models are grounded in axiomatic domain facts and compose them to solve complex, unseen tasks. To this end, we present a post-training pipeline, based on a combination of supervised fine-tuning and reinforcement learning (RL), in which knowledge graphs act as implicit reward models. By deriving novel reward signals from knowledge graph paths, we provide verifiable, scalable, and grounded supervision that encourages models to compose intermediate axioms rather than optimize only final answers during RL. We validate this approach in the medical domain, training a 14B model on short-hop reasoning paths (1-3 hops) and evaluating its zero-shot generalization to complex multi-hop queries (4-5 hops). Our experiments show that path-derived rewards act as a "compositional bridge", enabling our model to significantly outperform much larger models and frontier systems like GPT-5.2 and Gemini 3 Pro, on the most difficult reasoning tasks. Furthermore, we demonstrate the robustness of our approach to adversarial perturbations against option-shuffling stress tests. This work suggests that grounding the reasoning process in structured knowledge is a scalable and efficient path toward intelligent reasoning.

Transformers trained via Reinforcement Learning (RL) with outcome-based supervision can spontaneously develop the ability to generate intermediate reasoning steps (Chain-of-Thought). Yet the mechanism by which sparse rewards drive gradient descent to discover such systematic reasoning remains poorly understood. We address this by analyzing the gradient flow dynamics of single-layer Transformers on a synthetic graph traversal task that cannot be solved without Chain-of-Thought (CoT) but admits a simple iterative solution. We prove that despite training solely on final-answer correctness, gradient flow drives the model to converge to a structured, interpretable algorithm that iteratively traverses the graph vertex-by-vertex. We characterize the distributional properties required for this emergence, identifying the critical role of "simple examples": instances requiring fewer reasoning steps. When the training distribution places sufficient mass on these simpler instances, the model learns a generalizable traversal strategy that extrapolates to longer chains; when this mass vanishes, gradient-based learning becomes infeasible. We corroborate our theoretical results through experiments on synthetic data and with real-world language models on mathematical reasoning tasks, validating that our theoretical findings carry over to practical settings.

Python is one of the most popular programming languages; as such, projects written in Python involve an increasing number of diverse security vulnerabilities. However, existing state-of-the-art analysis tools for Python only support a few vulnerability types. Hence, there is a need to detect a large variety of vulnerabilities in Python projects. In this paper, we propose the SAGA approach to detect and locate vulnerabilities in Python source code in a versatile way. SAGA includes a source code parser able to extract control- and data-flow information and to represent it as a symbolic control-flow graph, as well as a domain-specific language defining static aspects of the source code and their evolution during graph traversals. We have leveraged this language to define a library of static aspects for integrity, confidentiality, and other security-related properties. We have evaluated SAGA on a dataset of 108 vulnerabilities, obtaining 100% sensitivity and 99.15% specificity, with only one false positive, while outperforming four common security analysis tools. This analysis was performed in less than 31 seconds, i.e., between 2.5 and 512.1 times faster than the baseline tools.

Critical domain knowledge typically resides with few experts, creating organizational bottlenecks in scalability and decision-making. Non-experts struggle to create effective visualizations, leading to suboptimal insights and diverting expert time. This paper investigates how to capture and embed human domain knowledge into AI agent systems through an industrial case study. We propose a software engineering framework to capture human domain knowledge for engineering AI agents in simulation data visualization by augmenting a Large Language Model (LLM) with a request classifier, Retrieval-Augmented Generation (RAG) system for code generation, codified expert rules, and visualization design principles unified in an agent demonstrating autonomous, reactive, proactive, and social behavior. Evaluation across five scenarios spanning multiple engineering domains with 12 evaluators demonstrates 206% improvement in output quality, with our agent achieving expert-level ratings in all cases versus baseline's poor performance, while maintaining superior code quality with lower variance. Our contributions are: an automated agent-based system for visualization generation and a validated framework for systematically capturing human domain knowledge and codifying tacit expert knowledge into AI agents, demonstrating that non-experts can achieve expert-level outcomes in specialized domains.

Agentic Reinforcement Learning (RL) has empowered Large Language Models (LLMs) to utilize tools like Python interpreters for complex problem-solving. However, for parameter-constrained models (e.g., 4B--7B), the exploration phase is often plagued by frequent execution failures, creating noisy trajectories that hinder policy optimization. Under standard outcome-based reward settings, this noise leads to a critical credit assignment issue, where erroneous actions are inadvertently reinforced alongside successful outcomes. Existing mitigations face a dilemma: dense rewards often trigger reward hacking, while supersampling incurs prohibitive computational costs. To address these challenges, we propose CLEANER. Distinct from external filtering methods, CLEANER exploits the model's intrinsic self-correction capabilities to eliminate error-contaminated context directly during data collection. At its core, the Similarity-Aware Adaptive Rollback (SAAR) mechanism autonomously constructs clean, purified trajectories by retrospectively replacing failures with successful self-corrections. Based on semantic similarity, SAAR adaptively regulates replacement granularity from shallow execution repairs to deep reasoning substitutions. By training on these self-purified paths, the model internalizes correct reasoning patterns rather than error-recovery loops. Empirical results on AIME24/25, GPQA, and LiveCodeBench show average accuracy gains of 6%, 3%, and 5% over baselines. Notably, CLEANER matches state-of-the-art performance using only one-third of the training steps, highlighting trajectory purification as a scalable solution for efficient agentic RL. Our models and code are available at GitHub

Metadata of libraries on the Python Package Index (PyPI)-including links to source code repositories and donation platforms-plays a critical role in supporting the transparency, trust, and sustainability of open-source libraries. Yet, many packages lack such metadata, and little is known about the underlying reasons. This paper presents a large-scale empirical study combining two targeted surveys sent to 50,000 PyPI authors and maintainers. We analyze more than 1,400 responses using large language model (LLM)-based topic modeling to uncover key motivations and barriers related to linking repositories and donation platforms. While repository URLs are often linked to foster collaboration, increase transparency, and enable issue tracking, some maintainers omit them due to oversight, laziness, or the perceived irrelevance to their project. Donation platform links are reported to support open source work or receive financial contributions, but are hindered by skepticism, technical friction, and organizational constraints. Cross-cutting challenges-such as outdated links, lack of awareness, and unclear guidance-affect both types of metadata. We further assess the robustness of our topic modeling pipeline across 30 runs (84% lexical and 89% semantic similarity) and validate topic quality with 23 expert raters (Randolph's kappa = 0.55). The study contributes empirical insights into PyPI's metadata practices and provides recommendations for improving them, while also demonstrating the effectiveness of our topic modeling approach for analyzing short-text survey responses.

Despite tremendous improvements in tasks such as image classification, object detection, and segmentation, the recognition of visual relationships, commonly modeled as the extraction of a graph from an image, remains a challenging task. We believe that this mainly stems from the fact that there is no canonical way to approach the visual graph recognition task. Most existing solutions are specific to a problem and cannot be transferred between different contexts out-of-the box, even though the conceptual problem remains the same. With broad applicability and simplicity in mind, in this paper we develop a method, \textbf{Gra}ph Recognition via \textbf{S}ubgraph \textbf{P}rediction (\textbf{GraSP}), for recognizing graphs in images. We show across several synthetic benchmarks and one real-world application that our method works with a set of diverse types of graphs and their drawings, and can be transferred between tasks without task-specific modifications, paving the way to a more unified framework for visual graph recognition.

In this paper, we study the vehicle routing problem with a finite time horizon. In this routing problem, the objective is to maximize the number of customer requests served within a finite time horizon. We present a novel routing network embedding module which creates local node embedding vectors and a context-aware global graph representation. The proposed Markov decision process for the vehicle routing problem incorporates the node features, the network adjacency matrix and the edge features as components of the state space. We incorporate the remaining finite time horizon into the network embedding module to provide a proper routing context to the embedding module. We integrate our embedding module with a policy gradient-based deep Reinforcement Learning framework to solve the vehicle routing problem with finite time horizon. We trained and validated our proposed routing method on real-world routing networks, as well as synthetically generated Euclidean networks. Our experimental results show that our method achieves a higher customer service rate than the existing routing methods. Additionally, the solution time of our method is significantly lower than that of the existing methods.

The ubiquity of Large Language Models (LLMs) is driving a paradigm shift where user convenience supersedes computational efficiency. This article defines the "Plausibility Trap": a phenomenon where individuals with access to Artificial Intelligence (AI) models deploy expensive probabilistic engines for simple deterministic tasks-such as Optical Character Recognition (OCR) or basic verification-resulting in significant resource waste. Through micro-benchmarks and case studies on OCR and fact-checking, we quantify the "efficiency tax"-demonstrating a ~6.5x latency penalty-and the risks of algorithmic sycophancy. To counter this, we introduce Tool Selection Engineering and the Deterministic-Probabilistic Decision Matrix, a framework to help developers determine when to use Generative AI and, crucially, when to avoid it. We argue for a curriculum shift, emphasizing that true digital literacy relies not only in knowing how to use Generative AI, but also on knowing when not to use it.

Multimodal large language models have demonstrated comparable performance to that of radiology trainees on multiple-choice board-style exams. However, to develop clinically useful multimodal LLM tools, high-quality benchmarks curated by domain experts are essential. To curate released and holdout datasets of 100 chest radiographic studies each and propose an artificial intelligence (AI)-assisted expert labeling procedure to allow radiologists to label studies more efficiently. A total of 13,735 deidentified chest radiographs and their corresponding reports from the MIDRC were used. GPT-4o extracted abnormal findings from the reports, which were then mapped to 12 benchmark labels with a locally hosted LLM (Phi-4-Reasoning). From these studies, 1,000 were sampled on the basis of the AI-suggested benchmark labels for expert review; the sampling algorithm ensured that the selected studies were clinically relevant and captured a range of difficulty levels. Seventeen chest radiologists participated, and they marked "Agree all", "Agree mostly" or "Disagree" to indicate their assessment of the correctness of the LLM suggested labels. Each chest radiograph was evaluated by three experts. Of these, at least two radiologists selected "Agree All" for 381 radiographs. From this set, 200 were selected, prioritizing those with less common or multiple finding labels, and divided into 100 released radiographs and 100 reserved as the holdout dataset. The holdout dataset is used exclusively by RSNA to independently evaluate different models. A benchmark of 200 chest radiographic studies with 12 benchmark labels was created and made publicly available https://imaging.rsna.org, with each chest radiograph verified by three radiologists. In addition, an AI-assisted labeling procedure was developed to help radiologists label at scale, minimize unnecessary omissions, and support a semicollaborative environment.

Federated Neural Architecture Search (FedNAS) aims to automate model design for privacy-preserving Federated Learning (FL) but currently faces two critical bottlenecks: unguided supernet training that yields suboptimal models, and costly multi-hour pipelines for post-training subnet discovery. We introduce DeepFedNAS, a novel, two-phase framework underpinned by a principled, multi-objective fitness function that synthesizes mathematical network design with architectural heuristics. Enabled by a re-engineered supernet, DeepFedNAS introduces Federated Pareto Optimal Supernet Training, which leverages a pre-computed Pareto-optimal cache of high-fitness architectures as an intelligent curriculum to optimize shared supernet weights. Subsequently, its Predictor-Free Search Method eliminates the need for costly accuracy surrogates by utilizing this fitness function as a direct, zero-cost proxy for accuracy, enabling on-demand subnet discovery in mere seconds. DeepFedNAS achieves state-of-the-art accuracy (e.g., up to 1.21% absolute improvement on CIFAR-100), superior parameter and communication efficiency, and a substantial ~61x speedup in total post-training search pipeline time. By reducing the pipeline from over 20 hours to approximately 20 minutes (including initial cache generation) and enabling 20-second individual subnet searches, DeepFedNAS makes hardware-aware FL deployments instantaneous and practical. The complete source code and experimental scripts are available at: https://github.com/bostankhan6/DeepFedNAS

Graph Foundation Models (GFMs) have emerged as a frontier in graph learning, which are expected to deliver transferable representations across diverse tasks. However, GFMs remain constrained by in-memory bottlenecks: they attempt to encode knowledge into model parameters, which limits semantic capacity, introduces heavy lossy compression with conflicts, and entangles graph representation with the knowledge in ways that hinder efficient adaptation, undermining scalability and interpretability. In this work,we propose RAG-GFM, a Retrieval-Augmented Generation aided Graph Foundation Model that offloads knowledge from parameters and complements parameterized learning. To externalize graph knowledge, we build a dual-modal unified retrieval module, where a semantic store from prefix-structured text and a structural store from centrality-based motif. To preserve heterogeneous information, we design a dual-view alignment objective that contrasts both modalities to capture both content and relational patterns. To enable efficient downstream adaptation, we perform in-context augmentation to enrich supporting instances with retrieved texts and motifs as contextual evidence. Extensive experiments on five benchmark graph datasets demonstrate that RAG-GFM consistently outperforms 13 state-of-the-art baselines in both cross-domain node and graph classification, achieving superior effectiveness and efficiency.

Promptable segmentation models such as SAM have established a powerful paradigm, enabling strong generalization to unseen objects and domains with minimal user input, including points, bounding boxes, and text prompts. Among these, bounding boxes stand out as particularly effective, often outperforming points while significantly reducing annotation costs. However, current training and evaluation protocols typically rely on synthetic prompts generated through simple heuristics, offering limited insight into real-world robustness. In this paper, we investigate the robustness of promptable segmentation models to natural variations in bounding box prompts. First, we conduct a controlled user study and collect thousands of real bounding box annotations. Our analysis reveals substantial variability in segmentation quality across users for the same model and instance, indicating that SAM-like models are highly sensitive to natural prompt noise. Then, since exhaustive testing of all possible user inputs is computationally prohibitive, we reformulate robustness evaluation as a white-box optimization problem over the bounding box prompt space. We introduce BREPS, a method for generating adversarial bounding boxes that minimize or maximize segmentation error while adhering to naturalness constraints. Finally, we benchmark state-of-the-art models across 10 datasets, spanning everyday scenes to medical imaging. Code - https://github.com/emb-ai/BREPS.

LLMs have advanced tool-using agents for real-world applications, yet they often lead to unexpected behaviors or results. Beyond obvious failures, the subtle issue of "intent deviation" severely hinders reliable evaluation and performance improvement. Existing post-training methods generally leverage either real system samples or virtual data simulated by LLMs. However, the former is costly due to reliance on hand-crafted user requests, while the latter suffers from distribution shift from the real tools in the wild. Additionally, both methods lack negative samples tailored to intent deviation scenarios, hindering effective guidance on preference learning. We introduce RISE, a "Real-to-Virtual" method designed to mitigate intent deviation. Anchoring on verified tool primitives, RISE synthesizes virtual trajectories and generates diverse negative samples through mutation on critical parameters. With synthetic data, RISE fine-tunes backbone LLMs via the two-stage training for intent alignment. Evaluation results demonstrate that data synthesized by RISE achieve promising results in eight metrics covering user requires, execution trajectories and agent responses. Integrating with training, RISE achieves an average 35.28% improvement in Acctask (task completion) and 23.27% in Accintent (intent alignment), outperforming SOTA baselines by 1.20--42.09% and 1.17--54.93% respectively.

Whisper has become the de-facto encoder for extracting general-purpose audio features in large audio-language models, where a 30-second clip is typically represented by 1500 frame features projected into an LLM. In contrast, audio-text embedding models like CLAP-based models have largely relied on alternative audio encoders (e.g., HTS-AT, PaSST), and have not leveraged Whisper effectively. We present WavLink, a compact audio-text embedding model that augments Whisper encoder with a learnable global token, trained jointly with a text encoder. Through a systematic study of design choices, including pretrained text encoders, loss functions, training modes, and data mixtures, we identify configurations that yield state-of-the-art retrieval performance. Our two-stage training recipe across three model sizes, combined with Matryoshka-style supervision, improves scalability, enabling 8x smaller embeddings with minimal performance drop. WavLink also demonstrates competitive performance on AIR-Bench with MCQs and zero-shot classification.

Generative Agent-Based Modeling (GABM) leverages Large Language Models to create autonomous agents that simulate human behavior in social media environments, demonstrating potential for modeling information propagation, influence processes, and network phenomena. While existing frameworks characterize agents through demographic attributes, personality traits, and interests, they lack mechanisms to encode behavioral dispositions toward platform actions, causing agents to exhibit homogeneous engagement patterns rather than the differentiated participation styles observed on real platforms. In this paper, we investigate the role of behavioral traits as an explicit characterization layer to regulate agents' propensities across posting, re-sharing, commenting, reacting, and inactivity. Through large-scale simulations involving 980 agents and validation against real-world social media data, we demonstrate that behavioral traits are essential to sustain heterogeneous, profile-consistent participation patterns and enable realistic content propagation dynamics through the interplay of amplification- and interaction-oriented profiles. Our findings establish that modeling how agents act-not only who they are-is necessary for advancing GABM as a tool for studying social media phenomena.

We expose a critical limitation in current approaches to machine unlearning in language models: despite the apparent success of unlearning algorithms, information about the forgotten data remains linearly decodable from internal representations. To systematically assess this discrepancy, we introduce an interpretable, information-theoretic framework for auditing unlearning using Partial Information Decomposition (PID). By comparing model representations before and after unlearning, we decompose the mutual information with the forgotten data into distinct components, formalizing the notions of unlearned and residual knowledge. Our analysis reveals that redundant information, shared across both models, constitutes residual knowledge that persists post-unlearning and correlates with susceptibility to known adversarial reconstruction attacks. Leveraging these insights, we propose a representation-based risk score that can guide abstention on sensitive inputs at inference time, providing a practical mechanism to mitigate privacy leakage. Our work introduces a principled, representation-level audit for unlearning, offering theoretical insight and actionable tools for safer deployment of language models.

The interoperability of data and intelligence across allied partners and their respective end-user groups is considered a foundational enabler to the collective defense capability--both conventional and hybrid--of NATO countries. Foreign Information Manipulation and Interference (FIMI) and related hybrid activities are conducted across various societal dimensions and infospheres, posing an ever greater challenge to the characterization of threats, sustaining situational awareness, and response coordination. Recent advances in AI have further led to the decreasing cost of AI-augmented trolling and interference activities, such as through the generation and amplification of manipulative content. Despite the introduction of the DISARM framework as a standardized metadata and analytical framework for FIMI, operationalizing it at the scale of social media remains a challenge. We propose a framework-agnostic agent-based operationalization of DISARM to investigate FIMI on social media. We develop a multi-agent pipeline in which specialized agentic AI components collaboratively (1) detect candidate manipulative behaviors, and (2) map these behaviors onto standard DISARM taxonomies in a transparent manner. We evaluated the approach on two real-world datasets annotated by domain practitioners. We demonstrate that our approach is effective in scaling the predominantly manual and heavily interpretive work of FIMI analysis, providing a direct contribution to enhancing the situational awareness and data interoperability in the context of operating in media and information-rich settings.

The relationship between the physical characteristics of the radiation field and biological damage is central to both radiotherapy and radioprotection, yet the link between spatial scales of energy deposition and biological effects remains not entirely understood. To address this, we developed an interpretable deep learning model that predicts cell survival after proton and carbon ion irradiation, leveraging sequential attention to highlight relevant features and provide insight into the contribution of different energy deposition scales. Trained and tested on the PIDE dataset, our model incorporates, beside LET, nanodosimetric and microdosimetric quantities simulated with MC-Startrack and Open-TOPAS, enabling multi-scale characterization. While achieving high predictive accuracy, our approach also emphasizes transparency in decision-making. We demonstrate high accuracy in predicting RBE for in vitro experiments. Multiple scales are utilized concurrently, with no single spatial scale being predominant. Quantities defined at smaller spatial domains generally have a greater influence, whereas the LET plays a lesser role.

Kilometer-scale Earth system models are essential for capturing local climate change. However, these models are computationally expensive and produce petabyte-scale outputs, which limits their utility for applications such as probabilistic risk assessment. Here, we present the Field-Space Autoencoder, a scalable climate emulation framework based on a spherical compression model that overcomes these challenges. By utilizing Field-Space Attention, the model efficiently operates on native climate model output and therefore avoids geometric distortions caused by forcing spherical data onto Euclidean grids. This approach preserves physical structures significantly better than convolutional baselines. By producing a structured compressed field, it serves as a good baseline for downstream generative emulation. In addition, the model can perform zero-shot super-resolution that maps low-resolution large ensembles and scarce high-resolution data into a shared representation. We train a generative diffusion model on these compressed fields. The model can simultaneously learn internal variability from abundant low-resolution data and fine-scale physics from sparse high-resolution data. Our work bridges the gap between the high volume of low-resolution ensemble statistics and the scarcity of high-resolution physical detail.

Large Language Models (LLMs) have proven highly effective in automating software engineering tasks, bridging natural language and code semantics to achieve notable results in code generation and summarization. However, their scale incurs substantial computational costs, making full fine-tuning impractical. Parameter-Efficient Fine-Tuning (PEFT) methods like QLoRA enable efficient specialization with lower resource demands. Recent studies show QLoRA-optimized Large Code Models (LCMs) perform strongly across diverse tasks, yet it remains unclear whether this effectiveness persists when a single model is QLoRA fine-tuned for multiple code-related tasks. The interaction between Multi-task fine-tuning and QLoRA optimization, and how transfer learning affects correctness and quality of generated artifacts, remains largely unexplored. We investigate Multi-task QLoRA fine-tuning across three representative tasks: code generation, translation, and summarization. We evaluate functional correctness through execution-based and similarity-based metrics, complemented by comprehensive code quality analysis--an aspect largely overlooked in prior work. Our findings show that Multi-task QLoRA effectively leverages transfer learning, achieving competitive or superior performance relative to both Single-task QLoRA and Multi-task full fine-tuning. Larger models demonstrate more consistent balance between correctness and quality, whereas smaller models preserve functionality but exhibit a higher incidence of quality-related issues.

Human cognition exhibits strong circadian modulation, yet its influence on high-dimensional semantic behavior remains poorly understood. Using large-scale Reddit data, we quantify time-of-day variation in language use by embedding text into a pretrained transformer model and measuring semantic entropy as an index of linguistic exploration-exploitation, for which we show a robust circadian rhythmicity that could be entrained by seasonal light cues. Distinguishing between local and global semantic entropy reveals a systematic temporal dissociation: local semantic exploration peaks in the morning, reflecting broader exploration of semantic space, whereas global semantic diversity peaks later in the day as submissions accumulate around already established topics, consistent with "rich-get-richer" dynamics. These patterns are not explained by sentiment or affective valence, indicating that semantic exploration captures a cognitive dimension distinct from mood. The observed temporal structure aligns with known diurnal patterns in neuromodulatory systems, suggesting that biological circadian rhythms extend to the semantic domain.

Effective decision-making in the real world depends on memory that is both stable and adaptive: environments change over time, and agents must retain relevant information over long horizons while also updating or overwriting outdated content when circumstances shift. Existing Reinforcement Learning (RL) benchmarks and memory-augmented agents focus primarily on retention, leaving the equally critical ability of memory rewriting largely unexplored. To address this gap, we introduce a benchmark that explicitly tests continual memory updating under partial observability, i.e. the natural setting where an agent must rely on memory rather than current observations, and use it to compare recurrent, transformer-based, and structured memory architectures. Our experiments reveal that classic recurrent models, despite their simplicity, demonstrate greater flexibility and robustness in memory rewriting tasks than modern structured memories, which succeed only under narrow conditions, and transformer-based agents, which often fail beyond trivial retention cases. These findings expose a fundamental limitation of current approaches and emphasize the necessity of memory mechanisms that balance stable retention with adaptive updating. Our work highlights this overlooked challenge, introduces benchmarks to evaluate it, and offers insights for designing future RL agents with explicit and trainable forgetting mechanisms. Code: https://quartz-admirer.github.io/Memory-Rewriting/

Parser-based log compression, which separates static tem- plates from dynamic variables, is a promising approach to exploit the unique structure of log data. However, its perfor- mance on complex production logs is often unsatisfactory. This performance gap coincides with a known degradation in the accuracy of its core log parsing component on such data, motivating our investigation into a foundational yet unverified question: does higher parsing accuracy necessarily lead to better compression ratio? To answer this, we conduct the first empirical study quanti- fying this relationship and find that a higher parsing accuracy does not guarantee a better compression ratio. Instead, our findings reveal that compression ratio is dictated by achiev- ing effective pattern-based grouping and encoding, i.e., the partitioning of tokens into low entropy, highly compressible groups. Guided by this insight, we design DeLog, a novel log com- pressor that implements a Pattern Signature Synthesis mecha- nism to achieve efficient pattern-based grouping. On 16 public and 10 production datasets, DeLog achieves state-of-the-art compression ratio and speed.

Bangla music is enrich in its own music cultures. Now a days music genre classification is very significant because of the exponential increase in available music, both in digital and physical formats. It is necessary to index them accordingly to facilitate improved retrieval. Automatically classifying Bangla music by genre is essential for efficiently locating specific pieces within a vast and diverse music library. Prevailing methods for genre classification predominantly employ conventional machine learning or deep learning approaches. This work introduces a novel music dataset comprising ten distinct genres of Bangla music. For the task of audio classification, we utilize a recurrent neural network (RNN) architecture. Specifically, a Long Short-Term Memory (LSTM) network is implemented to train the model and perform the classification. Feature extraction represents a foundational stage in audio data processing. This study utilizes Mel-Frequency Cepstral Coefficients (MFCCs) to transform raw audio waveforms into a compact and representative set of features. The proposed framework facilitates music genre classification by leveraging these extracted features. Experimental results demonstrate a classification accuracy of 78%, indicating the system's strong potential to enhance and streamline the organization of Bangla music genres.

Graph Neural Networks (GNNs) are neural networks that aim to process graph data, capturing the relationships and interactions between nodes using the message-passing mechanism. GNN quantization has emerged as a promising approach for reducing model size and accelerating inference in resource-constrained environments. Compared to quantization in LLMs, quantizing graph features is more emphasized in GNNs. Inspired by the above, we propose to leverage prompt learning, which manipulates the input data, to improve the performance of quantization-aware training (QAT) for GNNs. To mitigate the issue that prompting the node features alone can only make part of the quantized aggregation result optimal, we introduce Low-Rank Aggregation Prompting (LoRAP), which injects lightweight, input-dependent prompts into each aggregated feature to optimize the results of quantized aggregations. Extensive evaluations on 4 leading QAT frameworks over 9 graph datasets demonstrate that LoRAP consistently enhances the performance of low-bit quantized GNNs while introducing a minimal computational overhead.

Multi-agent systems (MAS) composed of large language models often exhibit improved problem-solving performance despite operating on identical information. In this work, we provide a formal explanation for this phenomenon grounded in operator theory and constrained optimization. We model each agent as enforcing a distinct family of validity constraints on a shared solution state, and show that a MAS implements a factorized composition of constraint-enforcement operators. Under mild conditions, these dynamics converge to invariant solution sets defined by the intersection of agent constraint sets. Such invariant structures are generally not dynamically accessible to a single agent applying all constraints simultaneously, even when expressive capacity and information are identical. We extend this result from exact constraint enforcement to soft constraints via proximal operators, and apply the formalism to contemporary text-based dialog systems.

Large Language Model (LLM)-based agents are widely used in real-world applications such as customer service, web navigation, and software engineering. As these systems become more autonomous and are deployed at scale, understanding why an agent takes a particular action becomes increasingly important for accountability and governance. However, existing research predominantly focuses on \textit{failure attribution} to localize explicit errors in unsuccessful trajectories, which is insufficient for explaining the reasoning behind agent behaviors. To bridge this gap, we propose a novel framework for \textbf{general agentic attribution}, designed to identify the internal factors driving agent actions regardless of the task outcome. Our framework operates hierarchically to manage the complexity of agent interactions. Specifically, at the \textit{component level}, we employ temporal likelihood dynamics to identify critical interaction steps; then at the \textit{sentence level}, we refine this localization using perturbation-based analysis to isolate the specific textual evidence. We validate our framework across a diverse suite of agentic scenarios, including standard tool use and subtle reliability risks like memory-induced bias. Experimental results demonstrate that the proposed framework reliably pinpoints pivotal historical events and sentences behind the agent behavior, offering a critical step toward safer and more accountable agentic systems.

Differential fuzzers detect bugs by executing identical inputs across distinct implementations of the same specification, such as JavaScript interpreters. Validating the outputs requires an oracle and for differential testing of JavaScript, these are constructed manually, making them expensive, time-consuming, and prone to false positives. Worse, when the specification evolves, this manual effort must be repeated. Inspired by the success of agentic systems in other SE domains, this paper introduces SmartOracle. SmartOracle decomposes the manual triage workflow into specialized Large Language Model (LLM) sub-agents. These agents synthesize independently gathered evidence from terminal runs and targeted specification queries to reach a final verdict. For historical benchmarks, SmartOracle achieves 0.84 recall with an 18% false positive rate. Compared to a sequential Gemini 2.5 Pro baseline, it improves triage accuracy while reducing analysis time by 4$\times$ and API costs by 10$\times$. In active fuzzing campaigns, SmartOracle successfully identified and reported previously unknown specification-level issues across major engines, including bugs in V8, JavaScriptCore, and GraalJS. The success of SmartOracle's agentic architecture on Javascript suggests it might be useful other software systems- a research direction we will explore in future work.

AI has revolutionised decision-making across various fields. Yet human judgement remains paramount for high-stakes decision-making. This has fueled explorations of collaborative decision-making between humans and AI systems, aiming to leverage the strengths of both. To explore this dynamic, researchers conduct empirical studies, investigating how humans use AI assistance for decision-making and how this collaboration impacts results. A critical aspect of conducting these studies is the role of participants, often recruited through crowdsourcing platforms. The validity of these studies hinges on the behaviours of the participants, hence effective incentives that can potentially affect these behaviours are a key part of designing and executing these studies. In this work, we aim to address the critical role of incentive design for conducting empirical human-AI decision-making studies, focusing on understanding, designing, and documenting incentive schemes. Through a thematic review of existing research, we explored the current practices, challenges, and opportunities associated with incentive design for human-AI decision-making empirical studies. We identified recurring patterns, or themes, such as what comprises the components of an incentive scheme, how incentive schemes are manipulated by researchers, and the impact they can have on research outcomes. Leveraging the acquired understanding, we curated a set of guidelines to aid researchers in designing effective incentive schemes for their studies, called the Incentive-Tuning Framework, outlining how researchers can undertake, reflect on, and document the incentive design process. By advocating for a standardised yet flexible approach to incentive design and contributing valuable insights along with practical tools, we hope to pave the way for more reliable and generalizable knowledge in the field of human-AI decision-making.

Traditional data masking techniques such as anonymization cannot achieve the expected privacy protection while ensuring data utility for privacy-preserving machine learning. Synthetic data plays an increasingly important role as it generates a large number of training samples and prevents information leakage in real data. The existing methods suffer from the repeating trade-off processes between privacy and utility. We propose a novel framework for differential privacy generation, which employs an Error Feedback Stochastic Gradient Descent(EFSGD) method and introduces a reconstruction loss and noise injection mechanism into the training process. We generate images with higher quality and usability under the same privacy budget as the related work. Extensive experiments demonstrate the effectiveness and generalization of our proposed framework for both grayscale and RGB images. We achieve state-of-the-art results over almost all metrics on three benchmarks: MNIST, Fashion-MNIST, and CelebA.

Modern CI/CD pipelines integrating agent-generated code exhibit a structural failure in responsibility attribution. Decisions are executed through formally correct approval processes, yet no entity possesses both the authority to approve those decisions and the epistemic capacity to meaningfully understand their basis. We define this condition as responsibility vacuum: a state in which decisions occur, but responsibility cannot be attributed because authority and verification capacity do not coincide. We show that this is not a process deviation or technical defect, but a structural property of deployments where decision generation throughput exceeds bounded human verification capacity. We identify a scaling limit under standard deployment assumptions, including parallel agent generation, CI-based validation, and individualized human approval gates. Beyond a throughput threshold, verification ceases to function as a decision criterion and is replaced by ritualized approval based on proxy signals. Personalized responsibility becomes structurally unattainable in this regime. We further characterize a CI amplification dynamic, whereby increasing automated validation coverage raises proxy signal density without restoring human capacity. Under fixed time and attention constraints, this accelerates cognitive offloading in the broad sense and widens the gap between formal approval and epistemic understanding. Additional automation therefore amplifies, rather than mitigates, the responsibility vacuum. We conclude that unless organizations explicitly redesign decision boundaries or reassign responsibility away from individual decisions toward batch- or system-level ownership, responsibility vacuum remains an invisible but persistent failure mode in scaled agent deployments.

Traditional defenses against Deep Leakage (DL) attacks in Federated Learning (FL) primarily focus on obfuscation, introducing noise, transformations or encryption to degrade an attacker's ability to reconstruct private data. While effective to some extent, these methods often still leak high-level information such as class distributions or feature representations, and are frequently broken by increasingly powerful denoising attacks. We propose a fundamentally different perspective on FL defense: framing it as a spoofing problem.We introduce SpooFL (Figure 1), a spoofing-based defense that deceives attackers into believing they have recovered the true training data, while actually providing convincing but entirely synthetic samples from an unrelated task. Unlike prior synthetic-data defenses that share classes or distributions with the private data and thus still leak semantic information, SpooFL uses a state-of-the-art generative model trained on an external dataset with no class overlap. As a result, attackers are misled into recovering plausible yet completely irrelevant samples, preventing meaningful data leakage while preserving FL training integrity. We implement the first example of such a spoofing defense, and evaluate our method against state-of-the-art DL defenses and demonstrate that it successfully misdirects attackers without compromising model performance significantly.

Current evaluation methods for Attributed Question Answering (AQA) suffer from \textit{attribution myopia}: they emphasize verification of isolated statements and their attributions but overlook the global logical integrity of long-form answers. Consequently, Large Language Models (LLMs) often produce factually grounded yet logically incoherent responses with elusive deductive gaps. To mitigate this limitation, we present \textsc{LogicScore}, a unified evaluation framework that shifts the paradigm from local assessment to global reasoning scrutiny. Grounded in Horn Rules, our approach integrates a backward verification mechanism to systematically evaluate three key reasoning dimensions: \textit{Completeness} (logically sound deduction), \textit{Conciseness} (non-redundancy), and \textit{Determinateness} (consistent answer entailment). Extensive experiments across three multi-hop QA datasets (HotpotQA, MusiQue, and 2WikiMultiHopQA) and over 20 LLMs (including GPT-5, Gemini-3-Pro, LLaMA3, and task-specific tuned models) reveal a critical capability gap: leading models often achieve high attribution scores (e.g., 92.85\% precision for Gemini-3 Pro) but struggle with global reasoning quality (e.g., 35.11\% Conciseness for Gemini-3 Pro). Our work establishes a robust standard for logical evaluation, highlighting the need to prioritize reasoning coherence alongside factual grounding in LLM development. Codes are available at: https://github.com/zhichaoyan11/LogicScore.

Large language models (LLMs) have demonstrated strong reasoning, planning, and communication abilities, enabling them to operate as autonomous agents in open environments. While single-agent systems remain limited in adaptability and coordination, recent progress has shifted attention toward multi-agent systems (MAS) composed of interacting LLMs that pursue cooperative, competitive, or mixed objectives. This emerging paradigm provides a powerful testbed for studying social dynamics and strategic behaviors among intelligent agents. However, current research remains fragmented and lacks a unifying theoretical foundation. To address this gap, we present a comprehensive survey of LLM-based multi-agent systems through a game-theoretic lens. By organizing existing studies around the four key elements of game theory: players, strategies, payoffs, and information, we establish a systematic framework for understanding, comparing, and guiding future research on the design and analysis of LLM-based MAS.

Deep learning models for brain tumor analysis require large and diverse datasets that are often siloed across healthcare institutions due to privacy regulations. We present a federated learning framework for brain tumor localization that enables multi-institutional collaboration without sharing sensitive patient data. Our method extends a hybrid Transformer-Graph Neural Network architecture derived from prior decoder-free supervoxel GNNs and is deployed within CAFEIN\textsuperscript{\textregistered}, CERN's federated learning platform designed for healthcare environments. We provide an explainability analysis through Transformer attention mechanisms that reveals which MRI modalities drive the model predictions. Experiments on the BraTS dataset demonstrate a key finding: while isolated training on individual client data triggers early stopping well before reaching full training capacity, federated learning enables continued model improvement by leveraging distributed data, ultimately matching centralized performance. This result provides strong justification for federated learning when dealing with complex tasks and high-dimensional input data, as aggregating knowledge from multiple institutions significantly benefits the learning process. Our explainability analysis, validated through rigorous statistical testing on the full test set (paired t-tests with Bonferroni correction), reveals that deeper network layers significantly increase attention to T2 and FLAIR modalities ($p<0.001$, Cohen's $d$=1.50), aligning with clinical practice.

The increasing deployment of deep learning systems requires systematic evaluation of their reliability in real-world scenarios. Traditional gradient-based adversarial attacks introduce small perturbations that rarely correspond to realistic failures and mainly assess robustness rather than functional behavior. Generative test generation methods offer an alternative but are often limited to simple datasets or constrained input domains. Although diffusion models enable high-fidelity image synthesis, their computational cost and limited controllability restrict their applicability to large-scale testing. We present HyNeA, a generative testing method that enables direct and efficient control over diffusion-based generation. HyNeA provides dataset-free controllability through hypernetworks, allowing targeted manipulation of the generative process without relying on architecture-specific conditioning mechanisms or dataset-driven adaptations such as fine-tuning. HyNeA employs a distinct training strategy that supports instance-level tuning to identify failure-inducing test cases without requiring datasets that explicitly contain examples of similar failures. This approach enables the targeted generation of realistic failure cases at substantially lower computational cost than search-based methods. Experimental results show that HyNeA improves controllability and test diversity compared to existing generative test generators and generalizes to domains where failure-labeled training data is unavailable.

The electric vehicle routing problem with time windows (EVRPTW) is a complex optimization problem in sustainable logistics, where routing decisions must minimize total travel distance, fleet size, and battery usage while satisfying strict customer time constraints. Although deep reinforcement learning (DRL) has shown great potential as an alternative to classical heuristics and exact solvers, existing DRL models often struggle to maintain training stability-failing to converge or generalize when constraints are dense. In this study, we propose a curriculum-based deep reinforcement learning (CB-DRL) framework designed to resolve this instability. The framework utilizes a structured three-phase curriculum that gradually increases problem complexity: the agent first learns distance and fleet optimization (Phase A), then battery management (Phase B), and finally the full EVRPTW (Phase C). To ensure stable learning across phases, the framework employs a modified proximal policy optimization algorithm with phase-specific hyperparameters, value and advantage clipping, and adaptive learning-rate scheduling. The policy network is built upon a heterogeneous graph attention encoder enhanced by global-local attention and feature-wise linear modulation. This specialized architecture explicitly captures the distinct properties of depots, customers, and charging stations. Trained exclusively on small instances with N=10 customers, the model demonstrates robust generalization to unseen instances ranging from N=5 to N=100, significantly outperforming standard baselines on medium-scale problems. Experimental results confirm that this curriculum-guided approach achieves high feasibility rates and competitive solution quality on out-of-distribution instances where standard DRL baselines fail, effectively bridging the gap between neural speed and operational reliability.

Open-domain Relational Triplet Extraction (ORTE) is the foundation for mining structured knowledge without predefined schemas. Despite the impressive in-context learning capabilities of Large Language Models (LLMs), existing methods are hindered by their reliance on static, heuristic-driven prompting strategies. Due to the lack of reflection mechanisms required to internalize erroneous signals, these methods exhibit vulnerability in semantic ambiguity, often making erroneous extraction patterns permanent. To address this bottleneck, we propose a Knowledge Reconstruction-driven Prompt Optimization (KRPO) framework to assist LLMs in continuously improving their extraction capabilities for complex ORTE task flows. Specifically, we design a self-evaluation mechanism based on knowledge restoration, which provides intrinsic feedback signals by projecting structured triplets into semantic consistency scores. Subsequently, we propose a prompt optimizer based on a textual gradient that can internalize historical experiences to iteratively optimize prompts, which can better guide LLMs to handle subsequent extraction tasks. Furthermore, to alleviate relation redundancy, we design a relation canonicalization memory that collects representative relations and provides semantically distinct schemas for the triplets. Extensive experiments across three datasets show that KRPO significantly outperforms strong baselines in the extraction F1 score.

Factorizable joint shift (FJS) was proposed as a type of distribution shift (or dataset shift) that comprises both covariate and label shift. Recently, it has been observed that FJS actually arises from consecutive label and covariate (or vice versa) shifts. Research into FJS so far has been confined to the case of categorical label spaces. We propose a framework for analysing distribution shift in the case of general label spaces, thus covering both classification and regression models. Based on the framework, we generalise existing results on FJS to general label spaces and propose a related extension of the expectation maximisation (EM) algorithm for class prior probabilities. We also take a fresh look at generalized label shift (GLS) in the case of general label spaces.

Driver distraction remains a leading contributor to motor vehicle crashes, necessitating rigorous evaluation of new in-vehicle technologies. This study assessed the visual and cognitive demands associated with an advanced Large Language Model (LLM) conversational agent (Gemini Live) during on-road driving, comparing it against handsfree phone calls, visual turn-by-turn guidance (low load baseline), and the Operation Span (OSPAN) task (high load anchor). Thirty-two licensed drivers completed five secondary tasks while visual and cognitive demands were measured using the Detection Response Task (DRT) for cognitive load, eye-tracking for visual attention, and subjective workload ratings. Results indicated that Gemini Live interactions (both single-turn and multi-turn) and hands-free phone calls shared similar levels of cognitive load, between that of visual turn-by-turn guidance and OSPAN. Exploratory analysis showed that cognitive load remained stable across extended multi-turn conversations. All tasks maintained mean glance durations well below the well-established 2-second safety threshold, confirming low visual demand. Furthermore, drivers consistently dedicated longer glances to the roadway between brief off-road glances toward the device during task completion, particularly during voice-based interactions, rendering longer total-eyes-off-road time findings less consequential. Subjective ratings mirrored objective data, with participants reporting low effort, demands, and perceived distraction for Gemini Live. These findings demonstrate that advanced LLM conversational agents, when implemented via voice interfaces, impose cognitive and visual demands comparable to established, low-risk hands-free benchmarks, supporting their safe deployment in the driving environment.

Explainable AI (XAI) is frequently positioned as a technical problem of revealing the inner workings of an AI model. This position is affected by unexamined onto-epistemological assumptions: meaning is treated as immanent to the model, the explainer is positioned outside the system, and a causal structure is presumed recoverable through computational techniques. In this paper, we draw on Barad's agential realism to develop an alternative onto-epistemology of XAI. We propose that interpretations are material-discursive performances that emerge from situated entanglements of the AI model with humans, context, and the interpretative apparatus. To develop this position, we read a comprehensive set of XAI methods through agential realism and reveal the assumptions and limitations that underpin several of these methods. We then articulate the framework's ethical dimension and propose design directions for XAI interfaces that support emergent interpretation, using a speculative text-to-music interface as a case study.

Mixture-of-Experts (MoE) architectures enable conditional computation by routing inputs to multiple expert subnetworks and are often motivated as a mechanism for scaling large language models. In this project, we instead study MoE behavior in an image classification setting, focusing on predictive performance, expert utilization, and generalization. We compare dense, SoftMoE, and SparseMoE classifier heads on the CIFAR10 dataset under comparable model capacity. Both MoE variants achieve slightly higher validation accuracy than the dense baseline while maintaining balanced expert utilization through regularization, avoiding expert collapse. To analyze generalization, we compute Hessian-based sharpness metrics at convergence, including the largest eigenvalue and trace of the loss Hessian, evaluated on both training and test data. We find that SoftMoE exhibits higher sharpness by these metrics, while Dense and SparseMoE lie in a similar curvature regime, despite all models achieving comparable generalization performance. Complementary loss surface perturbation analyses reveal qualitative differences in non-local behavior under finite parameter perturbations between dense and MoE models, which help contextualize curvature-based measurements without directly explaining validation accuracy. We further evaluate empirical inference efficiency and show that naively implemented conditional routing does not yield inference speedups on modern hardware at this scale, highlighting the gap between theoretical and realized efficiency in sparse MoE models.

Reinforcement learning (RL) has shown promising results in active flow control (AFC), yet progress in the field remains difficult to assess as existing studies rely on heterogeneous observation and actuation schemes, numerical setups, and evaluation protocols. Current AFC benchmarks attempt to address these issues but heavily rely on external computational fluid dynamics (CFD) solvers, are not fully differentiable, and provide limited 3D and multi-agent support. To overcome these limitations, we introduce FluidGym, the first standalone, fully differentiable benchmark suite for RL in AFC. Built entirely in PyTorch on top of the GPU-accelerated PICT solver, FluidGym runs in a single Python stack, requires no external CFD software, and provides standardized evaluation protocols. We present baseline results with PPO and SAC and release all environments, datasets, and trained models as public resources. FluidGym enables systematic comparison of control methods, establishes a scalable foundation for future research in learning-based flow control, and is available at https://github.com/safe-autonomous-systems/fluidgym.

We study in-context learning for nonparametric regression with $α$-Hölder smooth regression functions, for some $α>0$. We prove that, with $n$ in-context examples and $d$-dimensional regression covariates, a pretrained transformer with $Θ(\log n)$ parameters and $Ω\bigl(n^{2α/(2α+d)}\log^3 n\bigr)$ pretraining sequences can achieve the minimax-optimal rate of convergence $O\bigl(n^{-2α/(2α+d)}\bigr)$ in mean squared error. Our result requires substantially fewer transformer parameters and pretraining sequences than previous results in the literature. This is achieved by showing that transformers are able to approximate local polynomial estimators efficiently by implementing a kernel-weighted polynomial basis and then running gradient descent.

Batch inference workloads for causal transformer models frequently process sequences that share common prefixes, such as system prompts, few-shot examples, or shared queries. Standard inference engines treat each sequence independently, redundantly recomputing identical MLP activations for every copy of the shared prefix. We introduce RadixMLP, a technique that exploits the position-wise nature of MLPs, LayerNorms, linear projections, and embeddings to eliminate this redundancy. RadixMLP dynamically maps batches to a prefix trie, gathering shared segments into a compressed representation for position-wise computation and scattering results back only at attention boundaries. RadixMLP is stateless and operates within a single forward pass. In end-to-end serving benchmarks on MS~MARCO v1.1 with Qwen3 models (0.6B to 8B parameters), RadixMLP achieves 1.44-1.59$\times$ speedups in realistic reranking workloads, with up to $5\times$ speedups on synthetic benchmarks with longer shared prefixes. Our code is available at https://github.com/michaelfeil/radix-mlp.

Identifying combinations of players (that is, lineups) in basketball - and other sports - that perform well when they play together is one of the most important tasks in sports analytics. One of the main challenges associated with this task is the frequent substitutions that occur during a game, which results in highly sparse data. In particular, a National Basketball Association (NBA) team will use more than 600 lineups during a season, which translates to an average lineup having seen the court in approximately 25-30 possessions. Inevitably, any statistics that one collects for these lineups are going to be noisy, with low predictive value. Yet, there is no existing work (in the public at least) that addresses this problem. In this work, we propose a regression-based approach that controls for the opposition faced by each lineup, while it also utilizes information about the players making up the lineups. Our experiments show that L-RAPM provides improved predictive power than the currently used baseline, and this improvement increases as the sample size for the lineups gets smaller.

Data contamination undermines the validity of Large Language Model evaluation by enabling models to rely on memorized benchmark content rather than true generalization. While prior work has proposed contamination detection methods, these approaches are largely limited to English benchmarks, leaving multilingual contamination poorly understood. In this work, we investigate contamination dynamics in multilingual settings by fine-tuning several open-weight LLMs on varying proportions of Arabic datasets and evaluating them on original English benchmarks. To detect memorization, we extend the Tested Slot Guessing method with a choice-reordering strategy and incorporate Min-K% probability analysis, capturing both behavioral and distributional contamination signals. Our results show that translation into Arabic suppresses conventional contamination indicators, yet models still benefit from exposure to contaminated data, particularly those with stronger Arabic capabilities. This effect is consistently reflected in rising Mink% scores and increased cross-lingual answer consistency as contamination levels grow. To address this blind spot, we propose Translation-Aware Contamination Detection, which identifies contamination by comparing signals across multiple translated benchmark variants rather than English alone. The Translation-Aware Contamination Detection reliably exposes contamination even when English-only methods fail. Together, our findings highlight the need for multilingual, translation-aware evaluation pipelines to ensure fair, transparent, and reproducible assessment of LLMs.

Verifiable delegation in digital identity systems remains unresolved across centralized, federated, and self-sovereign identity (SSI) environments, particularly where both human users and autonomous AI agents must exercise and transfer authority without exposing primary credentials or private keys. We introduce a unified framework that enables bounded, auditable, and least-privilege delegation across heterogeneous identity ecosystems. The framework includes four key elements: Delegation Grants (DGs), first-class authorization artefacts that encode revocable transfers of authority with enforced scope reduction; a Canonical Verification Context (CVC) that normalizes verification requests into a single structured representation independent of protocols or credential formats; a layered reference architecture that separates trust anchoring, credential and proof validation, policy evaluation, and protocol mediation via a Trust Gateway; and an explicit treatment of blockchain anchoring as an optional integrity layer rather than a structural dependency. Together, these elements advance interoperable delegation and auditability and provide a foundation for future standardization, implementation, and integration of autonomous agents into trusted digital identity infrastructures.

Distributed computing has been widely applied in distributed edge networks for reducing the processing burden of high-dimensional data centralization, where a high-dimensional computational task is decomposed into multiple low-dimensional collaborative processing tasks or multiple edge nodes use distributed data to train a global model. However, the computing power of a single-edge node is limited, and collaborative computing will cause information leakage and excessive communication overhead. In this paper, we design a parallel collaborative distributed alternating direction method of multipliers (ADMM) and propose a three-phase parallel collaborative ADMM privacy computing (3P-ADMM-PC2) algorithm for distributed computing in edge networks, where the Paillier homomorphic encryption is utilized to protect data privacy during interactions. Especially, a quantization method is introduced, which maps the real numbers to a positive integer interval without affecting the homomorphic operations. To address the architectural mismatch between large-integer and Graphics Processing Unit (GPU) computing, we transform high-bitwidth computations into low-bitwidth matrix and vector operations. Thus the GPU can be utilized to implement parallel encryption and decryption computations with long keys. Finally, a GPU-accelerated 3P-ADMM-PC2 is proposed to optimize the collaborative computing tasks. Meanwhile, large-scale computational tasks are conducted in network topologies with varying numbers of edge nodes. Experimental results demonstrate that the proposed 3P-ADMM-PC2 has excellent mean square error performance, which is close to that of distributed ADMM without privacy-preserving. Compared to centralized ADMM and distributed ADMM implemented with Central Processing Unit (CPU) computation, the proposed scheme demonstrates a significant speedup ratio.

Reliable human--robot collaboration in emergency scenarios requires autonomous systems that can detect humans, infer navigation goals, and operate safely in dynamic environments. This paper presents HumanDiffusion, a lightweight image-conditioned diffusion planner that generates human-aware navigation trajectories directly from RGB imagery. The system combines YOLO-11--based human detection with diffusion-driven trajectory generation, enabling a quadrotor to approach a target person and deliver medical assistance without relying on prior maps or computationally intensive planning pipelines. Trajectories are predicted in pixel space, ensuring smooth motion and a consistent safety margin around humans. We evaluate HumanDiffusion in simulation and real-world indoor mock-disaster scenarios. On a 300-sample test set, the model achieves a mean squared error of 0.02 in pixel-space trajectory reconstruction. Real-world experiments demonstrate an overall mission success rate of 80% across accident-response and search-and-locate tasks with partial occlusions. These results indicate that human-conditioned diffusion planning offers a practical and robust solution for human-aware UAV navigation in time-critical assistance settings.

Modern machine learning systems are increasingly realised as multistage pipelines, yet existing transparency mechanisms typically operate at a model level: they describe what a system is and why it behaves as it does, but not how individual data samples are operationally recorded, tracked, and verified as they traverse the pipeline. This absence of verifiable, sample-level traceability leaves practitioners and users unable to determine whether a specific sample was used, when it was processed, or whether the corresponding records remain intact over time. We introduce FG-Trac, a model-agnostic framework that establishes verifiable, fine-grained sample-level traceability throughout machine learning pipelines. FG-Trac defines an explicit mechanism for capturing and verifying sample lifecycle events across preprocessing and training, computes contribution scores explicitly grounded in training checkpoints, and anchors these traces to tamper-evident cryptographic commitments. The framework integrates without modifying model architectures or training objectives, reconstructing complete and auditable data-usage histories with practical computational overhead. Experiments on a canonical convolutional neural network and a multimodal graph learning pipeline demonstrate that FG-Trac preserves predictive performance while enabling machine learning systems to furnish verifiable evidence of how individual samples were used and propagated during model execution.

Most existing time series classification methods adopt a discriminative paradigm that maps input sequences directly to one-hot encoded class labels. While effective, this paradigm struggles to incorporate contextual features and fails to capture semantic relationships among classes. To address these limitations, we propose InstructTime, a novel framework that reformulates time series classification as a multimodal generative task. Specifically, continuous numerical sequences, contextual textual features, and task instructions are treated as multimodal inputs, while class labels are generated as textual outputs by tuned language models. To bridge the modality gap, InstructTime introduces a time series discretization module that converts continuous sequences into discrete temporal tokens, together with an alignment projection layer and a generative self-supervised pre-training strategy to enhance cross-modal representation alignment. Building upon this framework, we further propose InstructTime++, which extends InstructTime by incorporating implicit feature modeling to compensate for the limited inductive bias of language models. InstructTime++ leverages specialized toolkits to mine informative implicit patterns from raw time series and contextual inputs, including statistical feature extraction and vision-language-based image captioning, and translates them into textual descriptions for seamless integration. Extensive experiments on multiple benchmark datasets demonstrate the superior performance of InstructTime++.

The performance of Language Models (LMs) on lower-resource, morphologically rich languages like Sinhala remains under-explored, particularly for Romanized Sinhala, which is prevalent in digital communication. This paper presents a comprehensive benchmark of modern LMs on a diverse corpus of Unicode and Romanized Sinhala. We evaluate open-source models using perplexity, a measure of how well a model predicts a text, and leading closed-source models via a qualitative analysis of sentence completion. Our findings reveal that the Mistral-Nemo-Base-2407 model achieves the strongest predictive performance on Unicode text and the Mistral-7B-v0.3 model for Romanized text. The results also highlight the strong all-around performance of the Llama-3.1-8B model for both scripts. Furthermore, a significant performance disparity exists among closed-source models: Gemini-1.5-pro and DeepSeek excel at Unicode generation, whereas Claude-3.5-Sonnet is superior at handling Romanized text. These results provide an essential guide for practitioners selecting models for Sinhala-specific applications and highlight the critical role of training data in handling script variations.

Unsupervised Environment Design (UED) seeks to automatically generate training curricula for reinforcement learning (RL) agents, with the goal of improving generalisation and zero-shot performance. However, designing effective curricula remains a difficult problem, particularly in settings where small subsets of environment parameterisations result in significant increases in the complexity of the required policy. Current methods struggle with a difficult credit assignment problem and rely on regret approximations that fail to identify challenging levels, both of which are compounded as the size of the environment grows. We propose Dynamic Environment Generation for UED (DEGen) to enable a denser level generator reward signal, reducing the difficulty of credit assignment and allowing for UED to scale to larger environment sizes. We also introduce a new regret approximation, Maximised Negative Advantage (MNA), as a significantly improved metric to optimise for, that better identifies more challenging levels. We show empirically that MNA outperforms current regret approximations and when combined with DEGen, consistently outperforms existing methods, especially as the size of the environment grows. We have made all our code available here: https://github.com/HarryMJMead/Dynamic-Environment-Generation-for-UED.

User interactions on e-commerce platforms are inherently diverse, involving behaviors such as clicking, favoriting, adding to cart, and purchasing. The transitions between these behaviors offer valuable insights into user-item interactions, serving as a key signal for un- derstanding evolving preferences. Consequently, there is growing interest in leveraging multi-behavior data to better capture user intent. Recent studies have explored sequential modeling of multi- behavior data, many relying on transformer-based architectures with polynomial time complexity. While effective, these approaches often incur high computational costs, limiting their applicability in large-scale industrial systems with long user sequences. To address this challenge, we propose the Transition-Aware Graph Attention Network (TGA), a linear-complexity approach for modeling multi-behavior transitions. Unlike traditional trans- formers that treat all behavior pairs equally, TGA constructs a structured sparse graph by identifying informative transitions from three perspectives: (a) item-level transitions, (b) category-level transitions, and (c) neighbor-level transitions. Built upon the structured graph, TGA employs a transition-aware graph Attention mechanism that jointly models user-item interactions and behav- ior transition types, enabling more accurate capture of sequential patterns while maintaining computational efficiency. Experiments show that TGA outperforms all state-of-the-art models while sig- nificantly reducing computational cost. Notably, TGA has been deployed in a large-scale industrial production environment, where it leads to impressive improvements in key business metrics.

Social media increasingly disseminates information through mixed image text posts, but rumors often exploit subtle inconsistencies and forged content, making detection based solely on post content difficult. Deep semantic mismatch rumors, which superficially align images and texts, pose particular challenges and threaten online public opinion. Existing multimodal rumor detection methods improve cross modal modeling but suffer from limited feature extraction, noisy alignment, and inflexible fusion strategies, while ignoring external factual evidence necessary for verifying complex rumors. To address these limitations, we propose a multimodal rumor detection model enhanced with external evidence and forgery features. The model uses a ResNet34 visual encoder, a BERT text encoder, and a forgery feature module extracting frequency-domain traces and compression artifacts via Fourier transformation. BLIP-generated image descriptions bridge image and text semantic spaces. A dual contrastive learning module computes contrastive losses between text image and text description pairs, improving detection of semantic inconsistencies. A gated adaptive feature-scaling fusion mechanism dynamically adjusts multimodal fusion and reduces redundancy. Experiments on Weibo and Twitter datasets demonstrate that our model outperforms mainstream baselines in macro accuracy, recall, and F1 score.

While large language models now handle million-token contexts, their capacity for reasoning across entire document repositories remains largely untested. Existing benchmarks are inadequate, as they are mostly limited to single long texts or rely on a "sparse retrieval" assumption-that answers can be derived from a few relevant chunks. This assumption fails for true corpus-level analysis, where evidence is highly dispersed across hundreds of documents and answers require global integration, comparison, and statistical aggregation. To address this critical gap, we introduce CorpusQA, a new benchmark scaling up to 10 million tokens, generated via a novel data synthesis framework. By decoupling reasoning from textual representation, this framework creates complex, computation-intensive queries with programmatically guaranteed ground-truth answers, challenging systems to perform holistic reasoning over vast, unstructured text without relying on fallible human annotation. We further demonstrate the utility of our framework beyond evaluation, showing that fine-tuning on our synthesized data effectively enhances an LLM's general long-context reasoning capabilities. Extensive experiments reveal that even state-of-the-art long-context LLMs struggle as input length increases, and standard retrieval-augmented generation systems collapse entirely. Our findings indicate that memory-augmented agentic architectures offer a more robust alternative, suggesting a critical shift is needed from simply extending context windows to developing advanced architectures for global information synthesis.

Time alters the visual appearance of entities in our world, like objects, places, and animals. Thus, for accurately generating contextually-relevant images, knowledge and reasoning about time can be crucial (e.g., for generating a landscape in spring vs. in winter). Yet, although substantial work exists on understanding and improving temporal knowledge in natural language processing, research on how temporal phenomena appear and are handled in text-to-image (T2I) models remains scarce. We address this gap with TempViz, the first data set to holistically evaluate temporal knowledge in image generation, consisting of 7.9k prompts and more than 600 reference images. Using TempViz, we study the capabilities of five T2I models across five temporal knowledge categories. Human evaluation shows that temporal competence is generally weak, with no model exceeding 75% accuracy across categories. Towards larger-scale studies, we also examine automated evaluation methods, comparing several established approaches against human judgments. However, none of these approaches provides a reliable assessment of temporal cues - further indicating the pressing need for future research on temporal knowledge in T2I.

Large-scale Vision-Language-Action (VLA) models offer semantic generalization but suffer from high inference latency, limiting them to low-frequency batch-and-execute paradigm. This frequency mismatch creates an execution blind spot, causing failures in dynamic environments where targets move during the open-loop execution window. We propose TIDAL (Temporally Interleaved Diffusion and Action Loop), a hierarchical framework that decouples semantic reasoning from high-frequency actuation. TIDAL operates as a backbone-agnostic module for diffusion-based VLAs, using a dual-frequency architecture to redistribute the computational budget. Specifically, a low-frequency macro-intent loop caches semantic embeddings, while a high-frequency micro-control loop interleaves single-step flow integration with execution. This design enables approximately 9 Hz control updates on edge hardware (vs. approximately 2.4 Hz baselines) without increasing marginal overhead. To handle the resulting latency shift, we introduce a temporally misaligned training strategy where the policy learns predictive compensation using stale semantic intent alongside real-time proprioception. Additionally, we address the insensitivity of static vision encoders to velocity by incorporating a differential motion predictor. TIDAL is architectural, making it orthogonal to system-level optimizations. Experiments show a 2x performance gain over open-loop baselines in dynamic interception tasks. Despite a marginal regression in static success rates, our approach yields a 4x increase in feedback frequency and extends the effective horizon of semantic embeddings beyond the native action chunk size. Under non-paused inference protocols, TIDAL remains robust where standard baselines fail due to latency.

LLMs are ubiquitous in modern NLP, and while their applicability extends to texts produced for democratic activities such as online deliberations or large-scale citizen consultations, ethical questions have been raised for their usage as analysis tools. We continue this line of research with two main goals: (a) to develop resources that can help standardize citizen contributions in public forums at the pragmatic level, and make them easier to use in topic modeling and political analysis; (b) to study how well this standardization can reliably be performed by small, open-weights LLMs, i.e. models that can be run locally and transparently with limited resources. Accordingly, we introduce Corpus Clarification as a preprocessing framework for large-scale consultation data that transforms noisy, multi-topic contributions into structured, self-contained argumentative units ready for downstream analysis. We present GDN-CC, a manually-curated dataset of 1,231 contributions to the French Grand Débat National, comprising 2,285 argumentative units annotated for argumentative structure and manually clarified. We then show that finetuned Small Language Models match or outperform LLMs on reproducing these annotations, and measure their usability for an opinion clustering task. We finally release GDN-CC-large, an automatically annotated corpus of 240k contributions, the largest annotated democratic consultation dataset to date.

Semantic communication is emerging as a key enabler for distributed edge intelligence due to its capability to convey task-relevant meaning. However, achieving communication-efficient training and robust inference over wireless links remains challenging. This challenge is further exacerbated for multi-modal edge inference (MMEI) by two factors: 1) prohibitive communication overhead for distributed learning over bandwidth-limited wireless links, due to the \emph{multi-modal} nature of the system; and 2) limited robustness under varying channels and noisy multi-modal inputs. In this paper, we propose a three-stage communication-aware distributed learning framework to improve training and inference efficiency while maintaining robustness over wireless channels. In Stage~I, devices perform local multi-modal self-supervised learning to obtain shared and modality-specific encoders without device--server exchange, thereby reducing the communication cost. In Stage~II, distributed fine-tuning with centralized evidential fusion calibrates per-modality uncertainty and reliably aggregates features distorted by noise or channel fading. In Stage~III, an uncertainty-guided feedback mechanism selectively requests additional features for uncertain samples, optimizing the communication--accuracy tradeoff in the distributed setting. Experiments on RGB--depth indoor scene classification show that the proposed framework attains higher accuracy with far fewer training communication rounds and remains robust to modality degradation or channel variation, outperforming existing self-supervised and fully supervised baselines.

This paper presents a method to automatically fix implicit data loss warnings in large C++ projects using Large Language Models (LLMs). Our approach uses the Language Server Protocol (LSP) to gather context, Tree-sitter to extract relevant code, and LLMs to make decisions and generate fixes. The method evaluates the necessity of range checks concerning performance implications and generates appropriate fixes. We tested this method in a large C++ project, resulting in a 92.73% acceptance rate of the fixes by human developers during the code review. Our LLM-generated fixes reduced the number of warning fix changes that introduced additional instructions due to range checks and exception handling by 39.09% compared to a baseline fix strategy. This result was 13.56% behind the optimal solutions created by human developers. These findings demonstrate that our LLM-based approach can reduce the manual effort to address compiler warnings while maintaining code quality and performance in a real-world scenario. Our automated approach shows promise for integration into existing development workflows, potentially improving code maintenance practices in complex C++ software projects.

This study explores the capacity of generative artificial intelligence (Gen AI) to contribute to the construction of peace narratives and the revitalization of musical heritage in Mali. The study has been made in a political and social context where inter-community tensions and social fractures motivate a search for new symbolic frameworks for reconciliation. The study empirically explores three questions: (1) how Gen AI can be used as a tool for musical creation rooted in national languages and traditions; (2) to what extent Gen AI systems enable a balanced hybridization between technological innovation and cultural authenticity; and (3) how AI-assisted musical co-creation can strengthen social cohesion and cultural sovereignty. The experimental results suggest that Gen AI, embedded in a culturally conscious participatory framework, can act as a catalyst for symbolic diplomacy, amplifying local voices instead of standardizing them. However, challenges persist regarding the availability of linguistic corpora, algorithmic censorship, and the ethics of generating compositions derived from copyrighted sources.

Single-channel speech enhancement algorithms are often used in resource-constrained embedded devices, where low latency and low complexity designs gain more importance. In recent years, researchers have proposed a wide variety of novel solutions to this problem. In particular, a recent deep learning model named ULCNet is among the state-of-the-art approaches in this domain. This paper proposes an adaptation of ULCNet, by replacing its GRU layers with FastGRNNs, to reduce both computational latency and complexity. Furthermore, this paper shows empirical evidence on the performance decay of FastGRNNs in long audio signals during inference due to internal state drifting, and proposes a novel approach based on a trainable complementary filter to mitigate it. The resulting model, Fast-ULCNet, performs on par with the state-of-the-art original ULCNet architecture on a speech enhancement task, while reducing its model size by more than half and decreasing its latency by 34% on average.

Modern Edge-to-Cloud (E2C) systems require fine-grained observability to ensure adaptive behavior and compliance with performance objectives across heterogeneous and dynamic environments. This work introduces an application-level observability framework that integrates developer-driven instrumentation and SLO-aware feedback for autonomous adaptation. By combining OpenTelemetry, Prometheus, K3s, and Chaos Mesh, the framework enables real-time monitoring and adaptive control across the continuum. A video processing use case demonstrates how application-level metrics guide automatic adjustments to maintain target frame rate, latency, and detection accuracy under variable workloads and injected faults. Preliminary results highlight improved scalability, fault tolerance, and responsiveness, providing a practical foundation for adaptive, SLO-compliant E2C applications.

Vision-Language Models (VLMs) enable multimodal reasoning for robotic perception and interaction, but their deployment in real-world systems remains constrained by latency, limited onboard resources, and privacy risks of cloud offloading. Edge intelligence within 6G, particularly Open RAN and Multi-access Edge Computing (MEC), offers a pathway to address these challenges by bringing computation closer to the data source. This work investigates the deployment of VLMs on ORAN/MEC infrastructure using the Unitree G1 humanoid robot as an embodied testbed. We design a WebRTC-based pipeline that streams multimodal data to an edge node and evaluate LLaMA-3.2-11B-Vision-Instruct deployed at the edge versus in the cloud under real-time conditions. Our results show that edge deployment preserves near-cloud accuracy while reducing end-to-end latency by 5\%. We further evaluate Qwen2-VL-2B-Instruct, a compact model optimized for resource-constrained environments, which achieves sub-second responsiveness, cutting latency by more than half but at the cost of accuracy.

Effective management of Type 1 Diabetes requires continuous glucose monitoring and precise insulin adjustments to prevent hyperglycemia and hypoglycemia. With the growing adoption of wearable glucose monitors and mobile health applications, accurate blood glucose prediction is essential for enhancing automated insulin delivery and decision-support systems. This paper presents a deep learning-based approach for personalized blood glucose prediction, leveraging patient-specific data to improve prediction accuracy and responsiveness in real-world scenarios. Unlike traditional generalized models, our method accounts for individual variability, enabling more effective subject-specific predictions. We compare Leave-One-Subject-Out Cross-Validation with a fine-tuning strategy to evaluate their ability to model patient-specific dynamics. Results show that personalized models significantly improve the prediction of adverse events, enabling more precise and timely interventions in real-world scenarios. To assess the impact of patient-specific data, we conduct experiments comparing a multimodal, patient-specific approach against traditional CGM-only methods. Additionally, we perform an ablation study to investigate model performance with progressively smaller training sets, identifying the minimum data required for effective personalization-an essential consideration for real-world applications where extensive data collection is often challenging. Our findings underscore the potential of adaptive, personalized glucose prediction models for advancing next-generation diabetes management, particularly in wearable and mobile health platforms, enhancing consumer-oriented diabetes care solutions.

Recent advances in large language models (LLMs) allow agents to represent actions as executable code, offering greater expressivity than traditional tool-calling. However, real-world tasks often demand both strategic planning and detailed implementation. Using a single agent for both leads to context pollution from debugging traces and intermediate failures, impairing long-horizon performance. We propose CodeDelegator, a multi-agent framework that separates planning from implementation via role specialization. A persistent Delegator maintains strategic oversight by decomposing tasks, writing specifications, and monitoring progress without executing code. For each sub-task, a new Coder agent is instantiated with a clean context containing only its specification, shielding it from prior failures. To coordinate between agents, we introduce Ephemeral-Persistent State Separation (EPSS), which isolates each Coder's execution state while preserving global coherence, preventing debugging traces from polluting the Delegator's context. Experiments on various benchmarks demonstrate the effectiveness of CodeDelegator across diverse scenarios.

Alerts are critical for detecting anomalies in large-scale cloud systems, ensuring reliability and user experience. However, current systems generate overwhelming volumes of alerts, degrading operational efficiency due to ineffective alert life-cycle management. This paper details the efforts of Company-X to optimize alert life-cycle management, addressing alert fatigue in cloud systems. We propose AlertGuardian, a framework collaborating large language models (LLMs) and lightweight graph models to optimize the alert life-cycle through three phases: Alert Denoise uses graph learning model with virtual noise to filter noise, Alert Summary employs Retrieval Augmented Generation (RAG) with LLMs to create actionable summary, and Alert Rule Refinement leverages multi-agent iterative feedbacks to improve alert rule quality. Evaluated on four real-world datasets from Company-X's services, AlertGuardian significantly mitigates alert fatigue (94.8\% alert reduction ratios) and accelerates fault diagnosis (90.5\% diagnosis accuracy). Moreover, AlertGuardian improves 1,174 alert rules, with 375 accepted by SREs (32% acceptance rate). Finally, we share success stories and lessons learned about alert life-cycle management after the deployment of AlertGuardian in Company-X.

Podcast script generation requires LLMs to synthesize structured, context-grounded dialogue from diverse inputs, yet systematic evaluation resources for this task remain limited. To bridge this gap, we introduce PodBench, a benchmark comprising 800 samples with inputs up to 21K tokens and complex multi-speaker instructions. We propose a multifaceted evaluation framework that integrates quantitative constraints with LLM-based quality assessment. Extensive experiments reveal that while proprietary models generally excel, open-source models equipped with explicit reasoning demonstrate superior robustness in handling long contexts and multi-speaker coordination compared to standard baselines. However, our analysis uncovers a persistent divergence where high instruction following does not guarantee high content substance. PodBench offers a reproducible testbed to address these challenges in long-form, audio-centric generation.

Recent debates on artificial intelligence increasingly emphasise questions of AI consciousness and moral status, yet there remains little agreement on how such properties should be evaluated. In this paper, we argue that awareness offers a more productive and methodologically tractable alternative. We introduce a practical method for evaluating awareness across diverse systems, where awareness is understood as encompassing a system's abilities to process, store and use information in the service of goal-directed action. Central to this approach is the claim that any evaluation aiming to capture the diversity of artificial systems must be domain-sensitive, deployable at any scale, multidimensional, and enable the prediction of task performance, while generalising to the level of abilities for the sake of comparison. Given these four desiderata, we outline a structured approach to evaluating and comparing awareness profiles across artificial systems with differing architectures, scales, and operational domains. By shifting the focus from artificial consciousness to being just aware enough, this approach aims to facilitate principled assessment, support design and oversight, and enable more constructive scientific and public discourse.

Multilingual retrieval-augmented generation (MRAG) requires models to effectively acquire and integrate beneficial external knowledge from multilingual collections. However, most existing studies employ a unitive process where queries of equivalent semantics across different languages are processed through a single-turn retrieval and subsequent optimization. Such a ``one-size-fits-all'' strategy is often suboptimal in multilingual settings, as the models occur to knowledge bias and conflict during the interaction with the search engine. To alleviate the issues, we propose LcRL, a multilingual search-augmented reinforcement learning framework that integrates a language-coupled Group Relative Policy Optimization into the policy and reward models. We adopt the language-coupled group sampling in the rollout module to reduce knowledge bias, and regularize an auxiliary anti-consistency penalty in the reward models to mitigate the knowledge conflict. Experimental results demonstrate that LcRL not only achieves competitive performance but is also appropriate for various practical scenarios such as constrained training data and retrieval over collections encompassing a large number of languages. Our code is available at https://github.com/Cherry-qwq/LcRL-Open.

We present SpatialMem, a memory-centric system that unifies 3D geometry, semantics, and language into a single, queryable representation. Starting from casually captured egocentric RGB video, SpatialMem reconstructs metrically scaled indoor environments, detects structural 3D anchors (walls, doors, windows) as the first-layer scaffold, and populates a hierarchical memory with open-vocabulary object nodes -- linking evidence patches, visual embeddings, and two-layer textual descriptions to 3D coordinates -- for compact storage and fast retrieval. This design enables interpretable reasoning over spatial relations (e.g., distance, direction, visibility) and supports downstream tasks such as language-guided navigation and object retrieval without specialized sensors. Experiments across three real-life indoor scenes demonstrate that SpatialMem maintains strong anchor-description-level navigation completion and hierarchical retrieval accuracy under increasing clutter and occlusion, offering an efficient and extensible framework for embodied spatial intelligence.

Background: Neuro-symbolic methods enhance the reliability of neural network classifiers through logical constraints, but they lack native support for ontologies. Objectives: We aim to develop a neuro-symbolic method that reliably outputs predictions consistent with a Description Logic ontology that formalizes domain-specific knowledge. Methods: We encode a Description Logic ontology as a circuit, a feed-forward differentiable computational graph that supports tractable execution of queries and transformations. We show that the circuit can be used to (i) generate synthetic datasets that capture the semantics of the ontology; (ii) efficiently perform deductive reasoning on a GPU; (iii) implement neuro-symbolic models whose predictions are approximately or provably consistent with the knowledge defined in the ontology. Results We show that the synthetic dataset generated using the circuit qualitatively captures the semantics of the ontology while being challenging for Machine Learning classifiers, including neural networks. Moreover, we show that compiling the ontology into a circuit is a promising approach for scalable deductive reasoning, with runtimes up to three orders of magnitude faster than available reasoners. Finally, we show that our neuro-symbolic classifiers reliably produce consistent predictions when compared to neural network baselines, maintaining competitive performances or even outperforming them. Conclusions By compiling Description Logic ontologies into circuits, we obtain a tighter integration between the Deep Learning and Knowledge Representation fields. We show that a single circuit representation can be used to tackle different challenging tasks closely related to real-world applications.

Reasoning models excel at complex tasks such as coding and mathematics, yet their inference is often slow and token-inefficient. To improve the inference efficiency, post-training quantization (PTQ) usually comes with the cost of large accuracy drops, especially for reasoning tasks under low-bit settings. In this study, we present a systematic empirical study of quantization-aware training (QAT) for reasoning models. Our key findings include: (1) Knowledge distillation is a robust objective for reasoning models trained via either supervised fine-tuning or reinforcement learning; (2) PTQ provides a strong initialization for QAT, improving accuracy while reducing training cost; (3) Reinforcement learning remains feasible for quantized models given a viable cold start and yields additional gains; and (4) Aligning the PTQ calibration domain with the QAT training domain accelerates convergence and often improves the final accuracy. Finally, we consolidate these findings into an optimized workflow (Reasoning-QAT), and show that it consistently outperforms state-of-the-art PTQ methods across multiple LLM backbones and reasoning datasets. For instance, on Qwen3-0.6B, it surpasses GPTQ by 44.53% on MATH-500 and consistently recovers performance in the 2-bit regime.

The Transiting Exoplanet Survey Satellite (TESS) Full-Frame Images (FFIs) provide photometric time series for millions of stars, enabling transit searches beyond the limited set of pre-selected 2-minute targets. However, FFIs present additional challenges for transit identification and vetting. In this work, we apply ExoMiner++ 2.0, an adaptation of the ExoMiner++ framework originally developed for TESS 2-minute data, to FFI light curves. The model is used to perform large-scale planet versus non-planet classification of Threshold Crossing Events across the sectors analyzed in this study. We construct a uniform vetting catalog of all evaluated signals and assess model performance under different observing conditions. We find that ExoMiner++ 2.0 generalizes effectively to the FFI domain, providing robust discrimination between planetary signals, astrophysical false positives, and instrumental artifacts despite the limitations inherent to longer cadence data. This work extends the applicability of ExoMiner++ to the full TESS dataset and supports future population studies and follow-up prioritization.

High-fidelity 4D dynamic facial avatar reconstruction from monocular video is a critical yet challenging task, driven by increasing demands for immersive virtual human applications. While Neural Radiance Fields (NeRF) have advanced scene representation, their capacity to capture high-frequency facial details, such as dynamic wrinkles and subtle textures from information-constrained monocular streams, requires significant enhancement. To tackle this challenge, we propose a novel hybrid neural radiance field framework, called Geometry-Aware-Transformer Enhanced NeRF (GAT-NeRF) for high-fidelity and controllable 4D facial avatar reconstruction, which integrates the Transformer mechanism into the NeRF pipeline. GAT-NeRF synergistically combines a coordinate-aligned Multilayer Perceptron (MLP) with a lightweight Transformer module, termed as Geometry-Aware-Transformer (GAT) due to its processing of multi-modal inputs containing explicit geometric priors. The GAT module is enabled by fusing multi-modal input features, including 3D spatial coordinates, 3D Morphable Model (3DMM) expression parameters, and learnable latent codes to effectively learn and enhance feature representations pertinent to fine-grained geometry. The Transformer's effective feature learning capabilities are leveraged to significantly augment the modeling of complex local facial patterns like dynamic wrinkles and acne scars. Comprehensive experiments unequivocally demonstrate GAT-NeRF's state-of-the-art performance in visual fidelity and high-frequency detail recovery, forging new pathways for creating realistic dynamic digital humans for multimedia applications.

We study a noisy linear observation model with an unknown permutation called permuted/shuffled linear regression, where responses and covariates are mismatched and the permutation forms a discrete, factorial-size parameter. This unknown permutation is a key component of the data-generating process, yet its statistical investigation remains challenging due to its discrete nature. In this study, we develop a general statistical inference framework on the permutation and regression coefficients. First, we introduce a localization step that reduces the permutation space to a small candidate set building on recent advances in the repro samples method, whose miscoverage decays polynomially with the number of Monte Carlo samples. Then, based on this localized set, we provide statistical inference procedures: a conditional Monte Carlo test of permutation structures with valid finite-sample Type-I error control. We also develop coefficient inference that remains valid under alignment uncertainty of permutations. For computational purposes, we develop a linear assignment problem computable in polynomial time complexity and demonstrate that its solution asymptotically converges to that of the conventional least squares problem with large computational cost. Extensions to partially permuted designs and ridge regularization are also discussed. Extensive simulations and an application to Beijing air-quality data corroborate finite-sample validity, strong power to detect mismatches, and practical scalability.

Explanations are essential in software engineering (SE) and requirements communication, helping stakeholders clarify ambiguities, justify design choices, and build shared understanding. Online Q&A forums such as Stack Overflow provide large-scale settings where such explanations are produced and evaluated, offering valuable insights into what makes them effective. While prior work has explored answer acceptance and voting behavior, little is known about which specific features make explanations genuinely useful. The relative influence of structural, contextual, and linguistic factors, such as content richness, timing, and sentiment, remains unclear. We analyzed 3,323 questions and 59,398 answers from Stack Overflow, combining text analysis and statistical modeling to examine how explanation attributes relate to perceived usefulness (normalized upvotes). Structural and contextual factors, especially explanation length, code inclusion, timing, and author reputation, show small to moderate positive effects. Sentiment polarity has negligible influence, suggesting that clarity and substance outweigh tone in technical communication. This study provides an empirical account of what drives perceived usefulness in developer explanations. It contributes methodological transparency through open data and replication materials, and conceptual insight by relating observed communication patterns to principles of requirements communication. The findings offer evidence-based implications for how developers and RE practitioners can craft clearer and more effective explanations, potentially supporting fairer communication in both open and organizational contexts. From an RE perspective, these determinants can be interpreted as practical signals for ambiguity reduction and rationale articulation in day-to-day requirements communication.

We introduce Strategic Doctrine Language Models (sdLM), a learning-system framework for multi-document strategic reasoning with doctrinal consistency constraints and calibrated uncertainty. The approach combines multi-document attention, temporal encoding, and a doctrine-consistency layer to improve long-horizon forecasting and plan plausibility while reducing severe doctrinal violations. We evaluate sdLM using (i) expert-panel scoring of strategic scenarios (N=47), (ii) doctrine consistency on 336 doctrine publications (12,847 statements), and (iii) geopolitical forecasting on 127 historical counterfactuals (1945-2020) across 12-60 month horizons. Across these benchmarks, sdLM achieves higher strategic quality and better calibration than strong general-purpose LLM baselines, and remains competitive with human experts on long-horizon judgments. We further report ablations, scaling trends, and deployment-oriented performance/latency characteristics to clarify which components drive improvements and how they translate to operational settings.

Software engineering is the invisible infrastructure of the digital age. Every breakthrough in artificial intelligence, quantum computing, photonics, and cybersecurity relies on advances in software engineering, yet the field is too often treated as a supportive digital component rather than as a strategic, enabling discipline. In policy frameworks, including major European programmes, software appears primarily as a building block within other technologies, while the scientific discipline of software engineering remains largely absent. This position paper argues that the long-term sustainability, dependability, and sovereignty of digital technologies depend on investment in software engineering research. It is a call to reclaim the identity of software engineering.

Memory-augmented language agents rely on embedding models for effective memory retrieval. However, existing training data construction overlooks a critical limitation: the hierarchical difficulty of negative samples and their natural distribution in human-agent interactions. In practice, some negatives are semantically close distractors while others are trivially irrelevant, and natural dialogue exhibits structured proportions of these types. Current approaches using synthetic or uniformly sampled negatives fail to reflect this diversity, limiting embedding models' ability to learn nuanced discrimination essential for robust memory retrieval. In this work, we propose a principled data construction framework HiNS that explicitly models negative sample difficulty tiers and incorporates empirically grounded negative ratios derived from conversational data, enabling the training of embedding models with substantially improved retrieval fidelity and generalization in memory-intensive tasks. Experiments show significant improvements: on LoCoMo, F1/BLEU-1 gains of 3.27%/3.30%(MemoryOS) and 1.95%/1.78% (Mem0); on PERSONAMEM, total score improvements of 1.19% (MemoryOS) and 2.55% (Mem0).

Black-box variational inference (BBVI) with Gaussian mixture families offers a flexible approach for approximating complex posterior distributions without requiring gradients of the target density. However, standard numerical optimization methods often suffer from instability and inefficiency. We develop a stable and efficient framework that combines three key components: (1) affine-invariant preconditioning via natural gradient formulations, (2) an exponential integrator that unconditionally preserves the positive definiteness of covariance matrices, and (3) adaptive time stepping to ensure stability and to accommodate distinct warm-up and convergence phases. The proposed approach has natural connections to manifold optimization and mirror descent. For Gaussian posteriors, we prove exponential convergence in the noise-free setting and almost-sure convergence under Monte Carlo estimation, rigorously justifying the necessity of adaptive time stepping. Numerical experiments on multimodal distributions, Neal's multiscale funnel, and a PDE-based Bayesian inverse problem for Darcy flow demonstrate the effectiveness of the proposed method.

On and off-ramps are understudied road sections even though they introduce a higher level of variation in highway interactions. Predicting vehicles' behavior in these areas can decrease the impact of uncertainty and increase road safety. In this paper, the difference between this Area of Interest (AoI) and a straight highway section is studied. Multi-layered LSTM architecture to train the AoI model with ExiD drone dataset is utilized. In the process, different prediction horizons and different models' workflow are tested. The results show great promise on horizons up to 4 seconds with prediction accuracy starting from about 76% for the AoI and 94% for the general highway scenarios on the maximum horizon.

Creating high-fidelity, real-time drivable 3D head avatars is a core challenge in digital animation. While 3D Gaussian Splashing (3D-GS) offers unprecedented rendering speed and quality, current animation techniques often rely on a "one-size-fits-all" global tuning approach, where all Gaussian primitives are uniformly driven by a single expression code. This simplistic approach fails to unravel the distinct dynamics of different facial regions, such as deformable skin versus rigid teeth, leading to significant blurring and distortion artifacts. We introduce Conditionally-Adaptive Gaussian Avatars (CAG-Avatar), a framework that resolves this key limitation. At its core is a Conditionally Adaptive Fusion Module built on cross-attention. This mechanism empowers each 3D Gaussian to act as a query, adaptively extracting relevant driving signals from the global expression code based on its canonical position. This "tailor-made" conditioning strategy drastically enhances the modeling of fine-grained, localized dynamics. Our experiments confirm a significant improvement in reconstruction fidelity, particularly for challenging regions such as teeth, while preserving real-time rendering performance.

Microtubules are cytoskeletal filaments that play essential roles in many cellular processes and are key therapeutic targets in several diseases. Accurate segmentation of microtubule networks is critical for studying their organization and dynamics but remains challenging due to filament curvature, dense crossings, and image noise. We present MTFlow, a novel time-conditioned flow-matching model for microtubule segmentation. Unlike conventional U-Net variants that predict masks in a single pass, MTFlow learns vector fields that iteratively transport noisy masks toward the ground truth, enabling interpretable, trajectory-based refinement. Our architecture combines a U-Net backbone with temporal embeddings, allowing the model to capture the dynamics of uncertainty resolution along filament boundaries. We trained and evaluated MTFlow on synthetic and real microtubule datasets and assessed its generalization capability on public biomedical datasets of curvilinear structures such as retinal blood vessels and nerves. MTFlow achieves competitive segmentation accuracy comparable to state-of-the-art models, offering a powerful and time-efficient tool for filamentous structure analysis with more precise annotations than manual or semi-automatic approaches.

This paper introduces KRROOD, a framework designed to bridge the integration gap between modern software engineering and Knowledge Representation & Reasoning (KR&R) systems. While Object-Oriented Programming (OOP) is the standard for developing complex applications, existing KR&R frameworks often rely on external ontologies and specialized languages that are difficult to integrate with imperative code. KRROOD addresses this by treating knowledge as a first-class programming abstraction using native class structures, bridging the gap between the logic programming and OOP paradigms. We evaluate the system on the OWL2Bench benchmark and a human-robot task learning scenario. Experimental results show that KRROOD achieves strong performance while supporting the expressive reasoning required for real-world autonomous systems.

Vision-Language Models show strong zero-shot performance for chest X-ray classification, but standard flat metrics fail to distinguish between clinically minor and severe errors. This work investigates how to quantify and mitigate abstraction errors by leveraging medical taxonomies. We benchmark several state-of-the-art VLMs using hierarchical metrics and introduce Catastrophic Abstraction Errors to capture cross-branch mistakes. Our results reveal substantial misalignment of VLMs with clinical taxonomies despite high flat performance. To address this, we propose risk-constrained thresholding and taxonomy-aware fine-tuning with radial embeddings, which reduce severe abstraction errors to below 2 per cent while maintaining competitive performance. These findings highlight the importance of hierarchical evaluation and representation-level alignment for safer and more clinically meaningful deployment of VLMs.

As large language models (LLMs) are increasingly applied to creative writing, their performance on culturally specific narrative tasks warrants systematic investigation. This study constructs the first Chinese film script continuation benchmark comprising 53 classic films, and designs a multi-dimensional evaluation framework comparing GPT-5.2 and Qwen-Max-Latest. Using a "first half to second half" continuation paradigm with 3 samples per film, we obtained 303 valid samples (GPT-5.2: 157, 98.7% validity; Qwen-Max: 146, 91.8% validity). Evaluation integrates ROUGE-L, Structural Similarity, and LLM-as-Judge scoring (DeepSeek-Reasoner). Statistical analysis of 144 paired samples reveals: Qwen-Max achieves marginally higher ROUGE-L (0.2230 vs 0.2114, d=-0.43); however, GPT-5.2 significantly outperforms in structural preservation (0.93 vs 0.75, d=0.46), overall quality (44.79 vs 25.72, d=1.04), and composite scores (0.50 vs 0.39, d=0.84). The overall quality effect size reaches large effect level (d>0.8). GPT-5.2 excels in character consistency, tone-style matching, and format preservation, while Qwen-Max shows deficiencies in generation stability. This study provides a reproducible framework for LLM evaluation in Chinese creative writing.

Melanoma detection is vital for early diagnosis and effective treatment. While deep learning models on dermoscopic images have shown promise, they require specialized equipment, limiting their use in broader clinical settings. This study introduces a multi-modal melanoma detection system using conventional photo images, making it more accessible and versatile. Our system integrates image data with tabular metadata, such as patient demographics and lesion characteristics, to improve detection accuracy. It employs a multi-modal neural network combining image and metadata processing and supports a two-step model for cases with or without metadata. A three-stage pipeline further refines predictions by boosting algorithms and enhancing performance. To address the challenges of a highly imbalanced dataset, specific techniques were implemented to ensure robust training. An ablation study evaluated recent vision architectures, boosting algorithms, and loss functions, achieving a peak Partial ROC AUC of 0.18068 (0.2 maximum) and top-15 retrieval sensitivity of 0.78371. Results demonstrate that integrating photo images with metadata in a structured, multi-stage pipeline yields significant performance improvements. This system advances melanoma detection by providing a scalable, equipment-independent solution suitable for diverse healthcare environments, bridging the gap between specialized and general clinical practices.

In supervised learning with distributional inputs in the two-stage sampling setup, relevant to applications like learning-based medical screening or causal learning, the inputs (which are probability distributions) are not accessible in the learning phase, but only samples thereof. This problem is particularly amenable to kernel-based learning methods, where the distributions or samples are first embedded into a Hilbert space, often using kernel mean embeddings (KMEs), and then a standard kernel method like Support Vector Machines (SVMs) is applied, using a kernel defined on the embedding Hilbert space. In this work, we contribute to the theoretical analysis of this latter approach, with a particular focus on classification with distributional inputs using SVMs. We establish a new oracle inequality and derive consistency and learning rate results. Furthermore, for SVMs using the hinge loss and Gaussian kernels, we formulate a novel variant of an established noise assumption from the binary classification literature, under which we can establish learning rates. Finally, some of our technical tools like a new feature space for Gaussian kernels on Hilbert spaces are of independent interest.

Machine Learning (ML) has deeply changed some fields recently, like Language and Vision and we may expect it to be relevant also to the analysis of of complex systems. Here we want to tackle the question of how and to which extent can one regress scale-free processes, i.e. processes displaying power law behavior, like earthquakes or avalanches? We are interested in predicting the large ones, i.e. rare events in the training set which therefore require extrapolation capabilities of the model. For this we consider two paradigmatic problems that are statistically self-similar. The first one is a 2-dimensional fractional Gaussian field obeying linear dynamics, self-similar by construction and amenable to exact analysis. The second one is the Abelian sandpile model, exhibiting self-organized criticality. The emerging paradigm of Geometric Deep Learning shows that including known symmetries into the model's architecture is key to success. Here one may hope to extrapolate only by leveraging scale invariance. This is however a peculiar symmetry, as it involves possibly non-trivial coarse-graining operations and anomalous scaling. We perform experiments on various existing architectures like U-net, Riesz network (scale invariant by construction), or our own proposals: a wavelet-decomposition based Graph Neural Network (with discrete scale symmetry), a Fourier embedding layer and a Fourier-Mellin Neural Operator. Based on these experiments and a complete characterization of the linear case, we identify the main issues relative to spectral biases and coarse-grained representations, and discuss how to alleviate them with the relevant inductive biases.

Fault injection is a key technique for assessing software reliability, enabling proactive detection of system defects before they manifest in production. However, the increasing complexity of microservice architectures leads to exponential growth in the fault-injection space, rendering traditional random injection inefficient. Recent lineage-driven approaches mitigate this problem through heuristic pruning, but they face two limitations. First, combinatorial-fault discovery remains bottlenecked by general-purpose SAT solvers, which fail to exploit the monotone and low-overlap structure of derived CNF formulas and typically rely on a static upper bound on fault size. Second, existing techniques provide limited post-injection guidance beyond reporting detected faults. To address these challenges, we propose FastFI, a fault-injection-guided framework to enhance the robustness of API call sites in microservice-based systems. FastFI features a DFS-based solver with dynamic fault injection to discover all valid combinatorial faults, and it leverages fault-injection results to identify critical APIs whose call sites should be hardened for robustness. Experiments on four representative microservice benchmarks show that FastFI reduces end-to-end fault-injection time by an average of 76.12\% compared to state-of-the-art baselines while maintaining acceptable resource overhead. Moreover, FastFI accurately identifies high-impact APIs and provides actionable guidance for call-site hardening.

Designing good reflection questions is pedagogically important but time-consuming and unevenly supported across teachers. This paper introduces a reflection-in-reflection framework for automated generation of reflection questions with large language models (LLMs). Our approach coordinates two role-specialized agents, a Student-Teacher and a Teacher-Educator, that engage in a Socratic multi-turn dialogue to iteratively refine a single question given a teacher-specified topic, key concepts, student level, and optional instructional materials. The Student-Teacher proposes candidate questions with brief rationales, while the Teacher-Educator evaluates them along clarity, depth, relevance, engagement, and conceptual interconnections, responding only with targeted coaching questions or a fixed signal to stop the dialogue. We evaluate the framework in an authentic lower-secondary ICT setting on the topic, using GPT-4o-mini as the backbone model and a stronger GPT- 4-class LLM as an external evaluator in pairwise comparisons of clarity, relevance, depth, and overall quality. First, we study how interaction design and context (dynamic vs.fixed iteration counts; presence or absence of student level and materials) affect question quality. Dynamic stopping combined with contextual information consistently outperforms fixed 5- or 10-step refinement, with very long dialogues prone to drift or over-complication. Second, we show that our two-agent protocol produces questions that are judged substantially more relevant and deeper, and better overall, than a one-shot baseline using the same backbone model.

We introduce RANDSMAPs (Random-feature/multi-scale neural decoders with Mass Preservation), numerical analysis-informed, explainable neural decoders designed to explicitly respect conservation laws when solving the challenging ill-posed pre-image problem in manifold learning. We start by proving the equivalence of vanilla random Fourier feature neural networks to Radial Basis Function interpolation and the double Diffusion Maps (based on Geometric Harmonics) decoders in the deterministic limit. We then establish the theoretical foundations for RANDSMAP and introduce its multiscale variant to capture structures across multiple scales. We formulate and derive the closed-form solution of the corresponding constrained optimization problem and prove the mass preservation property. Numerically, we assess the performance of RANDSMAP on three benchmark problems/datasets with mass preservation obtained by the Lighthill-Whitham-Richards traffic flow PDE with shock waves, 2D rotated MRI brain images, and the Hughes crowd dynamics PDEs. We demonstrate that RANDSMAPs yield high reconstruction accuracy at low computational cost and maintain mass conservation at single-machine precision. In its vanilla formulation, the scheme remains applicable to the classical pre-image problem, i.e., when mass-preservation constraints are not imposed.

Despite their practical success, it remains unclear why Mixture of Experts (MoE) models can outperform dense networks beyond sheer parameter scaling. We study an iso-parameter regime where inputs exhibit latent modular structure but are corrupted by feature noise, a proxy for noisy internal activations. We show that sparse expert activation acts as a noise filter: compared to a dense estimator, MoEs achieve lower generalization error under feature noise, improved robustness to perturbations, and faster convergence speed. Empirical results on synthetic data and real-world language tasks corroborate the theoretical insights, demonstrating consistent robustness and efficiency gains from sparse modular computation.

The scarcity of training data presents a fundamental challenge in applying deep learning to archaeological artifact classification, particularly for the rare types of Chinese porcelain. This study investigates whether synthetic images generated through Stable Diffusion with Low-Rank Adaptation (LoRA) can effectively augment limited real datasets for multi-task CNN-based porcelain classification. Using MobileNetV3 with transfer learning, we conducted controlled experiments comparing models trained on pure real data against those trained on mixed real-synthetic datasets (95:5 and 90:10 ratios) across four classification tasks: dynasty, glaze, kiln and type identification. Results demonstrate task-specific benefits: type classification showed the most substantial improvement (5.5\% F1-macro increase with 90:10 ratio), while dynasty and kiln tasks exhibited modest gains (3-4\%), suggesting that synthetic augmentation effectiveness depends on the alignment between generated features and task-relevant visual signatures. Our work contributes practical guidelines for deploying generative AI in archaeological research, demonstrating both the potential and limitations of synthetic data when archaeological authenticity must be balanced with data diversity.

While Large Language Models demonstrate remarkable proficiency in high-level semantic planning, they remain limited in handling fine-grained, low-level web component manipulations. To address this limitation, extensive research has focused on enhancing model grounding capabilities through techniques such as Reinforcement Learning. However, rather than compelling agents to adapt to human-centric interfaces, we propose constructing interaction interfaces specifically optimized for agents. This paper introduces Component Interface for Agent (CI4A), a semantic encapsulation mechanism that abstracts the complex interaction logic of UI components into a set of unified tool primitives accessible to agents. We implemented CI4A within Ant Design, an industrial-grade front-end framework, covering 23 categories of commonly used UI components. Furthermore, we developed a hybrid agent featuring an action space that dynamically updates according to the page state, enabling flexible invocation of available CI4A tools. Leveraging the CI4A-integrated Ant Design, we refactored and upgraded the WebArena benchmark to evaluate existing SoTA methods. Experimental results demonstrate that the CI4A-based agent significantly outperforms existing approaches, achieving a new SoTA task success rate of 86.3%, alongside substantial improvements in execution efficiency.

Recent advances in text-to-music generation (TTM) have yielded high-quality results, but often at the cost of extensive compute and the use of large proprietary internal data. To improve the affordability and openness of TTM training, an open-source generative model backbone that is more training- and data-efficient is needed. In this paper, we constrain the number of trainable parameters in the generative model to match that of the MusicGen-small benchmark (with about 300M parameters), and replace its Transformer backbone with the emerging class of state-space models (SSMs). Specifically, we explore different SSM variants for sequence modeling, and compare a single-stage SSM-based design with a decomposable two-stage SSM/diffusion hybrid design. All proposed models are trained from scratch on a purely public dataset comprising 457 hours of CC-licensed music, ensuring full openness. Our experimental findings are three-fold. First, we show that SSMs exhibit superior training efficiency compared to the Transformer counterpart. Second, despite using only 9% of the FLOPs and 2% of the training data size compared to the MusicGen-small benchmark, our model achieves competitive performance in both objective metrics and subjective listening tests based on MusicCaps captions. Finally, our scaling-down experiment demonstrates that SSMs can maintain competitive performance relative to the Transformer baseline even at the same training budget (measured in iterations), when the model size is reduced to four times smaller. To facilitate the democratization of TTM research, the processed captions, model checkpoints, and source code are available on GitHub via the project page: https://lonian6.github.io/ssmttm/.

Achieving bound consistency for the no-overlap constraint is known to be NP-complete. Therefore, several polynomial-time tightening techniques, such as edge finding, not-first-not-last reasoning, and energetic reasoning, have been introduced for this constraint. In this work, we derive the first bound-consistent algorithm for the no-overlap constraint. By building on the no-overlap MDD defined by Ciré and van Hoeve, we extract bounds of the time window of the jobs, allowing us to tighten start and end times in time polynomial in the number of nodes of the MDD. Similarly, to bound the size and time-complexity, we limit the width of the MDD to a threshold, creating a relaxed MDD that can also be used to relax the bound-consistent filtering. Through experiments on a sequencing problem with time windows and a just-in-time objective ($1 \mid r_j, d_j, \bar{d}_j \mid \sum E_j + \sum T_j$), we observe that the proposed filtering, even with a threshold on the width, achieves a stronger reduction in the number of nodes visited in the search tree compared to the previously proposed precedence-detection algorithm of Ciré and van Hoeve. The new filtering also appears to be complementary to classical propagation methods for the no-overlap constraint, allowing a substantial reduction in both the number of nodes and the solving time on several instances.

Recognizing and navigating client resistance is critical for effective mental health counseling, yet detecting such behaviors is particularly challenging in text-based interactions. Existing NLP approaches oversimplify resistance categories, ignore the sequential dynamics of therapeutic interventions, and offer limited interpretability. To address these limitations, we propose PsyFIRE, a theoretically grounded framework capturing 13 fine-grained resistance behaviors alongside collaborative interactions. Based on PsyFIRE, we construct the ClientResistance corpus with 23,930 annotated utterances from real-world Chinese text-based counseling, each supported by context-specific rationales. Leveraging this dataset, we develop RECAP, a two-stage framework that detects resistance and fine-grained resistance types with explanations. RECAP achieves 91.25% F1 for distinguishing collaboration and resistance and 66.58% macro-F1 for fine-grained resistance categories classification, outperforming leading prompt-based LLM baselines by over 20 points. Applied to a separate counseling dataset and a pilot study with 62 counselors, RECAP reveals the prevalence of resistance, its negative impact on therapeutic relationships and demonstrates its potential to improve counselors' understanding and intervention strategies.

Movie dubbing is the task of synthesizing speech from scripts conditioned on video scenes, requiring accurate lip sync, faithful timbre transfer, and proper modeling of character identity and emotion. However, existing methods face two major limitations: (1) high-quality multimodal dubbing datasets are limited in scale, suffer from high word error rates, contain sparse annotations, rely on costly manual labeling, and are restricted to monologue scenes, all of which hinder effective model training; (2) existing dubbing models rely solely on the lip region to learn audio-visual alignment, which limits their applicability to complex live-action cinematic scenes, and exhibit suboptimal performance in lip sync, speech quality, and emotional expressiveness. To address these issues, we propose FunCineForge, which comprises an end-to-end production pipeline for large-scale dubbing datasets and an MLLM-based dubbing model designed for diverse cinematic scenes. Using the pipeline, we construct the first Chinese television dubbing dataset with rich annotations, and demonstrate the high quality of these data. Experiments across monologue, narration, dialogue, and multi-speaker scenes show that our dubbing model consistently outperforms SOTA methods in audio quality, lip sync, timbre transfer, and instruction following. Code and demos are available at https://anonymous.4open.science/w/FunCineForge.

Video summarization is a crucial technique for social understanding, enabling efficient browsing of massive multimedia content and extraction of key information from social platforms. Most existing unsupervised summarization methods rely on Generative Adversarial Networks (GANs) to enhance keyframe selection and generate coherent, video summaries through adversarial training. However, such approaches primarily exploit unimodal features, overlooking the guiding role of semantic information in keyframe selection, and often suffer from unstable training. To address these limitations, we propose a novel Semantic-Guided Unsupervised Video Summarization method. Specifically, we design a novel frame-level semantic alignment attention mechanism and integrate it into a keyframe selector, which guides the Transformer-based generator within the adversarial framework to better reconstruct videos. In addition, we adopt an incremental training strategy to progressively update the model components, effectively mitigating the instability of GAN training. Experimental results demonstrate that our approach achieves superior performance on multiple benchmark datasets.

In recent years, significant progress has been made on algorithms for learning optimal decision trees, primarily in the context of binary features. Extending these methods to continuous features remains substantially more challenging due to the large number of potential splits for each feature. Recently, an elegant exact algorithm was proposed for learning optimal decision trees with continuous features; however, the rapidly increasing computational time limits its practical applicability to shallow depths (typically 3 or 4). It relies on a depth-first search optimization strategy that fully optimizes the left subtree of each split before exploring the corresponding right subtree. While effective in finding optimal solutions given sufficient time, this strategy can lead to poor anytime behavior: when interrupted early, the best-found tree is often highly unbalanced and suboptimal. In such cases, purely greedy methods such as C4.5 may, paradoxically, yield better solutions. To address this limitation, we propose an anytime, yet complete approach leveraging limited discrepancy search, distributing the computational effort more evenly across the entire tree structure, and thus ensuring that a high-quality decision tree is available at any interruption point. Experimental results show that our approach outperforms the existing one in terms of anytime performance.

Answer Set Programming (ASP) is a popular declarative reasoning and problem solving approach in symbolic AI. Its rule-based formalism makes it inherently attractive for explainable and interpretive reasoning, which is gaining importance with the surge of Explainable AI (XAI). A number of explanation approaches and tools for ASP have been developed, which often tackle specific explanatory settings and may not cover all scenarios that ASP users encounter. In this survey, we provide, guided by an XAI perspective, an overview of types of ASP explanations in connection with user questions for explanation, and describe how their coverage by current theory and tools. Furthermore, we pinpoint gaps in existing ASP explanations approaches and identify research directions for future work.

Post-training pretrained Autoregressive models (ARMs) into Masked Diffusion models (MDMs) has emerged as a cost-effective strategy to overcome the limitations of sequential generation. However, the internal algorithmic transformations induced by this paradigm shift remain unexplored, leaving it unclear whether post-trained MDMs acquire genuine bidirectional reasoning capabilities or merely repackage autoregressive heuristics. In this work, we address this question by conducting a comparative circuit analysis of ARMs and their MDM counterparts. Our analysis reveals a systematic "mechanism shift" dependent on the structural nature of the task. Structurally, we observe a distinct divergence: while MDMs largely retain autoregressive circuitry for tasks dominated by local causal dependencies, they abandon initialized pathways for global planning tasks, exhibiting distinct rewiring characterized by increased early-layer processing. Semantically, we identify a transition from sharp, localized specialization in ARMs to distributed integration in MDMs. Through these findings, we conclude that diffusion post-training does not merely adapt model parameters but fundamentally reorganizes internal computation to support non-sequential global planning.

Chain-of-Thought (CoT) prompting has achieved remarkable success in unlocking the reasoning capabilities of Large Language Models (LLMs). Although CoT prompting enhances reasoning, its verbosity imposes substantial computational overhead. Recent works often focus exclusively on outcome alignment and lack supervision on the intermediate reasoning process. These deficiencies obscure the analyzability of the latent reasoning chain. To address these challenges, we introduce Render-of-Thought (RoT), the first framework to reify the reasoning chain by rendering textual steps into images, making the latent rationale explicit and traceable. Specifically, we leverage the vision encoders of existing Vision Language Models (VLMs) as semantic anchors to align the vision embeddings with the textual space. This design ensures plug-and-play implementation without incurring additional pre-training overhead. Extensive experiments on mathematical and logical reasoning benchmarks demonstrate that our method achieves 3-4x token compression and substantial inference acceleration compared to explicit CoT. Furthermore, it maintains competitive performance against other methods, validating the feasibility of this paradigm. Our code is available at https://github.com/TencentBAC/RoT

Federated learning (FL) enables collaborative model training without sharing raw data in edge environments, but is constrained by limited communication bandwidth and heterogeneous client data distributions. Prototype-based FL mitigates this issue by exchanging class-wise feature prototypes instead of full model parameters; however, existing methods still suffer from suboptimal generalization under severe communication constraints. In this paper, we propose RefProtoFL, a communication-efficient FL framework that integrates External-Referenced Prototype Alignment (ERPA) for representation consistency with Adaptive Probabilistic Update Dropping (APUD) for communication efficiency. Specifically, we decompose the model into a private backbone and a lightweight shared adapter, and restrict federated communication to the adapter parameters only. To further reduce uplink cost, APUD performs magnitude-aware Top-K sparsification, transmitting only the most significant adapter updates for server-side aggregation. To address representation inconsistency across heterogeneous clients, ERPA leverages a small server-held public dataset to construct external reference prototypes that serve as shared semantic anchors. For classes covered by public data, clients directly align local representations to public-induced prototypes, whereas for uncovered classes, alignment relies on server-aggregated global reference prototypes via weighted averaging. Extensive experiments on standard benchmarks demonstrate that RefProtoFL attains higher classification accuracy than state-of-the-art prototype-based FL baselines.

The effectiveness of collision-free trajectory planners depends on the quality and diversity of training data, especially for rare scenarios. A widely used approach to improve dataset diversity involves generating realistic synthetic traffic scenarios. However, producing such scenarios remains difficult due to the precision required when scripting them manually or generating them in a single pass. Natural language offers a flexible way to describe scenarios, but existing text-to-simulation pipelines often rely on static snippet retrieval, limited grammar, single-pass decoding, or lack robust executability checks. Moreover, they depend heavily on constrained LLM prompting with minimal post-processing. To address these limitations, we introduce ARISE - Adaptive Refinement and Iterative Scenario Engineering, a multi-stage tool that converts natural language prompts into executable Scenic scripts through iterative LLM-guided refinement. After each generation, ARISE tests script executability in simulation software, feeding structured diagnostics back to the LLM until both syntactic and functional requirements are met. This process significantly reduces the need for manual intervention. Through extensive evaluation, ARISE outperforms the baseline in generating semantically accurate and executable traffic scenarios with greater reliability and robustness.

Agentic workflows that use autonomous AI Agents powered by Large Language Models (LLMs) and Model Context Protocol (MCP) servers is rapidly rising. This introduces challenges in scalable cloud deployment and state management. Traditional hosting on Virtual Machines (VMs) is resource-intensive and lacks elasticity. Functions-as-a-Service (FaaS) platforms offer modularity, autoscaling and cost efficiency but are inherently stateless. In this paper, we present the FAME, a FaaS-based architecture for orchestrating MCP-enabled agentic workflows. FAME decomposes agentic patterns such as ReAct into composable agents: Planner, Actor and Evaluator, that are each a FaaS function built using LangGraph and are orchestrated as a FaaS workflow. This enables modular composition as AWS Step Functions and avoids function timeouts seen for monolithic agentic workflows. To address context persistence across user requests in a conversation, FAME automates agent memory persistence and injection using DynamoDB. It also optimizes MCP server deployment through AWS Lambda wrappers, caches tool outputs in S3 and proposes function fusion strategies. We evaluate FAME on two representative applications, on research paper summarization and log analytics, under diverse memory and caching configurations. Results show up to 13x latency reduction, 88% fewer input tokens and 66% in cost savings, along with improved workflow completion rates. This demonstrates the viability of serverless platforms for hosting complex, multi-agent AI workflows at scale.

We compare different circuits implementing distributed versions of quantum computations, using entangled pairs only, and using distributed fan-out operations (using GHZ states). We highlight the advantages of using distributed fan-out operations in terms of reductions in circuit depth and (possibly) entanglement resources. We note that distributed fan-out operations (or notably, distributed GHZ states) could be a ``primitive'' building block for distributed quantum operations in the same way as entangled pairs are, if distributed GHZ states could be realized efficiently.

AI models for drug discovery and chemical literature mining must interpret molecular images and generate outputs consistent with 3D geometry and stereochemistry. Most molecular language models rely on strings or graphs, while vision-language models often miss stereochemical details and struggle to map continuous 3D structures into discrete tokens. We propose DeepMoLM: Deep Molecular Language M odeling, a dual-view framework that grounds high-resolution molecular images in geometric invariants derived from molecular conformations. DeepMoLM preserves high-frequency evidence from 1024 $\times$ 1024 inputs, encodes conformer neighborhoods as discrete Extended 3-Dimensional Fingerprints, and fuses visual and geometric streams with cross-attention, enabling physically grounded generation without atom coordinates. DeepMoLM improves PubChem captioning with a 12.3% relative METEOR gain over the strongest generalist baseline while staying competitive with specialist methods. It produces valid numeric outputs for all property queries and attains MAE 13.64 g/mol on Molecular Weight and 37.89 on Complexity in the specialist setting. On ChEBI-20 description generation from images, it exceeds generalist baselines and matches state-of- the-art vision-language models. Code is available at https://github.com/1anj/DeepMoLM.

Software systems that run for long periods often suffer from software aging, which is typically caused by Aging-Related Bugs (ARBs). To mitigate the risk of ARBs early in the development phase, ARB prediction has been introduced into software aging research. However, due to the difficulty of collecting ARBs, within-project ARB prediction faces the challenge of data scarcity, leading to the proposal of cross-project ARB prediction. This task faces two major challenges: 1) domain adaptation issue caused by distribution difference between source and target projects; and 2) severe class imbalance between ARB-prone and ARB-free samples. Although various methods have been proposed for cross-project ARB prediction, existing approaches treat the input metrics independently and often neglect the rich inter-metric dependencies, which can lead to overlapping information and misjudgment of metric importance, potentially affecting the model's performance. Moreover, they typically use cross-entropy as the loss function during training, which cannot distinguish the difficulty of sample classification. To overcome these limitations, we propose ARFT-Transformer, a transformer-based cross-project ARB prediction framework that introduces a metric-level multi-head attention mechanism to capture metric interactions and incorporates Focal Loss function to effectively handle class imbalance. Experiments conducted on three large-scale open-source projects demonstrate that ARFT-Transformer on average outperforms state-of-the-art cross-project ARB prediction methods in both single-source and multi-source cases, achieving up to a 29.54% and 19.92% improvement in Balance metric.

Despite the widespread success of Graph Neural Networks (GNNs), understanding the reasons behind their specific predictions remains challenging. Existing explainability methods face a trade-off that gradient-based approaches are computationally efficient but often ignore structural interactions, while game-theoretic techniques capture interactions at the cost of high computational overhead and potential deviation from the model's true reasoning path. To address this gap, we propose FSX (Message Flow Sensitivity Enhanced Structural Explainer), a novel hybrid framework that synergistically combines the internal message flows of the model with a cooperative game approach applied to the external graph data. FSX first identifies critical message flows via a novel flow-sensitivity analysis: during a single forward pass, it simulates localized node perturbations and measures the resulting changes in message flow intensities. These sensitivity-ranked flows are then projected onto the input graph to define compact, semantically meaningful subgraphs. Within each subgraph, a flow-aware cooperative game is conducted, where node contributions are evaluated fairly through a Shapley-like value that incorporates both node-feature importance and their roles in sustaining or destabilizing the identified critical flows. Extensive evaluation across multiple datasets and GNN architectures demonstrates that FSX achieves superior explanation fidelity with significantly reduced runtime, while providing unprecedented insights into the structural logic underlying model predictions--specifically, how important sub-structures exert influence by governing the stability of key internal computational pathways.

Although text-to-audio generation has made remarkable progress in realism and diversity, the development of evaluation metrics has not kept pace. Widely-adopted approaches, typically based on embedding similarity like CLAPScore, effectively measure general relevance but remain limited in fine-grained semantic alignment and compositional reasoning. To address this, we introduce AQAScore, a backbone-agnostic evaluation framework that leverages the reasoning capabilities of audio-aware large language models (ALLMs). AQAScore reformulates assessment as a probabilistic semantic verification task; rather than relying on open-ended text generation, it estimates alignment by computing the exact log-probability of a "Yes" answer to targeted semantic queries. We evaluate AQAScore across multiple benchmarks, including human-rated relevance, pairwise comparison, and compositional reasoning tasks. Experimental results show that AQAScore consistently achieves higher correlation with human judgments than similarity-based metrics and generative prompting baselines, showing its effectiveness in capturing subtle semantic inconsistencies and scaling with the capability of underlying ALLMs.

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated significant improvement in offline video understanding. However, extending these capabilities to streaming video inputs, remains challenging, as existing models struggle to simultaneously maintain stable understanding performance, real-time responses, and low GPU memory overhead. To address this challenge, we propose HERMES, a novel training-free architecture for real-time and accurate understanding of video streams. Based on a mechanistic attention investigation, we conceptualize KV cache as a hierarchical memory framework that encapsulates video information across multiple granularities. During inference, HERMES reuses a compact KV cache, enabling efficient streaming understanding under resource constraints. Notably, HERMES requires no auxiliary computations upon the arrival of user queries, thereby guaranteeing real-time responses for continuous video stream interactions, which achieves 10$\times$ faster TTFT compared to prior SOTA. Even when reducing video tokens by up to 68% compared with uniform sampling, HERMES achieves superior or comparable accuracy across all benchmarks, with up to 11.4% gains on streaming datasets.

Document extraction is a core component of digital workflows, yet existing vision-language models (VLMs) predominantly favor high-resource languages. Thai presents additional challenges due to script complexity from non-latin letters, the absence of explicit word boundaries, and the prevalence of highly unstructured real-world documents, limiting the effectiveness of current open-source models. This paper presents Typhoon OCR, an open VLM for document extraction tailored for Thai and English. The model is fine-tuned from vision-language backbones using a Thai-focused training dataset. The dataset is developed using a multi-stage data construction pipeline that combines traditional OCR, VLM-based restructuring, and curated synthetic data. Typhoon OCR is a unified framework capable of text transcription, layout reconstruction, and document-level structural consistency. The latest iteration of our model, Typhoon OCR V1.5, is a compact and inference-efficient model designed to reduce reliance on metadata and simplify deployment. Comprehensive evaluations across diverse Thai document categories, including financial reports, government forms, books, infographics, and handwritten documents, show that Typhoon OCR achieves performance comparable to or exceeding larger frontier proprietary models, despite substantially lower computational cost. The results demonstrate that open vision-language OCR models can achieve accurate text extraction and layout reconstruction for Thai documents, reaching performance comparable to proprietary systems while remaining lightweight and deployable.

We present PCL-Reasoner-V1.5, a 32-billion-parameter large language model (LLM) for mathematical reasoning. The model is built upon Qwen2.5-32B and refined via supervised fine-tuning (SFT) followed by reinforcement learning (RL). A central innovation is our proposed offline RL method, which provides superior training stability and efficiency over standard online RL methods such as GRPO. Our model achieves state-of-the-art performance among models post-trained on Qwen2.5-32B, attaining average accuracies of 90.9% on AIME 2024 and 85.6% on AIME 2025. Our work demonstrates offline RL as a stable and efficient paradigm for advancing reasoning in LLMs. All experiments were conducted on Huawei Ascend 910C NPUs.

Fidelity estimation is a critical yet resource-intensive step in testing quantum programs on noisy intermediate-scale quantum (NISQ) devices, where the required number of measurements is difficult to predefine due to hardware noise, device heterogeneity, and transpilation-induced circuit transformations. We present QuFid, an adaptive and noise-aware framework that determines measurement budgets online by leveraging circuit structure and runtime statistical feedback. QuFid models a quantum program as a directed acyclic graph (DAG) and employs a control-flow-aware random walk to characterize noise propagation along gate dependencies. Backend-specific effects are captured via transpilation-induced structural deformation metrics, which are integrated into the random-walk formulation to induce a noise-propagation operator. Circuit complexity is then quantified through the spectral characteristics of this operator, providing a principled and lightweight basis for adaptive measurement planning. Experiments on 18 quantum benchmarks executed on IBM Quantum backends show that QuFid significantly reduces measurement cost compared to fixed-shot and learning-based baselines, while consistently maintaining acceptable fidelity bias.

Optimizing the advertiser's cumulative value of winning impressions under budget constraints poses a complex challenge in online advertising, under the paradigm of AI-Generated Bidding (AIGB). Advertisers often have personalized objectives but limited historical interaction data, resulting in few-shot scenarios where traditional reinforcement learning (RL) methods struggle to perform effectively. Large Language Models (LLMs) offer a promising alternative for AIGB by leveraging their in-context learning capabilities to generalize from limited data. However, they lack the numerical precision required for fine-grained optimization. To address this limitation, we introduce GRPO-Adaptive, an efficient LLM post-training strategy that enhances both reasoning and numerical precision by dynamically updating the reference policy during training. Built upon this foundation, we further propose DARA, a novel dual-phase framework that decomposes the decision-making process into two stages: a few-shot reasoner that generates initial plans via in-context prompting, and a fine-grained optimizer that refines these plans using feedback-driven reasoning. This separation allows DARA to combine LLMs' in-context learning strengths with precise adaptability required by AIGB tasks. Extensive experiments on both real-world and synthetic data environments demonstrate that our approach consistently outperforms existing baselines in terms of cumulative advertiser value under budget constraints.

Optimally sequencing experimental assays in drug discovery is a high-stakes planning problem under severe uncertainty and resource constraints. A primary obstacle for standard reinforcement learning (RL) is the absence of an explicit environment simulator or transition data $(s, a, s')$; planning must rely solely on a static database of historical outcomes. We introduce the Implicit Bayesian Markov Decision Process (IBMDP), a model-based RL framework designed for such simulator-free settings. IBMDP constructs a case-guided implicit model of transition dynamics by forming a nonparametric belief distribution using similar historical outcomes. This mechanism enables Bayesian belief updating as evidence accumulates and employs ensemble MCTS planning to generate stable policies that balance information gain toward desired outcomes with resource efficiency. We validate IBMDP through comprehensive experiments. On a real-world central nervous system (CNS) drug discovery task, IBMDP reduced resource consumption by up to 92\% compared to established heuristics while maintaining decision confidence. To rigorously assess decision quality, we also benchmarked IBMDP in a synthetic environment with a computable optimal policy. Our framework achieves significantly higher alignment with this optimal policy than a deterministic value iteration alternative that uses the same similarity-based model, demonstrating the superiority of our ensemble planner. IBMDP offers a practical solution for sequential experimental design in data-rich but simulator-poor domains.

Proximal Policy Optimization (PPO) is a widely used reinforcement learning algorithm known for its stability and sample efficiency, but it often suffers from premature convergence due to limited exploration. In this paper, we propose POEM (Proximal Policy Optimization with Evolutionary Mutations), a novel modification to PPO that introduces an adaptive exploration mechanism inspired by evolutionary algorithms. POEM enhances policy diversity by monitoring the Kullback-Leibler (KL) divergence between the current policy and a moving average of previous policies. When policy changes become minimal, indicating stagnation, POEM triggers an adaptive mutation of policy parameters to promote exploration. We evaluate POEM on four OpenAI Gym environments: CarRacing, MountainCar, BipedalWalker, and LunarLander. Through extensive fine-tuning using Bayesian optimization techniques and statistical testing using Welch's t-test, we find that POEM significantly outperforms PPO on three of the four tasks (BipedalWalker: t=-2.0642, p=0.0495; CarRacing: t=-6.3987, p=0.0002; MountainCar: t=-6.2431, p<0.0001), while performance on LunarLander is not statistically significant (t=-1.8707, p=0.0778). Our results highlight the potential of integrating evolutionary principles into policy gradient methods to overcome exploration-exploitation tradeoffs.

Autonomous driving is a highly challenging domain that requires reliable perception and safe decision-making in complex scenarios. Recent vision-language models (VLMs) demonstrate reasoning and generalization abilities, opening new possibilities for autonomous driving; however, existing benchmarks and metrics overemphasize perceptual competence and fail to adequately assess decision-making processes. In this work, we present AutoDriDM, a decision-centric, progressive benchmark with 6,650 questions across three dimensions - Object, Scene, and Decision. We evaluate mainstream VLMs to delineate the perception-to-decision capability boundary in autonomous driving, and our correlation analysis reveals weak alignment between perception and decision-making performance. We further conduct explainability analyses of models' reasoning processes, identifying key failure modes such as logical reasoning errors, and introduce an analyzer model to automate large-scale annotation. AutoDriDM bridges the gap between perception-centered and decision-centered evaluation, providing guidance toward safer and more reliable VLMs for real-world autonomous driving.

Reinforcement Learning with Verifiable Rewards (RLVR) has demonstrated promising gains in enhancing the reasoning capabilities of large language models. However, its dependence on domain-specific verifiers significantly restricts its applicability to open and general domains. Recent efforts such as RLPR have extended RLVR to general domains, enabling training on broader datasets and achieving improvements over RLVR. However, a notable limitation of these methods is their tendency to overfit to reference answers, which constrains the model's ability to generate diverse outputs. This limitation is particularly pronounced in open-ended tasks such as writing, where multiple plausible answers exist. To address this, we propose DARL, a simple yet effective reinforcement learning framework that encourages the generation of diverse answers within a controlled deviation range from the reference while preserving alignment with it. Our framework is fully compatible with existing general reinforcement learning methods and can be seamlessly integrated without additional verifiers. Extensive experiments on thirteen benchmarks demonstrate consistent improvements in reasoning performance. Notably, DARL surpasses RLPR, achieving average gains of 1.3 points on six reasoning benchmarks and 9.5 points on seven general benchmarks, highlighting its effectiveness in improving both reasoning accuracy and output diversity.

Despite substantial progress in fact-verification benchmarks, claims grounded in large-scale structured data remain underexplored. In this work, we introduce ClaimDB, the first fact-verification benchmark where the evidence for claims is derived from compositions of millions of records and multiple tables. ClaimDB consists of 80 unique real-life databases covering a wide range of domains, from governance and healthcare to media, education and the natural sciences. At this scale, verification approaches that rely on "reading" the evidence break down, forcing a timely shift toward reasoning in executable programs. We conduct extensive experiments with 30 state-of-the-art proprietary and open-source (below 70B) LLMs and find that none exceed 83% accuracy, with more than half below 55%. Our analysis also reveals that both closed- and open-source models struggle with abstention -- the ability to admit that there is no evidence to decide -- raising doubts about their reliability in high-stakes data analysis. We release the benchmark, code, and the LLM leaderboard at https://claimdb.github.io .

Semantic ID learning is a key interface in Generative Recommendation (GR) models, mapping items to discrete identifiers grounded in side information, most commonly via a pretrained text encoder. However, these text encoders are primarily optimized for well-formed natural language. In real-world recommendation data, item descriptions are often symbolic and attribute-centric, containing numerals, units, and abbreviations. These text encoders can break these signals into fragmented tokens, weakening semantic coherence and distorting relationships among attributes. Worse still, when moving to multimodal GR, relying on standard text encoders introduces an additional obstacle: text and image embeddings often exhibit mismatched geometric structures, making cross-modal fusion less effective and less stable. In this paper, we revisit representation design for Semantic ID learning by treating text as a visual signal. We conduct a systematic empirical study of OCR-based text representations, obtained by rendering item descriptions into images and encoding them with vision-based OCR models. Experiments across four datasets and two generative backbones show that OCR-text consistently matches or surpasses standard text embeddings for Semantic ID learning in both unimodal and multimodal settings. Furthermore, we find that OCR-based Semantic IDs remain robust under extreme spatial-resolution compression, indicating strong robustness and efficiency in practical deployments.

Tool-Integrated Reasoning (TIR) has significantly enhanced the capabilities of Large Language Models (LLMs), yet current agents tend to exhibit cognitive offloading, redundantly invoking external tools even for simple tasks. In this paper, we suggest that true agentic intelligence requires not just tool invocation, but the adaptive wisdom to discern when to use them. We propose AdaTIR, a framework that shifts the paradigm from static tool invocation to difficulty-aware reasoning internalization. By introducing a difficulty-aware efficiency reward, AdaTIR dynamically adjusts tool budgets based on task complexity--internalizing reasoning for simple tasks while selectively invoking tools for complex tasks. Furthermore, we identify a sign reversal problem where tool penalties outweigh correctness rewards, mistakenly penalizing correct rollouts with negative advantages. To resolve this, we propose Clipped Advantage Shaping (CAS), which ensures that correctness remains the primary objective while using efficiency as a secondary constraint. Empirical results demonstrate that AdaTIR reduces tool calls by up to 97.6% on simple tasks and 28.2% on complex challenges while maintaining or enhancing accuracy. Notably, AdaTIR successfully internalizes reasoning, outperforming baselines by 4.8% on AIME 2024 even when tool access is strictly disabled.

Training Large Reasoning Model (LRM) is usually unstable and unpredictable, especially on hard problems or weak foundation models. We found that the current post-training scaling strategy can still improve on these cases. We propose CoScale-RL, a novel scaling strategy with better data and computational efficiency. We first scale up solutions to make problems solvable. The core idea is to collect multiple solutions for each problem, rather than simply enlarging the dataset. Then, we scale up rollout computation to stabilize Reinforcement Learning. We further leverage a model merge technique called Re-distillation to sustain or even improve computational efficiency when scaling up. Our method significantly improves data and computational efficiency, with an average 3.76$\times$ accuracy improvement on four benchmarks. CoScale-RL is able to improve an LRM's ability boundary without an extensive SFT dataset. Our method provides a new scaling direction to further improve LRM's reasoning ability.

Graph unlearning (GU), which removes nodes, edges, or features from trained graph neural networks (GNNs), is crucial in Web applications where graph data may contain sensitive, mislabeled, or malicious information. However, existing GU methods lack a clear understanding of the key factors that determine unlearning effectiveness, leading to three fundamental limitations: (1) impractical and inaccurate GU difficulty assessment due to test-access requirements and invalid assumptions, (2) ineffectiveness on hard-to-unlearn tasks, and (3) misaligned evaluation protocols that overemphasize easy tasks and fail to capture true forgetting capability. To address these issues, we establish GNN memorization as a new perspective for understanding graph unlearning and propose MGU, a Memorization-guided Graph Unlearning framework. MGU achieves three key advances: it provides accurate and practical difficulty assessment across different GU tasks, develops an adaptive strategy that dynamically adjusts unlearning objectives based on difficulty levels, and establishes a comprehensive evaluation protocol that aligns with practical requirements. Extensive experiments on ten real-world graphs demonstrate that MGU consistently outperforms state-of-the-art baselines in forgetting quality, computational efficiency, and utility preservation.

Symbolic Regression aims to automatically identify compact and interpretable mathematical expressions that model the functional relationship between input and output variables. Most existing search-based symbolic regression methods typically rely on the fitting error to inform the search process. However, in the vast expression space, numerous candidate expressions may exhibit similar error values while differing substantially in structure, leading to ambiguous search directions and hindering convergence to the underlying true function. To address this challenge, we propose a novel framework named EGRL-SR (Experience-driven Goal-conditioned Reinforcement Learning for Symbolic Regression). In contrast to traditional error-driven approaches, EGRL-SR introduces a new perspective: leveraging precise historical trajectories and optimizing the action-value network to proactively guide the search process, thereby achieving a more robust expression search. Specifically, we formulate symbolic regression as a goal-conditioned reinforcement learning problem and incorporate hindsight experience replay, allowing the action-value network to generalize common mapping patterns from diverse input-output pairs. Moreover, we design an all-point satisfaction binary reward function that encourages the action-value network to focus on structural patterns rather than low-error expressions, and concurrently propose a structure-guided heuristic exploration strategy to enhance search diversity and space coverage. Experiments on public benchmarks show that EGRL-SR consistently outperforms state-of-the-art methods in recovery rate and robustness, and can recover more complex expressions under the same search budget. Ablation results validate that the action-value network effectively guides the search, with both the reward function and the exploration strategy playing critical roles.

Large language models (LLMs) are increasingly used as judges to evaluate agent performance, particularly in non-verifiable settings where judgments rely on agent trajectories including chain-of-thought (CoT) reasoning. This paradigm implicitly assumes that the agent's CoT faithfully reflects both its internal reasoning and the underlying environment state. We show this assumption is brittle: LLM judges are highly susceptible to manipulation of agent reasoning traces. By systematically rewriting agent CoTs while holding actions and observations fixed, we demonstrate that manipulated reasoning alone can inflate false positive rates of state-of-the-art VLM judges by up to 90% across 800 trajectories spanning diverse web tasks. We study manipulation strategies spanning style-based approaches that alter only the presentation of reasoning and content-based approaches that fabricate signals of task progress, and find that content-based manipulations are consistently more effective. We evaluate prompting-based techniques and scaling judge-time compute, which reduce but do not fully eliminate susceptibility to manipulation. Our findings reveal a fundamental vulnerability in LLM-based evaluation and highlight the need for judging mechanisms that verify reasoning claims against observable evidence.

Federated Rank Learning (FRL) is a promising Federated Learning (FL) paradigm designed to be resilient against model poisoning attacks due to its discrete, ranking-based update mechanism. Unlike traditional FL methods that rely on model updates, FRL leverages discrete rankings as a communication parameter between clients and the server. This approach significantly reduces communication costs and limits an adversary's ability to scale or optimize malicious updates in the continuous space, thereby enhancing its robustness. This makes FRL particularly appealing for applications where system security and data privacy are crucial, such as web-based auction and bidding platforms. While FRL substantially reduces the attack surface, we demonstrate that it remains vulnerable to a new class of local model poisoning attack, i.e., fine-grained control attacks. We introduce the Edge Control Attack (ECA), the first fine-grained control attack tailored to ranking-based FL frameworks. Unlike conventional denial-of-service (DoS) attacks that cause conspicuous disruptions, ECA enables an adversary to precisely degrade a competitor's accuracy to any target level while maintaining a normal-looking convergence trajectory, thereby avoiding detection. ECA operates in two stages: (i) identifying and manipulating Ascending and Descending Edges to align the global model with the target model, and (ii) widening the selection boundary gap to stabilize the global model at the target accuracy. Extensive experiments across seven benchmark datasets and nine Byzantine-robust aggregation rules (AGRs) show that ECA achieves fine-grained accuracy control with an average error of only 0.224%, outperforming the baseline by up to 17x. Our findings highlight the need for stronger defenses against advanced poisoning attacks. Our code is available at: https://github.com/Chenzh0205/ECA

Learning Path Recommendation (LPR) aims to generate personalized sequences of learning items that maximize long-term learning effect while respecting pedagogical principles and operational constraints. Although large language models (LLMs) offer rich semantic understanding for free-form recommendation, applying them to long-horizon LPR is challenging due to (i) misalignment with pedagogical objectives such as the Zone of Proximal Development (ZPD) under sparse, delayed feedback, (ii) scarce and costly expert demonstrations, and (iii) multi-objective interactions among learning effect, difficulty scheduling, length controllability, and trajectory diversity. To address these issues, we propose IB-GRPO (Indicator-Based Group Relative Policy Optimization), an indicator-guided alignment approach for LLM-based LPR. To mitigate data scarcity, we construct hybrid expert demonstrations via Genetic Algorithm search and teacher RL agents and warm-start the LLM with supervised fine-tuning. Building on this warm-start, we design a within-session ZPD alignment score for difficulty scheduling. IB-GRPO then uses the $I_{ε+}$ dominance indicator to compute group-relative advantages over multiple objectives, avoiding manual scalarization and improving Pareto trade-offs. Experiments on ASSIST09 and Junyi using the KES simulator with a Qwen2.5-7B backbone show consistent improvements over representative RL and LLM baselines.

Qualitative research often contains personal, contextual, and organizational details that pose privacy risks if not handled appropriately. Manual anonymization is time-consuming, inconsistent, and frequently omits critical identifiers. Existing automated tools tend to rely on pattern matching or fixed rules, which fail to capture context and may alter the meaning of the data. This study uses local LLMs to build a reliable, repeatable, and context-aware anonymization process for detecting and anonymizing sensitive data in qualitative transcripts. We introduce a Structured Framework for Adaptive Anonymizer (SFAA) that includes three steps: detection, classification, and adaptive anonymization. The SFAA incorporates four anonymization strategies: rule-based substitution, context-aware rewriting, generalization, and suppression. These strategies are applied based on the identifier type and the risk level. The identifiers handled by the SFAA are guided by major international privacy and research ethics standards, including the GDPR, HIPAA, and OECD guidelines. This study followed a dual-method evaluation that combined manual and LLM-assisted processing. Two case studies were used to support the evaluation. The first includes 82 face-to-face interviews on gamification in organizations. The second involves 93 machine-led interviews using an AI-powered interviewer to test LLM awareness and workplace privacy. Two local models, LLaMA and Phi were used to evaluate the performance of the proposed framework. The results indicate that the LLMs found more sensitive data than a human reviewer. Phi outperformed LLaMA in finding sensitive data, but made slightly more errors. Phi was able to find over 91% of the sensitive data and 94.8% kept the same sentiment as the original text, which means it was very accurate, hence, it does not affect the analysis of the qualitative data.

Natural walking enhances immersion in virtual environments (VEs), but physical space limitations and obstacles hinder exploration, especially in large virtual scenes. Redirected Walking (RDW) techniques mitigate this by subtly manipulating the virtual camera to guide users away from physical collisions within pre-defined VEs. However, RDW efficacy diminishes significantly when substantial geometric divergence exists between the physical and virtual environments, leading to unavoidable collisions. Existing scene generation methods primarily focus on object relationships or layout aesthetics, often neglecting the crucial aspect of physical compatibility required for effective RDW. To address this, we introduce HCVR (High Compatibility Virtual Reality Environment Generation), a novel framework that generates virtual scenes inherently optimized for alignment-based RDW controllers. HCVR first employs ENI++, a novel, boundary-sensitive metric to evaluate the incompatibility between physical and virtual spaces by comparing rotation-sensitive visibility polygons. Guided by the ENI++ compatibility map and user prompts, HCVR utilizes a Large Language Model (LLM) for context-aware 3D asset retrieval and initial layout generation. The framework then strategically adjusts object selection, scaling, and placement to maximize coverage of virtually incompatible regions, effectively guiding users towards RDW-feasible paths. User studies evaluating physical collisions and layout quality demonstrate HCVR's effectiveness with HCVR-generated scenes, resulting in 22.78 times fewer physical collisions and received 35.89\% less on ENI++ score compared to LLM-based generation with RDW, while also receiving 12.5\% higher scores on user feedback to layout design.

Supervised deep learning models often achieve excellent performance within their training distribution but struggle to generalize beyond it. In cancer histopathology, for example, a convolutional neural network (CNN) may classify cancer severity accurately for cancer types represented in its training data, yet fail on related but unseen types. Although adenocarcinomas from different organs share morphological features that might support limited cross-domain generalization, addressing domain shift directly is necessary for robust performance. Domain adaptation offers a way to transfer knowledge from labeled data in one cancer type to unlabeled data in another, helping mitigate the scarcity of annotated medical images. This work evaluates cross-domain classification performance among lung, colon, breast, and kidney adenocarcinomas. A ResNet50 trained on any single adenocarcinoma achieves over 98% accuracy on its own domain but shows minimal generalization to others. Ensembling multiple supervised models does not resolve this limitation. In contrast, converting the ResNet50 into a domain adversarial neural network (DANN) substantially improves performance on unlabeled target domains. A DANN trained on labeled breast and colon data and adapted to unlabeled lung data reaches 95.56% accuracy. We also examine the impact of stain normalization on domain adaptation. Its effects vary by target domain: for lung, accuracy drops from 95.56% to 66.60%, while for breast and colon targets, stain normalization boosts accuracy from 49.22% to 81.29% and from 78.48% to 83.36%, respectively. Finally, using Integrated Gradients reveals that DANNs consistently attribute importance to biologically meaningful regions such as densely packed nuclei, indicating that the model learns clinically relevant features and can apply them to unlabeled cancer types.

Object detection in video and image surveillance is a well-established yet rapidly evolving task, strongly influenced by recent deep learning advancements. This review summarises modern techniques by examining architectural innovations, generative model integration, and the use of temporal information to enhance robustness and accuracy. Unlike earlier surveys, it classifies methods based on core architectures, data processing strategies, and surveillance specific challenges such as dynamic environments, occlusions, lighting variations, and real-time requirements. The primary goal is to evaluate the current effectiveness of semantic object detection, while secondary aims include analysing deep learning models and their practical applications. The review covers CNN-based detectors, GAN-assisted approaches, and temporal fusion methods, highlighting how generative models support tasks such as reconstructing missing frames, reducing occlusions, and normalising illumination. It also outlines preprocessing pipelines, feature extraction progress, benchmarking datasets, and comparative evaluations. Finally, emerging trends in low-latency, efficient, and spatiotemporal learning approaches are identified for future research.

Given a monocular video, the goal of video re-rendering is to generate views of the scene from a novel camera trajectory. Existing methods face two distinct challenges. Geometrically unconditioned models lack spatial awareness, leading to drift and deformation under viewpoint changes. On the other hand, geometrically-conditioned models depend on estimated depth and explicit reconstruction, making them susceptible to depth inaccuracies and calibration errors. We propose to address these challenges by using the implicit geometric knowledge embedded in the latent space of a large 4D reconstruction model to condition the video generation process. These latents capture scene structure in a continuous space without explicit reconstruction. Therefore, they provide a flexible representation that allows the pretrained diffusion prior to regularize errors more effectively. By jointly conditioning on these latents and source camera poses, we demonstrate that our model achieves state-of-the-art results on the video re-rendering task. Project webpage is https://lavr-4d-scene-rerender.github.io/

The ongoing decarbonisation of power systems is driving an increasing reliance on distributed energy resources, which introduces complex and nonlinear interactions that are difficult to capture in conventional optimisation models. As a result, machine learning based surrogate modelling has emerged as a promising approach, but integrating machine learning models such as ReLU deep neural networks (DNNs) directly into optimisation often results in nonconvex and computationally intractable formulations. This paper proposes a linear programming (LP) reformulation for a class of convexified ReLU DNNs with non-negative weight matrices beyond the first layer, enabling a tight and tractable embedding of learned surrogate models in optimisation. We evaluate the method using a case study on learning the prosumer's responsiveness within an aggregator bidding problem in the Danish tertiary capacity market. The proposed reformulation is benchmarked against state-of-the-art alternatives, including piecewise linearisation (PWL), MIP-based embedding, and other LP relaxations. Across multiple neural network architectures and market scenarios, the convexified ReLU DNN achieves solution quality comparable to PWL and MIP-based reformulations while significantly improving computational performance and preserving model fidelity, unlike penalty-based reformulations. The results demonstrate that convexified ReLU DNNs offer a scalable and reliable methodology for integrating learned surrogate models in optimisation, with applicability to a wide range of emerging power system applications.

Hyperparameter tuning is a critical yet computationally expensive step in training neural networks, particularly when the search space is high dimensional and nonconvex. Metaheuristic optimization algorithms are often used for this purpose due to their derivative free nature and robustness against local optima. In this work, we propose Golden Eagle Genetic Optimization (GEGO), a hybrid metaheuristic that integrates the population movement strategy of Golden Eagle Optimization with the genetic operators of selection, crossover, and mutation. The main novelty of GEGO lies in embedding genetic operators directly into the iterative search process of GEO, rather than applying them as a separate evolutionary stage. This design improves population diversity during search and reduces premature convergence while preserving the exploration behavior of GEO. GEGO is evaluated on standard unimodal, multimodal, and composite benchmark functions from the CEC2017 suite, where it consistently outperforms its constituent algorithms and several classical metaheuristics in terms of solution quality and robustness. The algorithm is further applied to hyperparameter tuning of artificial neural networks on the MNIST dataset, where GEGO achieves improved classification accuracy and more stable convergence compared to GEO and GA. These results indicate that GEGO provides a balanced exploration-exploitation tradeoff and is well suited for hyperparameter optimization under constrained computational settings.

The rapid advancement of Large Language Model (LLM)-based Multi-Agent Systems (MAS) has introduced significant security vulnerabilities, where malicious influence can propagate virally through inter-agent communication. Conventional safeguards often rely on a binary paradigm that strictly distinguishes between benign and attack agents, failing to account for infected agents i.e., benign entities converted by attack agents. In this paper, we propose Infection-Aware Guard, INFA-Guard, a novel defense framework that explicitly identifies and addresses infected agents as a distinct threat category. By leveraging infection-aware detection and topological constraints, INFA-Guard accurately localizes attack sources and infected ranges. During remediation, INFA-Guard replaces attackers and rehabilitates infected ones, avoiding malicious propagation while preserving topological integrity. Extensive experiments demonstrate that INFA-Guard achieves state-of-the-art performance, reducing the Attack Success Rate (ASR) by an average of 33%, while exhibiting cross-model robustness, superior topological generalization, and high cost-effectiveness.

Reliable forecasting of prosumer flexibility is critical for demand response aggregators participating in frequency controlled ancillary services market, where strict reliability requirements such as the P90 standard are enforced. Limited historical data, dependence on exogeneous factors, and heterogenous prosumer behaviour introduce significant epistemic uncertainty, making deterministic or poorly calibrated probabilistic models unsuitable for market bidding. This paper proposes the use of scalable uncertainty quantification framework that integrates Monte Carlo dropout (MCD) with conformal prediction (CP) to produce calibrated, finite sample prediction intervals for aggregated prosumer flexibility. The proposed framework is applied to a behind-the-meter aggregator participating in the Danish manual frequency restoration reserve capacity market. A large-scale synthetic dataset is generated using a modified industry-grade home energy management system, combined with publicly available load, solar, price, activation and device-level data. The resulting machine learning surrogate model captures aggregate prosumer price responsiveness and provides uncertainty-aware estimates suitable for market bidding. Multiple multivariate CP strategies are evaluated and benchmarked against conventional MCD-based methods. Results show that standalone MCD systematically overestimates available flexibility and violates P90 compliance, whereas the proposed MCD-CP framework achieves reliable coverage with controlled conservatism. When embedded in aggregator bidding model, conformalised methods substantially reduce overbidding risk and achieve upto 70% of perfect-information profit while satisfying regulatory reliability constraints, providing practical, computationally efficient, and market-compliant solution for aggregator flexibility forecasting under uncertainty.

Graph-based retrieval-augmented generation (GraphRAG) systems construct knowledge graphs over document collections to support multi-hop reasoning. While prior work shows that GraphRAG responses may leak retrieved subgraphs, the feasibility of query-efficient reconstruction of the hidden graph structure remains unexplored under realistic query budgets. We study a budget-constrained black-box setting where an adversary adaptively queries the system to steal its latent entity-relation graph. We propose AGEA (Agentic Graph Extraction Attack), a framework that leverages a novelty-guided exploration-exploitation strategy, external graph memory modules, and a two-stage graph extraction pipeline combining lightweight discovery with LLM-based filtering. We evaluate AGEA on medical, agriculture, and literary datasets across Microsoft-GraphRAG and LightRAG systems. Under identical query budgets, AGEA significantly outperforms prior attack baselines, recovering up to 90% of entities and relationships while maintaining high precision. These results demonstrate that modern GraphRAG systems are highly vulnerable to structured, agentic extraction attacks, even under strict query limits.

This work addresses the computational challenge of enforcing privacy for agentic Large Language Models (LLMs), where privacy is governed by the contextual integrity framework. Indeed, existing defenses rely on LLM-mediated checking stages that add substantial latency and cost, and that can be undermined in multi-turn interactions through manipulation or benign-looking conversational scaffolding. Contrasting this background, this paper makes a key observation: internal representations associated with privacy-violating intent can be separated from benign requests using linear structure. Using this insight, the paper proposes NeuroFilter, a guardrail framework that operationalizes contextual integrity by mapping norm violations to simple directions in the model's activation space, enabling detection even when semantic filters are bypassed. The proposed filter is also extended to capture threats arising during long conversations using the concept of activation velocity, which measures cumulative drift in internal representations across turns. A comprehensive evaluation across over 150,000 interactions and covering models from 7B to 70B parameters, illustrates the strong performance of NeuroFilter in detecting privacy attacks while maintaining zero false positives on benign prompts, all while reducing the computational inference cost by several orders of magnitude when compared to LLM-based agentic privacy defenses.

Large language models (LLMs) reason over discrete token ID sequences, yet modern subword tokenizers routinely produce non-unique encodings: multiple token ID sequences can detokenize to identical surface strings. This representational mismatch creates an unmeasured fragility wherein reasoning processes can fail. LLMs may treat two internal representations as distinct "words" even when they are semantically identical at the text level. In this work, we show that tokenization can betray LLM reasoning through one-to-many token ID mappings. We introduce a tokenization-consistency probe that requires models to replace designated target words in context while leaving all other content unchanged. The task is intentionally simple at the surface level, enabling us to attribute failures to tokenizer-detokenizer artifacts rather than to knowledge gaps or parameter limitations. Through analysis of over 11000 replacement trials across state-of-the-art open-source LLMs, we find a non-trivial rate of outputs exhibit phantom edits: cases where models operate under the illusion of correct reasoning, a phenomenon arising from tokenizer-induced representational defects. We further analyze these cases and provide a taxonomy of eight systematic tokenizer artifacts, including whitespace-boundary shifts and intra-word resegmentation. These findings indicate that part of apparent reasoning deficiency originates in the tokenizer layer, motivating tokenizer-level remedies before incurring the cost of training ever-larger models on ever-larger corpora.

Missing data in single-cell sequencing datasets poses significant challenges for extracting meaningful biological insights. However, existing imputation approaches, which often assume uniformity and data completeness, struggle to address cases with large patches of missing data. In this paper, we present CROT, an optimal transport-based imputation algorithm designed to handle patch-based missing data in tabular formats. Our approach effectively captures the underlying data structure in the presence of significant missingness. Notably, it achieves superior imputation accuracy while significantly reducing runtime, demonstrating its scalability and efficiency for large-scale datasets. This work introduces a robust solution for imputation in heterogeneous, high-dimensional datasets with structured data absence, addressing critical challenges in both biological and clinical data analysis. Our code is available at Anomalous Github.

While multi-agent systems (MAS) promise elevated intelligence through coordination of agents, current approaches to automatic MAS design under-deliver. Such shortcomings stem from two key factors: (1) methodological complexity - agent orchestration is performed using sequential, code-level execution that limits global system-level holistic reasoning and scales poorly with agent complexity - and (2) efficacy uncertainty - MAS are deployed without understanding if there are tangible benefits compared to single-agent systems (SAS). We propose MAS-Orchestra, a training-time framework that formulates MAS orchestration as a function-calling reinforcement learning problem with holistic orchestration, generating an entire MAS at once. In MAS-Orchestra, complex, goal-oriented sub-agents are abstracted as callable functions, enabling global reasoning over system structure while hiding internal execution details. To rigorously study when and why MAS are beneficial, we introduce MASBENCH, a controlled benchmark that characterizes tasks along five axes: Depth, Horizon, Breadth, Parallel, and Robustness. Our analysis reveals that MAS gains depend critically on task structure, verification protocols, and the capabilities of both orchestrator and sub-agents, rather than holding universally. Guided by these insights, MAS-Orchestra achieves consistent improvements on public benchmarks including mathematical reasoning, multi-hop QA, and search-based QA. Together, MAS-Orchestra and MASBENCH enable better training and understanding of MAS in the pursuit of multi-agent intelligence.

We propose a stochastic trust-region method for unconstrained nonconvex optimization that incorporates stochastic variance-reduced gradients (SVRG) to accelerate convergence. Unlike classical trust-region methods, the proposed algorithm relies solely on stochastic gradient information and does not require function value evaluations. The trust-region radius is adaptively adjusted based on a radius-control parameter and the stochastic gradient estimate. Under mild assumptions, we establish that the algorithm converges in expectation to a first-order stationary point. Moreover, the method achieves iteration and sample complexity bounds that match those of SVRG-based first-order methods, while allowing stochastic and potentially gradient-dependent second-order information. Extensive numerical experiments demonstrate that incorporating SVRG accelerates convergence, and that the use of trust-region methods and Hessian information further improves performance. We also highlight the impact of batch size and inner-loop length on efficiency, and show that the proposed method outperforms SGD and Adam on several machine learning tasks.

Remote direct memory access (RDMA) allows a machine to directly read from and write to the memory of remote machine, enabling high-throughput, low-latency data transfer. Ensuring correctness of RDMA programs has only recently become possible with the formalisation of $\text{RDMA}^\text{TSO}$ semantics (describing the behaviour of RDMA networking over a TSO CPU). However, this semantics currently lacks a formalisation of remote synchronisation, meaning that the implementations of common abstractions such as locks cannot be verified. In this paper, we close this gap by presenting $\text{RDMA}^{\text{TSO}}_{\text{RMW}}$, the first semantics for remote `read-modify-write' (RMW) instructions over TSO. It turns out that remote RMW operations are weak and only ensure atomicity against other remote RMWs. We therefore build a set of composable synchronisation abstractions starting with the $\text{RDMA}^{\text{WAIT}}_{\text{RMW}}$ library. Underpinned by $\text{RDMA}^{\text{WAIT}}_{\text{RMW}}$, we then specify, implement and verify three classes of remote locks that are suitable for different scenarios. Additionally, we develop the notion of a strong RDMA model, $\text{RDMA}^{\text{SC}}_{\text{RMW}}$, which is akin to sequential consistency in shared memory architectures. Our libraries are built to be compatible with an existing set of high-performance libraries called LOCO, which ensures compositionality and verifiability.

The increasing availability of high-resolution satellite imagery, together with advances in deep learning, creates new opportunities for enhancing forest monitoring workflows. Two central challenges in this domain are pixel-level change detection and semantic change interpretation, particularly for complex forest dynamics. While large language models (LLMs) are increasingly adopted for data exploration, their integration with vision-language models (VLMs) for remote sensing image change interpretation (RSICI) remains underexplored, especially beyond urban environments. We introduce Forest-Chat, an LLM-driven agent designed for integrated forest change analysis. The proposed framework enables natural language querying and supports multiple RSICI tasks, including change detection, change captioning, object counting, deforestation percentage estimation, and change reasoning. Forest-Chat builds upon a multi-level change interpretation (MCI) vision-language backbone with LLM-based orchestration, and incorporates zero-shot change detection via a foundation change detection model together with an interactive point-prompt interface to support fine-grained user guidance. To facilitate adaptation and evaluation in forest environments, we introduce the Forest-Change dataset, comprising bi-temporal satellite imagery, pixel-level change masks, and multi-granularity semantic change captions generated through a combination of human annotation and rule-based methods. Experimental results demonstrate that Forest-Chat achieves strong performance on Forest-Change and on LEVIR-MCI-Trees, a tree-focused subset of LEVIR-MCI, for joint change detection and captioning, highlighting the potential of interactive, LLM-driven RSICI systems to improve accessibility, interpretability, and analytical efficiency in forest change analysis.

Credit default risk arises from complex interactions among borrowers, financial institutions, and transaction-level behaviors. While strong tabular models remain highly competitive in credit scoring, they may fail to explicitly capture cross-entity dependencies embedded in multi-table financial histories. In this work, we construct a massive-scale heterogeneous graph containing over 31 million nodes and more than 50 million edges, integrating borrower attributes with granular transaction-level entities such as installment payments, POS cash balances, and credit card histories. We evaluate heterogeneous graph neural networks (GNNs), including heterogeneous GraphSAGE and a relation-aware attentive heterogeneous GNN, against strong tabular baselines. We find that standalone GNNs provide limited lift over a competitive gradient-boosted tree baseline, while a hybrid ensemble that augments tabular features with GNN-derived customer embeddings achieves the best overall performance, improving both ROC-AUC and PR-AUC. We further observe that contrastive pretraining can improve optimization stability but yields limited downstream gains under generic graph augmentations. Finally, we conduct structured explainability and fairness analyses to characterize how relational signals affect subgroup behavior and screening-oriented outcomes.

This paper presents a semi-supervised learning framework for Gaussian mixture modelling under a Missing at Random (MAR) mechanism. The method explicitly parameterizes the missingness mechanism by modelling the probability of missingness as a function of classification uncertainty. To quantify classification uncertainty, we introduce margin confidence and incorporate the Aranda Ordaz (AO) link function to flexibly capture the asymmetric relationships between uncertainty and missing probability. Based on this formulation, we develop an efficient Expectation Conditional Maximization (ECM) algorithm that jointly estimates all parameters appearing in both the Gaussian mixture model (GMM) and the missingness mechanism, and subsequently imputes the missing labels by a Bayesian classifier derived from the fitted mixture model. This method effectively alleviates the bias induced by ignoring the missingness mechanism while enhancing the robustness of semi-supervised learning. The resulting uncertainty-aware framework delivers reliable classification performance in realistic MAR scenarios with substantial proportions of missing labels.

We study an online linear programming (OLP) model in which inventory is not provided upfront but instead arrives gradually through an exogenous stochastic replenishment process. This replenishment-based formulation captures operational settings, such as e-commerce fulfillment, perishable supply chains, and renewable-powered systems, where resources are accumulated gradually and initial inventories are small or zero. The introduction of dispersed, uncertain replenishment fundamentally alters the structure of classical OLPs, creating persistent stockout risk and eliminating advance knowledge of the total budget. We develop new algorithms and regret analyses for three major distributional regimes studied in the OLP literature: bounded distributions, finite-support distributions, and continuous-support distributions with a non-degeneracy condition. For bounded distributions, we design an algorithm that achieves $\widetilde{\mathcal{O}}(\sqrt{T})$ regret. For finite-support distributions with a non-degenerate induced LP, we obtain $\mathcal{O}(\log T)$ regret, and we establish an $Ω(\sqrt{T})$ lower bound for degenerate instances, demonstrating a sharp separation from the classical setting where $\mathcal{O}(1)$ regret is achievable. For continuous-support, non-degenerate distributions, we develop a two-stage accumulate-then-convert algorithm that achieves $\mathcal{O}(\log^2 T)$ regret, comparable to the $\mathcal{O}(\log T)$ regret in classical OLPs. Together, these results provide a near-complete characterization of the optimal regret achievable in OLP with replenishment. Finally, we empirically evaluate our algorithms and demonstrate their advantages over natural adaptations of classical OLP methods in the replenishment setting.

Recent advances in embodied intelligence have leveraged massive scaling of data and model parameters to master natural-language command following and multi-task control. In contrast, biological systems demonstrate an innate ability to acquire skills rapidly from sparse experience. Crucially, current robotic policies struggle to replicate the dynamic stability, reflexive responsiveness, and temporal memory inherent in biological motion. Here we present Neuromorphic Vision-Language-Action (NeuroVLA), a framework that mimics the structural organization of the bio-nervous system between the cortex, cerebellum, and spinal cord. We adopt a system-level bio-inspired design: a high-level model plans goals, an adaptive cerebellum module stabilizes motion using high-frequency sensors feedback, and a bio-inspired spinal layer executes lightning-fast actions generation. NeuroVLA represents the first deployment of a neuromorphic VLA on physical robotics, achieving state-of-the-art performance. We observe the emergence of biological motor characteristics without additional data or special guidance: it stops the shaking in robotic arms, saves significant energy(only 0.4w on Neuromorphic Processor), shows temporal memory ability and triggers safety reflexes in less than 20 milliseconds.

Speech Emotion Recognition models typically use single categorical labels, overlooking the inherent ambiguity of human emotions. Ambiguous Emotion Recognition addresses this by representing emotions as probability distributions, but progress is limited by unreliable ground-truth distributions inferred from sparse human annotations. This paper explores whether Large Audio-Language Models (ALMs) can mitigate the annotation bottleneck by generating high-quality synthetic annotations. We introduce a framework leveraging ALMs to create Synthetic Perceptual Proxies, augmenting human annotations to improve ground-truth distribution reliability. We validate these proxies through statistical analysis of their alignment with human distributions and evaluate their impact by fine-tuning ALMs with the augmented emotion distributions. Furthermore, to address class imbalance and enable unbiased evaluation, we propose DiME-Aug, a Distribution-aware Multimodal Emotion Augmentation strategy. Experiments on IEMOCAP and MSP-Podcast show that synthetic annotations enhance emotion distribution, especially in low-ambiguity regions where annotation agreement is high. However, benefits diminish for highly ambiguous emotions with greater human disagreement. This work provides the first evidence that ALMs could address annotation scarcity in ambiguous emotion recognition, but highlights the need for more advanced prompting or generation strategies to handle highly ambiguous cases.

Deploying learning-based controllers across heterogeneous robots is challenging due to platform differences, inconsistent interfaces, and inefficient middleware. To address these issues, we present UniCon, a lightweight framework that standardizes states, control flow, and instrumentation across platforms. It decomposes workflows into execution graphs with reusable components, separating system states from control logic to enable plug-and-play deployment across various robot morphologies. Unlike traditional middleware, it prioritizes efficiency through batched, vectorized data flow, minimizing communication overhead and improving inference latency. This modular, data-oriented approach enables seamless sim-to-real transfer with minimal re-engineering. We demonstrate that UniCon reduces code redundancy when transferring workflows and achieves higher inference efficiency compared to ROS-based systems. Deployed on over 12 robot models from 7 manufacturers, it has been successfully integrated into ongoing research projects, proving its effectiveness in real-world scenarios.

Search agents have emerged as a pivotal paradigm for solving open-ended, knowledge-intensive reasoning tasks. However, training these agents via Reinforcement Learning (RL) faces a critical dilemma: interacting with live commercial Web APIs is prohibitively expensive, while relying on static data snapshots often introduces noise due to data misalignment. This misalignment generates corrupted reward signals that destabilize training by penalizing correct reasoning or rewarding hallucination. To address this, we propose SearchGym, a simulation environment designed to bootstrap robust search agents. SearchGym employs a rigorous generative pipeline to construct a verifiable knowledge graph and an aligned document corpus, ensuring that every reasoning task is factually grounded and strictly solvable. Building on this controllable environment, we introduce SearchGym-RL, a curriculum learning methodology that progressively optimizes agent policies through purified feedback, evolving from basic interactions to complex, long-horizon planning. Extensive experiments across the Llama and Qwen families demonstrate strong Sim-to-Real generalization. Notably, our Qwen2.5-7B-Base model trained within SearchGym surpasses the web-enhanced ASearcher baseline across nine diverse benchmarks by an average relative margin of 10.6%. Our results validate that high-fidelity simulation serves as a scalable and highly cost-effective methodology for developing capable search agents.

This paper addresses the challenge of deadline-aware online scheduling for jobs in hybrid cloud environments, where jobs may run on either cost-effective but unreliable spot instances or more expensive on-demand instances, under hard deadlines. We first establish a fundamental limit for existing (predominantly-) deterministic policies, proving a worst-case competitive ratio of $Ω(K)$, where $K$ is the cost ratio between on-demand and spot instances. We then present a novel randomized scheduling algorithm, ROSS, that achieves a provably optimal competitive ratio of $\sqrt{K}$ under reasonable deadlines, significantly improving upon existing approaches. Extensive evaluations on real-world trace data from Azure and AWS demonstrate that ROSS effectively balances cost optimization and deadline guarantees, consistently outperforming the state-of-the-art by up to $30\%$ in cost savings, across diverse spot market conditions.

Privacy-preserving model co-training in medical research is often hindered by server-dependent architectures incompatible with protected hospital data systems and by the predominant focus on relative effect measures (hazard ratios) which lack clinical interpretability for absolute survival risk assessment. We propose FedRD, a communication-efficient framework for federated risk difference estimation in distributed survival data. Unlike typical federated learning frameworks (e.g., FedAvg) that require persistent server connections and extensive iterative communication, FedRD is server-independent with minimal communication: one round of summary statistics exchange for the stratified model and three rounds for the unstratified model. Crucially, FedRD provides valid confidence intervals and hypothesis testing--capabilities absent in FedAvg-based frameworks. We provide theoretical guarantees by establishing the asymptotic properties of FedRD and prove that FedRD (unstratified) is asymptotically equivalent to pooled individual-level analysis. Simulation studies and real-world clinical applications across different countries demonstrate that FedRD outperforms local and federated baselines in both estimation accuracy and prediction performance, providing an architecturally feasible solution for absolute risk assessment in privacy-restricted, multi-site clinical studies.

Asynchronous Many-Task (AMT) runtimes offer a productive alternative to the Message Passing Interface (MPI). However, the diverse AMT landscape makes fair comparisons challenging. Task Bench, proposed by Slaughter et al., addresses this challenge through a parameterized framework for evaluating parallel programming systems. This work integrates two recent cluster AMTs, Itoyori and ItoyoriFBC, into Task Bench for comprehensive evaluation against MPI and HPX. Itoyori employs a Partitioned Global Address Space (PGAS) model with RDMA-based work stealing, while ItoyoriFBC extends it with futurebased synchronization. We evaluate these systems in terms of both performance and programmer productivity. Performance is assessed across various configurations, including compute-bound kernels, weak scaling, and both imbalanced and communication-intensive patterns. Performance is quantified using application efficiency, i.e., the percentage of maximum performance achieved, and the Minimum Effective Task Granularity (METG), i.e., the smallest task duration before runtime overheads dominate. Programmer productivity is quantified using Lines of Code (LOC) and the Number of Library Constructs (NLC). Our results reveal distinct trade-offs. MPI achieves the highest efficiency for regular, communication-light workloads but requires verbose, lowlevel code. HPX maintains stable efficiency under load imbalance across varying node counts, yet ranks last in productivity metrics, demonstrating that AMTs do not inherently guarantee improved productivity over MPI. Itoyori achieves the highest efficiency in communication-intensive configurations while leading in programmer productivity. ItoyoriFBC exhibits slightly lower efficiency than Itoyori, though its future-based synchronization offers potential for expressing irregular workloads.

In clinical practice, medical segmentation datasets are often limited and heterogeneous, with variations in modalities, protocols, and anatomical targets across institutions. Existing deep learning models struggle to jointly learn from such diverse data, often sacrificing either generalization or domain-specific knowledge. To overcome these challenges, we propose a joint training method called Universal Harmonization (U-Harmony), which can be integrated into deep learning-based architectures with a domain-gated head, enabling a single segmentation model to learn from heterogeneous datasets simultaneously. By integrating U-Harmony, our approach sequentially normalizes and then denormalizes feature distributions to mitigate domain-specific variations while preserving original dataset-specific knowledge. More appealingly, our framework also supports universal modality adaptation, allowing the seamless learning of new imaging modalities and anatomical classes. Extensive experiments on cross-institutional brain lesion datasets demonstrate the effectiveness of our approach, establishing a new benchmark for robust and adaptable 3D medical image segmentation models in real-world clinical settings.

Large Language Models (LLMs) achieve competitive performance across diverse natural language processing (NLP) tasks, yet pretraining is computationally demanding, making optimizer efficiency an important practical consideration. Muon accelerates LLM pretraining via orthogonal momentum updates that serve as a matrix analogue of the element-wise sign operator. Motivated by the recent perspective that Adam is a variance-adaptive sign update algorithm, we propose two variants of Muon, Muon-NSR and Muon-VS, which apply variance-adaptive normalization to momentum before orthogonalization. Muon-NSR applies noise-to-signal ratio (NSR) modulation, while Muon-VS performs variance-based scaling without introducing additional hyperparameters. Experiments on GPT-2 and LLaMA pretraining demonstrate that our proposed methods accelerate convergence and consistently achieve lower validation loss than both competitive, well-tuned AdamW and Muon baselines. For example, on the LLaMA-1.2B model, Muon-NSR and Muon-VS reduce the iterations required to reach the target validation loss by $1.36\times$ relative to the well-tuned Muon following the recent benchmark.

A large number of heuristics have been proposed to optimize the reinforcement fine-tuning of LLMs. However, inconsistent claims are made from time to time, making this area elusive. Reflecting on this situation, two fundamental questions still lack a clear understanding: 1) what is the role of each optimizing choice? 2) which ones are the bottlenecks? This paper aims to shed light on them, and it faces the challenge of several entangled confounding factors in the fine-tuning process. To tackle this challenge, we propose a bottom-up experiment pipeline. The bottom layer is composed of a minimalist configuration: one training data, one rollout per round and the reward directly serve as the learning signal without advantage function design. This minimalist configuration connects to multi-armed bandit learning with extremely large discrete action space, which offers theories to corroborate the experiment findings. The up procedure of the experiment pipeline expanding the minimalist configuration layer by layer, examining the role of each design choice. Experimental results on three LLMs and two reasoning datasets not only reveal new understanding of the design choice but also yield essential insights to shape the area.

Large language models (LLMs) have recently been applied to binary decompilation, yet they still treat code as plain text and ignore the graphs that govern program control flow. This limitation often yields syntactically fragile and logically inconsistent output, especially for optimized binaries. This paper presents \textsc{HELIOS}, a framework that reframes LLM-based decompilation as a structured reasoning task. \textsc{HELIOS} summarizes a binary's control flow and function calls into a hierarchical text representation that spells out basic blocks, their successors, and high-level patterns such as loops and conditionals. This representation is supplied to a general-purpose LLM, along with raw decompiler output, optionally combined with a compiler-in-the-loop that returns error messages when the generated code fails to build. On HumanEval-Decompile for \texttt{x86\_64}, \textsc{HELIOS} raises average object file compilability from 45.0\% to 85.2\% for Gemini~2.0 and from 71.4\% to 89.6\% for GPT-4.1~Mini. With compiler feedback, compilability exceeds 94\% and functional correctness improves by up to 5.6 percentage points over text-only prompting. Across six architectures drawn from x86, ARM, and MIPS, \textsc{HELIOS} reduces the spread in functional correctness while keeping syntactic correctness consistently high, all without fine-tuning. These properties make \textsc{HELIOS} a practical building block for reverse engineering workflows in security settings where analysts need recompilable, semantically faithful code across diverse hardware targets.

We study the optimal design of additive mechanisms for vector-valued queries under $ε$-differential privacy (DP). Given only the sensitivity of a query and a norm-monotone cost function measuring utility loss, we ask which noise distribution minimizes expected cost among all additive $ε$-DP mechanisms. Using convex rearrangement theory, we show that this infinite-dimensional optimization problem admits a reduction to a one-dimensional compact and convex family of radially symmetric distributions whose extreme points are the staircase distributions. As a consequence, we prove that for any dimension, any norm, and any norm-monotone cost function, there exists an $ε$-DP staircase mechanism that is optimal among all additive mechanisms. This result resolves a conjecture of Geng, Kairouz, Oh, and Viswanath, and provides a geometric explanation for the emergence of staircase mechanisms as extremal solutions in differential privacy.

Infrastructure as Code (IaC) enables automated provisioning of large-scale cloud and on-premise environments, reducing the need for repetitive manual setup. However, this automation is a double-edged sword: a single misconfiguration in IaC scripts can propagate widely, leading to severe system downtime and security risks. Prior studies have shown that IaC scripts often contain security smells--bad coding patterns that may introduce vulnerabilities--and have proposed static analyzers based on symbolic rules to detect them. Yet, our preliminary analysis reveals that rule-based detection alone tends to over-approximate, producing excessive false positives and increasing the burden of manual inspection. In this paper, we present IntelliSA, an intelligent static analyzer for IaC security smell detection that integrates symbolic rules with neural inference. IntelliSA applies symbolic rules to over-approximate potential smells for broad coverage, then employs neural inference to filter false positives. While an LLM can effectively perform this filtering, reliance on LLM APIs introduces high cost and latency, raises data governance concerns, and limits reproducibility and offline deployment. To address the challenges, we adopt a knowledge distillation approach: an LLM teacher generates pseudo-labels to train a compact student model--over 500x smaller--that learns from the teacher's knowledge and efficiently classifies false positives. We evaluate IntelliSA against two static analyzers and three LLM baselines (Claude-4, Grok-4, and GPT-5) using a human-labeled dataset including 241 security smells across 11,814 lines of real-world IaC code. Experimental results show that IntelliSA achieves the highest F1 score (83%), outperforming baselines by 7-42%. Moreover, IntelliSA demonstrates the best cost-effectiveness, detecting 60% of security smells while inspecting less than 2% of the codebase.

The requirement for expert annotations limits the effectiveness of deep learning for medical image analysis. Although 3D self-supervised methods like volume contrast learning (VoCo) are powerful and partially address the labeling scarcity issue, their high computational cost and memory consumption are barriers. We propose 2D-VoCo, an efficient adaptation of the VoCo framework for slice-level self-supervised pre-training that learns spatial-semantic features from unlabeled 2D CT slices via contrastive learning. The pre-trained CNN backbone is then integrated into a CNN-LSTM architecture to classify multi-organ injuries. In the RSNA 2023 Abdominal Trauma dataset, 2D-VoCo pre-training significantly improves mAP, precision, recall, and RSNA score over training from scratch. Our framework provides a practical method to reduce the dependency on labeled data and enhance model performance in clinical CT analysis. We release the code for reproducibility. https://github.com/tkz05/2D-VoCo-CT-Classifier

Counterfactual explanations (CFEs) provide human-centric interpretability by identifying the minimal, actionable changes required to alter a machine learning model's prediction. Therefore, CFs can be used as (i) interventions for abnormality prevention and (ii) augmented data for training robust models. We conduct a comprehensive evaluation of CF generation using large language models (LLMs), including GPT-4 (zero-shot and few-shot) and two open-source models-BioMistral-7B and LLaMA-3.1-8B, in both pretrained and fine-tuned configurations. Using the multimodal AI-READI clinical dataset, we assess CFs across three dimensions: intervention quality, feature diversity, and augmentation effectiveness. Fine-tuned LLMs, particularly LLaMA-3.1-8B, produce CFs with high plausibility (up to 99%), strong validity (up to 0.99), and realistic, behaviorally modifiable feature adjustments. When used for data augmentation under controlled label-scarcity settings, LLM-generated CFs substantially restore classifier performance, yielding an average 20% F1 recovery across three scarcity scenarios. Compared with optimization-based baselines such as DiCE, CFNOW, and NICE, LLMs offer a flexible, model-agnostic approach that generates more clinically actionable and semantically coherent counterfactuals. Overall, this work demonstrates the promise of LLM-driven counterfactuals for both interpretable intervention design and data-efficient model training in sensor-based digital health. Impact: SenseCF fine-tunes an LLM to generate valid, representative counterfactual explanations and supplement minority class in an imbalanced dataset for improving model training and boosting model robustness and predictive performance

Most personal wellbeing apps present summative dashboards of health and physical activity metrics, yet many users struggle to translate this information into meaningful understanding. These apps commonly support engagement through goals, reminders, and structured targets, which can reinforce comparison, judgment, and performance anxiety. To explore a complementary approach that prioritizes self-reflection, we design KRIYA, an AI wellbeing companion that supports co-interpretive engagement with personal wellbeing data. KRIYA aims to collaborate with users to explore questions, explanations, and future scenarios through features such as Comfort Zone, Detective Mode, and What-If Planning. We conducted semi-structured interviews with 18 college students interacting with a KRIYA prototype using hypothetical data. Our findings show that through KRIYA interaction, users framed engaging with wellbeing data as interpretation rather than performance, experienced reflection as supportive or pressuring depending on emotional framing, and developed trust through transparency. We discuss design implications for AI companions that support curiosity, self-compassion, and reflective sensemaking of personal health data.

Accurate forecasting of daily attendance is vital for managing transportation, crowd flows, and services at large-scale international events such as Expo 2025 Osaka, Kansai, Japan. However, existing approaches often rely on multi-source external data (such as weather, traffic, and social media) to improve accuracy, which can lead to unreliable results when historical data are insufficient. To address these challenges, we propose a Transformer-based framework that leverages reservation dynamics, i.e., ticket bookings and subsequent updates within a time window, as a proxy for visitors' attendance intentions, under the assumption that such intentions are eventually reflected in reservation patterns. This design avoids the complexity of multi-source integration while still capturing external influences like weather and promotions implicitly embedded in reservation dynamics. We construct a dataset combining entrance records and reservation dynamics and evaluate the model under both single-channel (total attendance) and two-channel (separated by East and West gates) settings. Results show that separately modeling East and West gates consistently improves accuracy, particularly for short- and medium-term horizons. Ablation studies further confirm the importance of the encoder-decoder structure, inverse-style embedding, and adaptive fusion module. Overall, our findings indicate that reservation dynamics offer a practical and informative foundation for attendance forecasting in large-scale international events.

Social understanding abilities are crucial for multimodal large language models (MLLMs) to interpret human social interactions. We introduce Social Caption, a framework grounded in interaction theory to evaluate social understanding abilities of MLLMs along three dimensions: Social Inference (SI), the ability to make accurate inferences about interactions; Holistic Social Analysis (HSA), the ability to generate comprehensive descriptions of interactions; Directed Social Analysis (DSA), the ability to extract relevant social information from interactions. We analyze factors influencing model performance in social understanding, such as scale, architectural design, and spoken context. Experiments with MLLM judges contribute insights about scaling automated evaluation of multimodal social understanding.

Existing video inference (VI) enhancement methods typically aim to improve performance by scaling up model sizes and employing sophisticated network architectures. While these approaches demonstrated state-of-the-art performance, they often overlooked the trade-off of resource efficiency and inference effectiveness, leading to inefficient resource utilization and suboptimal inference performance. To address this problem, a fuzzy controller (FC-r) is developed based on key system parameters and inference-related metrics. Guided by the FC-r, a VI enhancement framework is proposed, where the spatiotemporal correlation of targets across adjacent video frames is leveraged. Given the real-time resource conditions of the target device, the framework can dynamically switch between models of varying scales during VI. Experimental results demonstrate that the proposed method effectively achieves a balance between resource utilization and inference performance.

Multi-agent systems face a fundamental architectural flaw: agent identity is bound to network location. When agents migrate between providers, scale across instances, or federate across organizations, URI-based identity schemes break references, fragment audit trails, and require centralized coordination. We propose the agent:// URI scheme, which decouples identity from topology through three orthogonal components: a trust root establishing organizational authority, a hierarchical capability path enabling semantic discovery, and a sortable unique identifier providing stable reference. The scheme enables capability-based discovery through DHT key derivation, where queries return agents by what they do rather than where they are. Trust-root scoping prevents cross-organization pollution while permitting federation when desired. Cryptographic attestation via PASETO tokens binds capability claims to agent identity, enabling verification without real-time contact with the issuing authority. We evaluate the scheme across four dimensions: capability expressiveness (100% coverage on 369 production tools with zero collision), discovery precision (F1=1.0 across 10,000 agents), identity stability (formal proofs of migration invariance), and performance (all operations under 5 microseconds). The agent:// URI scheme provides a formally-specified, practically-evaluated foundation for decentralized agent identity and capability-based discovery.

Large language models (LLMs) are increasingly deployed as intelligent tutoring systems, yet research on optimizing LLMs specifically for educational contexts remains limited. Recent works have proposed reinforcement learning approaches for training LLM tutors, but these methods focus solely on optimizing visible responses while neglecting the model's internal thinking process. We introduce PedagogicalRL-Thinking, a framework that extends pedagogical alignment to reasoning LLMs in education through two novel approaches: (1) Pedagogical Reasoning Prompting, which guides internal reasoning using domain-specific educational theory rather than generic instructions; and (2) Thinking Reward, which explicitly evaluates and reinforces the pedagogical quality of the model's reasoning traces. Our experiments reveal that domain-specific, theory-grounded prompting outperforms generic prompting, and that Thinking Reward is most effective when combined with pedagogical prompting. Furthermore, models trained only on mathematics tutoring dialogues show improved performance on educational benchmarks not seen during training, while preserving the base model's factual knowledge. Our quantitative and qualitative analyses reveal that pedagogical thinking reward produces systematic reasoning trace changes, with increased pedagogical reasoning and more structured instructional decision-making in the tutor's thinking process.

MITRE ATT&CK is a cybersecurity knowledge base that organizes threat actor and cyber-attack information into a set of tactics describing the reasons and goals threat actors have for carrying out attacks, with each tactic having a set of techniques that describe the potential methods used in these attacks. One major application of ATT&CK is the use of its tactic and technique hierarchy by security specialists as a framework for annotating cyber-threat intelligence reports, vulnerability descriptions, threat scenarios, inter alia, to facilitate downstream analyses. To date, the tagging process is still largely done manually. In this technical note, we provide a stratified "task space" characterization of the MITRE ATT&CK text tagging task for organizing previous efforts toward automation using AIML methods, while also clarifying pathways for constructing new methods. To illustrate one of the pathways, we use the task space strata to stage-wise construct our own multi-label hierarchical classification models for the text tagging task via experimentation over general cyber-threat intelligence text -- using shareable computational tools and publicly releasing the models to the security community (via https://github.com/jpmorganchase/MITRE_models). Our multi-label hierarchical approach yields accuracy scores of roughly 94% at the tactic level, as well as accuracy scores of roughly 82% at the technique level. The models also meet or surpass state-of-the-art performance while relying only on classical machine learning methods -- removing any dependence on LLMs, RAG, agents, or more complex hierarchical approaches. Moreover, we show that GPT-4o model performance at the tactic level is significantly lower (roughly 60% accuracy) than our own approach. We also extend our baseline model to a corpus of threat scenarios for financial applications produced by subject matter experts.

We report direct spectroscopic evidence of antimagnons, i.e., negative-energy spin waves identified by their signature inverted dispersion with Brillouin light scattering (BLS) spectroscopy. We investigate an ultrathin BiYIG film with a perpendicular magnetized anisotropy that compensates the demagnetizing field. By injecting a spin-orbit torque, the magnetization is driven into auto-oscillation and eventually into a non-equilibrium reversed state above a secondary current threshold ($\sim$1.2$\times$10$^7$~A/cm$^2$). The dispersion is measured by wavevector-resolved BLS and exhibits a sharp change from an upward dispersion to a downward one, in agreement with theoretical predictions and micromagnetic simulations. Around the threshold current, we observe the coexistence of conventional magnons and antimagnons. Our work establishes antimagnons with inverted dispersion and is a first step towards exploring novel phenomena and applications due to magnon-antimagnon coupling, such as magnon amplification and magnon-antimagnon entanglement, which are part of the emerging field of antimagnonics.

We study the effect of the choice of embedding geometry on the entropy of random geometric graph ensembles with soft connection functions. First we show that when the connection range is small, the entropy is dependent only on the dimension of the geometry and not the shape, but for large connection ranges the boundaries of the domain matter. Next, we formulate the problem of estimating entropy as a problem of estimating the average degree of a graph with the binary entropy function as its connection function. We use this formulation to study the effect of boundaries on the entropy, and to estimate the entropy of soft random geometric graphs in complicated geometries where a closed form pair distance density is not available.

Perovskite oxides are attractive for reactions in photo/electrocatalytic schemes, and extrinsic doping is a common strategy for tuning their properties. It is widely known that extrinsic dopants impact the structure and stability of perovskite surfaces, but an atomic-scale view is missing. Here, noncontact atomic force microscopy (ncAFM) and photoelectron spectroscopy (XPS/PES) are used to combine microscopic and spectroscopic evidence of cobalt adsorption, incorporation, and clustering at surfaces of two prototypical perovskites SrTiO3 and KTaO3. A number of different sub-ML coverages and temperatures of annealing were investigated. Several common features are observed: cobalt shows a strong preference for ionic nature (+2 and +3 charge states), and remains dispersed as single atoms to a certain extent in both perovskites. Two competing mechanisms are observed upon annealing: coalescence into clusters with a mixed metallic/ionic character, and incorporation into the surface and subsurface regions. The latter is more pronounced in SrTiO3, where a cobalt-stabilized surface reconstruction is identified, whereas for KTaO3 cobalt likely incorporates in the near-surface region.

Colloidal quantum dots (cQDs) have emerged as a cornerstone of next-generation optoelectronics, offering unparalleled spectral tunability and solution-processability. However, the transition from laboratory-scale devices to sustainable industrial manufacturing is fundamentally hindered by spin-coating workflows, which are intrinsically wasteful and restricted to planar geometries. These limitations are particularly acute for high-performance cQDs containing regulated elements such as lead, cadmium, or mercury, where poor material utilization exacerbates both environmental burden and cost. Here we report a biphasic dip-coating strategy that redefines the material efficiency of nanocrystal film fabrication. By utilizing an immiscible underlayer to displace ~88% of the active reservoir volume, we demonstrate a deposition geometry that decouples material consumption from total precursor volume. Infrared PbS photodetectors fabricated via this approach maintain their performance against spin-coated benchmarks while reducing ink consumption by up to 20-fold. Our technoeconomic analysis reveals that this biphasic architecture achieves cost parity at film thicknesses an order of magnitude lower than conventional monophasic dip-coating. Our results establish a low-waste framework for solution-processed materials, providing a viable pathway for the resource-efficient manufacturing of optoelectronic devices.

Semiflexible polymers in poor solvents exhibit a rich variety of collapsed morphologies, including globules, toroids, and rodlike bundles, arising from the competition between attractive interactions and chain stiffness. Computer simulations and experiments on stiff and conjugated polymers have revealed complex morphological crossovers, yet a unified theoretical description remains incomplete. Here we develop a coarse-grained, field-theoretic free-energy framework for linear polymers with variable stiffness that captures these morphologies and their transitions within a common description. The theory is built on three key ingredients: a density field describing monomer attraction and excluded-volume effects, a nematic order parameter accounting for orientational ordering in dense regions, and the bending rigidity of a worm-like chain. Using simple variational ansatzes for competing morphologies, we derive analytic expressions for their free energies and identify the boundaries separating coil, globule, toroidal, and rodlike conformational regimes as functions of the reduced attraction strength and the effective persistence length. The resulting phase-diagram topology provides a transparent free-energy-based framework for interpreting morphology diagrams observed in simulations and experiments on semiflexible polymers in poor solvents. We find the possibility of the existence of a triple point involving globules, rods and toroids.

The multicritical generalizations of the Lee-Yang universality class arise as renormalization-group fixed points of scalar field theories with complex $i\varphi^{2n+1}$ interaction, $n\in\mathbb{N}$, just below their upper critical dimension. It has been recently conjectured that their continuation to two dimensions corresponds to the non-unitary conformal minimal models $\mathcal{M}(2,2n+3)$. Motivated by that, we revisit the functional renormalization group approach to complex $\mathcal{P}\mathcal{T}$-symmetric scalar field theories in the Local Potential Approximation, without or with wavefunction renormalization (LPA and LPA' respectively), aiming to explore the fate of the $i\varphi^{2n+1}$ theories from their upper critical dimension to two dimensions. The $i\varphi^{2n+1}$ fixed points are identified using a perturbative expansion of the functional fixed-point equation near their upper critical dimensions, and they are followed to lower dimensions by numerical integration of the full equation. A peculiar feature of the complex $\mathcal{P}\mathcal{T}$-symmetric potentials is that the fixed points are characterized by real but negative anomalous dimensions $η$, and in low dimension $d$, this can lead to a change of sign of the scaling dimensions $Δ=(d-2+η)/2$, thus requiring a novel analysis of the analytical properties of the functional fixed-point equations. We are able to follow the Lee-Yang universality class ($n=1$) down to two dimensions, and numerically determine the scaling dimension of the fundamental field as a function of $d$. On the other hand, within the LPA', multicritical Lee-Yang fixed points with $n>1$ cannot be continued to $d=2$ due to the existence of unexpected non-perturbative fixed points that annihilate with the $i\varphi^{2n+1}$ fixed points.

A central open question in chiral-induced spin selectivity (CISS) is whether weak electron-phonon coupling in a helical molecular junction can generate a sizable spin polarization in two-terminal transport without invoking additional strong symmetry-breaking ingredients. We address this question by implementing a self-consistent nonequilibrium Green's function (NEGF) calculation for a helical tight-binding model with spin-orbit coupling and electron-phonon interactions. The electron-phonon self-energies are evaluated self-consistently, and the transport signal is extracted using the standard magnetization-reversal protocol with a spin-polarized analyzer lead. We benchmark a fully self-consistent NEGF within the self-consistent Born approximation (SCBA) treatment for both global and local electron-phonon couplings against commonly used approximations, including diagonal self-energy schemes. We quantify how the resulting transport regime and spin polarization depend on phonon frequency, coupling strength, bias, temperature, and system size. In contrast to large polarizations and anomalous size trends reported under approximate treatments, the fully self-consistent calculation yields negligible spin polarization, additionally the electron-phonon coupling mainly renormalizes the spectrum, and transport remains quasi-ballistic across the explored parameter range.

The theoretical relationship between the macroscopic textural quality and microscopic structural quality appearing in the phase inversion processes from fresh cream via whipped cream to butter is revealed by the multi-scale simulation of coupled map lattice (CML) based on the mesoscopic elementary processes of the emulsion interfaces. Using the Young-Laplace equation, we derive the microscopic particle quantities of the size and density of air bubbles and butter grains in an emulsion from the macroscopic rheological quantities of the overrun and viscosity of the emulsion. In doing so, we focus on the size determined by the "tug-of-war" between air bubbles and butter grains via their cohesion pressures, and on the density determined by the "costume change" of the emulsion molecular complexes (clad particles, e.g., butter grain-clad air bubbles) to their suitable size. Using the obtained microscopic particle quantities, we now propose a microscopic state diagram, the size-density plane, in addition to the previously proposed macroscopic state diagram, the viscosity-overrun plane. These state diagrams reveal that while the two well-known different phase inversion processes at high and low whipping temperatures appear as the two parallel processes of viscosity dominance and overrun dominance in the viscosity-overrun plane, they appear as the two orthogonal processes of isodensity/size dominance and isosize/density dominance in the size-density plane. This theoretical simulation result is significant for the quality design of butter because it demonstrates that differences in macroscopic textural quality can be easily controlled by differences in microscopic structural quality.

The formation and entrapment of gas bubbles during solidification significantly influence the microstructure and mechanical properties of materials, from metallic alloys to ice. While gas segregation at the solidification front is well-documented, the real-time dynamics of bubble nucleation, growth, and engulfment-and their dependence on solidification velocity-remain poorly understood. In this study, we use in situ cryo-confocal fluorescence microscopy to investigate the coupled gas-bubble dynamics at the solidification front of carbonated water, systematically varying the solidification velocity ($V = 1-20 μm/s$) while maintaining a constant thermal gradient ($G = 15 K/mm$). Our experiments reveal that bubble nucleation is governed by a characteristic nucleation time, which emerges from the interplay between gas diffusion ahead of the front, nucleation kinetics, and bubble growth, all competing with the advancing solidification front. These results allow us to estimate the critical gas concentration for bubbles nucleation in carbonated water. These results offer a detailed understanding of the mechanisms controlling bubble nucleation and entrapment during solidification at constant thermal gradient. They contribute to the development of strategies to control bubble formation in industrial processes where the presence of bubbles can either be detrimental or intentionally harnessed.

We investigate the dynamics of a two-qubit open quantum system, in particular the two-qubit spin-boson model in the strong coupling regime, coupled to two thermal bosonic baths under non-Markovian and non-equilibrium conditions. Two complementary approaches, the Hierarchical Equations of Motion (HEOM) and Reaction Coordinate Mapping (RCM), are employed to examine various coupling regimes between the qubits and their respective baths. The dynamical features of the model and the impact of the tunneling amplitude on quantum coherence of the system are probed using the $l_1$-norm of coherence. The model is further shown to have non-Markovian evolution. The nontrivial task of calculating entropy production in the strong-coupling regime is performed using auxiliary density operators in HEOM. Motivated by the realization of a quantum thermal device in the strong-coupling regime, the non-equilibrium steady-state behavior of the system is investigated. Furthermore, the relationship between the heat and spin currents and the tunneling amplitude is probed.

The Falicov-Kimball model (FKM) is long known to be the simplest model of correlated fermions exhibiting a novel Mott-like quantum critical point (QCP) assocaited with a {\it continuous} MIT in dimensions $D \geq 3$. It is also known to be isomorphic to an {\it annealed} binary-alloy disorder model. Notwithstanding extensive numerical studies for the FKM, analytic insight into the microscopic processes spawning novel Mott-like quantum criticality is scarce. Here, we develop a fully analytic theory for the Mott-like quantum criticality in the FKM on a hierarchical Cayley tree (Bethe lattice) by utilizing a single input from a 2-site cluster-dynamical mean-field theory (CDMFT). We find that density fluctuation modes acquire anomalous dimensions, originating from infra-red power-law singular cluster self-energies. Interestingly, we uncover, at $T=0$, that this {\it sub-diffusive} metal with glassy dynamics separating a weakly ergodic metal from a non-ergodic insulator shrinks to a single point, namely the Mott-like QCP, at least on the Bethe lattice. We detail the consequences of this anomalous quantum criticality for a range of thermal and dynamical responses in a variety of physical systems that can be effectively modelled by the FKM.

We consider a phase-separating mixture of active and passive fluids and explore morphological asymmetries of the emerging dominantly bicontinous dynamic emulsion. Two-dimensional numerical simulations reveal that the geometric and topological asymmetries can solely be explained by self-organized flows in the active region. As in inertial turbulence an inverse energy cascade in the active region leads to the formation of condensates. The size of these mesocales vortices is determined by the locally available space in the emulsion. As these condensates accumulate energy they impact the fluctuation of the surrounding interface and thus form a tight coupling between the flow field and the dynamic morphology. While explored for active/passive systems the symmetry-breaking mechanism can be generalized to heterogeneous active systems and proposes a way to control the morphology of various functional soft materials.

Iontronic neuromorphic computing has emerged as a rapidly expanding paradigm. The arrival of angstrom-confined iontronic devices enables ultra-low power consumption with dynamics and memory timescales that intrinsically align well with signals of natural origin, a challenging combination for conventional (solid-state) neuromorphic materials. However, comparisons to earlier conventional substrates and evaluations of concrete application domains remain a challenge for iontronics. Here we propose a pathway toward iontronic circuits that can address established time series benchmark tasks, enabling performance comparisons and highlighting possible application domains for efficient real-time time series processing. We model a Kirchhoff-governed circuit with iontronic memristors as edges, while the dynamic internal voltages serve as output vector for a linear readout function, during which energy consumption is also logged. All these aspects are integrated into the open-source pyontronics package. Without requiring input encoding or virtual timing mechanisms, our simulations demonstrate prediction performance comparable to various earlier solid-state reservoirs, notably with an exceptionally low energy consumption of over 5 orders of magnitude lower. These results suggest a pathway of iontronic technologies for ultra-low-power real-time neuromorphic computation.

We comment on the preprint arXiv:2511.04744 by Andolina et al.

The Mollow triplet is the definitive spectral signature of an optically dressed quantum emitter. We predict that for emitters coupled to localized phonons, this signature is not confined to the zero-phonon line. Under a strong resonant drive, we show that Mollow triplets are strikingly replicated on the associated phonon sidebands -a surprising result, given that phonon sidebands are typically viewed as incoherent, inelastic scattering pathways. These vibronic Mollow triplets are a direct fingerprint of dynamically generated dressed states that hybridize the emitter's electronic, photonic, and vibrational degrees of freedom. We develop a scalable analytical formalism to model this effect in complex, multi-mode molecular systems, such as dibenzoterrylene. Our work provides the precise driving conditions for observing these novel spectral features, establishing a new signature of coherence in vibronically coupled systems.

DNA nanostar (DNAns) hydrogels are promising materials for in vivo applications, including tissue regeneration and drug and antibody delivery. However, a systematic and quantitative understanding of the design principles controlling their degradation is lacking. Here, we investigate hydrogels made of three-armed DNAns with varying flexible joints, arm lengths, and mesh sizes and use restriction enzymes to cut the DNAns structures while monitoring the gel's degradation. We discover that (i) removing flexible joints, (ii) increasing arm length, or (iii) relocating the RE site along a DNA linker markedly accelerates hydrogel degradation. In contrast, non-specific endonucleases, e.g. DNaseI, quicly degrade DNAns hydrogels regardless of design. Importantly, the release of antibodies from DNAns hydrogels can be modulated by the action of different enzymes, confirming that programmable degradation can be leveraged for responsive drug-delivery systems. These findings provide a better understanding of the design principles for DNAns-based scaffolds with tunable degradation, cargo release, and responsive rheology.

Semimetals exhibit intriguing characteristics attributed to the coexistence of both electrons and holes. In rhombohedral multilayer graphene, a strong trigonal warping effect gives rise to a semi-metallic state near the Fermi surface, offering unique opportunities to explore the interplay of semi-metallic properties with strong correlations and topologies. Here, the observation of quarter semimetals in rhombohedral multilayer graphene by introducing spin-orbit coupling (SOC) is reported. The semi-metallic characteristics of rhombohedral graphene manifest as nearly vanished Hall resistance and parabolic longitudinal resistance. The strong correlations arising from the surface flat band lead to spontaneous symmetry breaking. SOC proximitized by WSe2 further lifts the valley degeneracy, resulting in the spontaneous time-reversal symmetry breaking, as evidenced by the hysteretic anomalous Hall effect. The coexistence of fully polarized electrons and holes allows for the observation of a non-monotonic temperature dependence of the anomalous Hall resistance. Furthermore, the application of moderate magnetic fields induces a phase transition from quarter semimetals to Chern insulators. These findings establish rhombohedral multilayer graphene as an ideal platform for studying strong correlations and topologies in semimetals.

We study spin-dependent transport in an extended Su-Schrieffer-Heeger chain with cosine modulated nearest- and next-nearest-neighbor hopping using the nonequilibrium Green's function formalism. Suitable tuning of the hopping parameters yields a complete separation of spin channels and perfect spin polarization over broad energy windows. The inclusion of next-nearest-neighbor hopping enhances both tunability and robustness, while systematic phase-diagram analyses reveal quantized polarization across extended regions of parameter space rather than at isolated fine-tuned points. These characteristics persist for larger system sizes, establishing the extended SSH model as a versatile platform for controllable spin-polarized transport.

Integration of high-quality photon emitters at specific locations within nanophotonic structures or optoelectronic devices is a key to innovating on-chip optical control and quantum technologies. Halide perovskite nanoparticles have great potential as single photon emitters with high quantum efficiency. To achieve their full potential, they must be embedded in a host material that ensures chemical stability and passivates surface defects. A previous experiment on a CsPbBr3-Cs4PbBr6 nanocomposite film suggested possibility that electron beam irradiation can be used to control positions of CsPbBr3 nano-light sources in the Cs4PbBr6 host although the effects of electron beam irradiation are not fully understood. Here, we fabricate a Cs4PbBr6-CsBr film, not containing the CsPbBr3 phase, and provide direct evidence that CsPbBr3 nanoparticles can be locally generated in the Cs4PbBr6 host by irradiation with a focused electron beam. We further demonstrate perovskite nano-light source arrays with submicron spacing using this method.

Using grain-resolved LES-DEM simulations, Zhang et al. (J. Geophys. Res. Earth Surf. 130, e2024JF007937, 2025) aimed to validate a grain-shape-corrected bedload transport equation proposed earlier by the same group. It states that grain shape effects are captured through a modified Shields number that depends, among others, on the drag coefficient, $C_{D_\mathrm{settle}}$, determined from the force balance for a grain settling in a fluid at rest. To independently vary $C_{D_\mathrm{settle}}$ in their simulations, the authors changed the boundary conditions on the grains' surfaces: By artificially shifting the locations of the no-slip conditions from the actual grain surface to a virtual surface a distance $l$ into the grain interior, they hoped to well approximate Navier-slip conditions with a slip length $l$. Here, we argue that this approximation is appropriate only if the thickness of the boundary layer that forms around the virtual surface is much larger than $l$, which we demonstrate was not the case for the authors' simulations. In particular, using independent DNS-DEM grain settling simulations for the same hydrodynamic conditions, we directly show that this approximation substantially overestimates the value of $C_{D_\mathrm{settle}}$ of a Navier-slip sphere. This implies that the conditions created with their artificial method do not correspond to physically realistic scenarios and therefore do not support the authors' grain shape correction. To support this conclusion, we demonstrate that their entire numerical data can be alternatively explained by a simple null hypothesis model, without grain shape correction, based on the virtual-grain rather than the actual-grain size.

We study planar Josephson junctions formed on the surface of a three-dimensional topological insulator (Fu-Kane proposal) and examine the experimentally relevant parameter regimes in which the effective velocity of the emergent one-dimensional Majorana modes approaches zero. We show that the frequently employed Fu-Kane effective theory breaks down in this case. As parameters like the chemical potential or the width of the junction are tuned, instances of vanishing effective velocity mark the emergence of additional 'Dirac cones' at zero energy and finite momentum. If the junction is subjected to an external magnetic field, Josephson vortices may then bind a number of zero modes in addition to the topological Majorana mode. The additional zero modes are 'symmetry-protected' and can be lifted by a broken mirror symmetry (which is to be expected in realistic scenarios) as well as by an in-plane magnetization (or Zeeman field). We note that the ensuing presence of additional low-energy Andreev states can significantly contribute to measured quantities like the Josephson current or microwave absorption spectra.

One-dimensional topological Anderson insulators provide a paradigm for disorder-induced topological phases in which the underlying system turns from a trivial to a topological phase. It is widely recognized that the latter vanishes at large disorder amplitude. Here, and contrary to the general belief, we provide evidence for a successive disorder-driven topological transitions in a single-wall nanotube, culminating in a topological Anderson phase that remains unexpectedly robust at strong disorder. This phenomenon is confirmed by analysis of the corresponding topological invariant, which increases stepwise as disorder increases, giving evidence for the emergence of edge states. We experimentally implement these topological Anderson staircase phase transitions in a one-dimensional topolectrical circuit, where the persistence of edge states is revealed by node-voltage measurements. The robustness of the edge states is corroborated by numerical calculations of their localization properties. Our work opens the road to topological disordertronics, where topological phases can be tuned by disorder.

We study anomalous localization in a one-dimensional non-Hermitian quasicrystal with a spatially disordered imaginary gauge field. The system is a generalized Aubry-André-Harper (AAH) chain with asymmetric nearest- and next-nearest-neighbor hoppings generated by a Bernoulli imaginary gauge field and a quasiperiodic onsite potential. In the standard non-Hermitian AAH limit, the system undergoes a transition from a fully erratic non-Hermitian skin effect (ENHSE) phase to a fully localized phase. We show that the fractal dimension cannot distinguish these phases, whereas the Lyapunov exponent and center-of-mass fluctuations provide sharp diagnostics. This transition is accompanied by a complex-to-real spectral change under periodic boundary conditions and a topological change of the spectral winding number. With next-nearest-neighbor hopping, we uncover an anomalous mobility edge separating Anderson-localized states from ENHSE states, rather than extended states. This mobility edge is captured by an energy-dependent winding number that vanishes in the localized regime. Finally, we propose a dynamical probe based on wave-packet expansion: for typical disorder realizations, the dynamics shows winding-controlled drift and disorder-selected pinning or boundary-wrapping recurrence, while disorder averaging restores Hermitian-like transport. These results offer practical spectral, topological, and dynamical diagnostics of anomalous localization and mobility edges in non-Hermitian quasicrystals.

Integrated optical microresonators exploiting either second-order ($χ^{(2)}$) or third-order ($χ^{(3)}$) nonlinearities have become key platforms for frequency conversion, low-noise microwave photonics, and quantum entanglement generation. Here, we present an analytic theory of Kerr-enhanced three-wave mixing amplification in an electro-optic microresonator with both $χ^{(2)}$ and $χ^{(3)}$ nonlinearities. We demonstrate that Kerr dressing hybridizes the optical sidebands, renormalizing the $χ^{(2)}$ couplings and detunings. As a result the system exhibits gain in regions where analogous bare $χ^{(2)}$ or $χ^{(3)}$ amplifiers are subthreshold. Time-domain Langevin simulations confirm this threshold reduction, mapping a practical design window for experiments.

In 2D field effect transistors the gate electrostatically dopes the 2D semiconductor (2DSC) channel, tuning the Fermi level. In principle, Kelvin probe force microscopy (KPFM) can detect the Fermi level, and its dependence on gate bias as well as position, potentially directly yielding band gaps, contact barriers, spatial nonuniformities, and sub-gap densities of states in such devices. However, KPFM relies on an oscillating probe voltage which itself electrostatically dopes the 2DSC, potentially creating a nonlinear response. Here, we show that when a suitably thin hBN back-gate dielectric is used, the KPFM signal agrees well with expectations, as explained by a quasistatic charge-balance model. Corresponding experimental results show excellent consistency with the literature values of the bandgaps of monolayer and trilayer WSe2. With this approach, the widely available technique of KPFM should find improved utility and new uses in the study of 2D devices.

Contact tracing (CT) plays a pivotal role in controlling early epidemic spread, particularly when a novel infectious disease emerges. However, the quantitative impact of missing information -- such as untraced cases or unnotified contacts -- on the effectiveness of CT remains insufficiently understood. Using a stochastic agent-based model with sociodemographics from metropolitan areas in South Korea, we simulate how different forms of information loss affect epidemic spreading dynamics. We construct information-loss scenarios based on two types: infector-omission (IO) and contact-omission (CO), including selective (SCO) and uniform (UCO) scenarios; IO corresponds to the omission of infected individuals (nodes) from the tracing process, leading to the loss of all movement trajectories and downstream transmission links originating from them, whereas CO corresponds to the omission of specific contact events (edges), in which infected individuals are identified but some of their transmission links fail to be detected or notified. The sensitivity of epidemic dynamics to increasing omission rates differs markedly between the two types: IO scenarios exhibit substantially stronger and more abrupt changes in transmission structure and epidemic outcomes, whereas CO scenarios produce more gradual effects. In both scenarios, the magnitude of these effects varies across cities, with a lower-population city (Busan) showing greater tolerance to information loss than the largest city (Seoul), underscoring the importance of regional tailoring in CT strategies. Both IO and CO scenarios also lead to an increase in the transmission network diameter as information loss grows, indicating that a small network diameter reflects effective contact tracing that limits the depth of transmission chains.

Excitation energy transfer (EET) in an organic/inorganic nanohybrid system, composed of a single \textit{para}-sexiphenyl (6P) molecule physisorbed on a finite-sized MoS$_2$ nanoflake, is investigated theoretically. % The electronic structure of the MoS$_2$ nanoflake is described by using an 11-band tight-binding model, in which edge states are passivated with H atoms to restore a well-defined bandgap. % Within a configuration-interaction scheme, excitonic states are constructed and, for computational efficiency, approximated by uncorrelated electron-hole pairs in the relevant high-energy window. % The EET rates are evaluated via Fermi's golden rule, incorporating Coulomb coupling, thermal broadening, and spectral overlap between the molecular excitation and the MoS$_2$ nanoflake's electron-hole pairs. % Our results reveal that energy transfer from the molecule to the nanoflake is the dominant process, and its efficiency depends strongly on the size of the MoS$_2$ nanoflake, as well as the molecule's vertical distance and lateral position relative to the nanoflake.

Semiconductor-s-type superconductor nanowires host spinful fermions and cannot be reduced to a single spinless Kitaev chain hosting single Majorana zero mode. Instead, such systems can be converted into two coupled p-wave Kitaev-like chains associated with different spin sectors. Using the bond Fermion transformation and exact diagonalization, we analyze parity resolved spectra and local spectral functions, demonstrating that zero-energy modes strongly localized at the system boundaries emerge only in one effective chain. Inter-chain coupling lifts parity degeneracy and redistributes the low-energy spectral weight, providing a controlled framework to assess the stability of Majorana-like modes in the finite spinful nanowires.

Electronic phases of matter, such as magnetism and superconductivity, are defined and distinguished by their order parameters that quantify the spontaneous symmetry breaking underlying each phase. The simplest cases are the uniform magnetisation of ferromagnets and isotropic gap function of conventional superconductors. Unconventional superconductors often have a nodal gap function, where the gap changes sign at nodes on the Fermi surface. This concept of unconventional or nodal order parameter symmetry has recently been extended to numerous magnetic systems, including altermagnets, in which up- and down-spin species are non-degenerate around the Fermi surface. Here we demonstrate that magnetic quantum oscillation measurements can provide a direct, bulk-sensitive, 3D mapping of the order parameter in an unconventional magnet. By rotating a magnetic field through high- and low-symmetry directions of the CrSb Brillouin zone, we show that this material's altermagetic band structure leads to the loss of mirror symmetry for each spin-split Fermi sheet away from highly symmetric nodal orientations. In momentum space, the difference between up and down spins follows the profile of the $\mathcal{Y}_{4}^{-3}=zy(3x^2-y^2)$ spherical harmonic - analogous to a $g$-orbital of the hydrogen atom. While notoriously difficult to resolve in unconventional superconductors, our work demonstrates that the order parameter symmetry of unconventional magnets can be conclusively determined through quantum-oscillatory quasiparticle spectroscopy. Our results empirically establish CrSb as a prototypical $g$-wave metallic altermagnet, which in pristine form possesses low residual resistivities down to $\sim$1 $μΩ$cm, opening numerous avenues for next-generation spintronic device applications

Recent research suggests the possibility of the two-dimensional breathing-Kagome magnet Nb$_3$Cl$_8$ hosting a quantum spin liquid state, warranting further study into its magnetic properties. Using ab initio calculations, we show that monolayer Nb$_3$Cl$_8$ has short-range antiferromagnetic correlations among Nb$_3$ trimers with S = 1/2, and becomes magnetically frustrated due to the underlying effective triangular lattice geometry, and is evidenced by a frustration index of f > 1. The high-temperature susceptibility shows a negative Weiss temperature from Monte Carlo calculations. Considering spin-orbit coupling, we investigate the magnetic anisotropy, including anisotropic exchange, single-ion anisotropy and the Dzyaloshinskii-Moriya interaction using the four-state energy mapping formalism. Although the elements have relatively small atomic numbers, the Dzyaloshinskii-Moriya interaction is comparable in magnitude to the anisotropic exchange. Additionally, we show that biaxial strain tunes the short-range correlations between antiferromagnetic, paramagnetic and ferromagnetic. These findings strengthen our understanding of Nb$_3$Cl$_8$ and advance its applications in current condensed matter physics and materials science research, including nanoscale mechanical and spintronics applications.

Mechanical metamaterials can be designed to exhibit unique mechanical properties, including tunable auxetic behavior as well as multi-stability, which arise from the geometry and configuration of the constituent building blocks. Lattice-based metamaterials, in particular, provide lightweight platforms where local instabilities can dictate the global response, with applications in energy routing, vibration isolation, and impact mitigation. In underdamped structures, perturbations have been found to propagate as nonlinear waves, e.g., transition waves or solitons. Here we investigate the opposite limit of overdamped, highly dissipative lattice metamaterials. Focusing on three-dimensional structures, we uncover how buckling instabilities, triggered by compression, propagate as fronts that shape the macroscopic behavior. We demonstrate in experiments on 3D-printed simple cubic lattices how global and local buckling modes can be controlled via the lattice geometry. By incorporating viscoelastic dissipation into a 3D-continuum model, we show that strain-driven buckling fronts obey coupled reaction-diffusion equations. The diffusion and reaction coefficients, determined by local geometry, material properties, and strain, select the propagation direction and enable steering of the fronts. This establishes a predictive and experimentally validated framework for the control of cascading mechanical instabilities in lattice-based metamaterials.

A central unresolved question in the theory of superconductivity is why only a small subset of metallic elements exhibit a superconducting state, whereas many others remain strictly normal. Neither the conventional Bardeen Cooper Schrieffer (BCS) framework nor its extensions involving charge density wave (CDW) or pair density wave (PDW) order provide a predictive or material-selective criterion capable of distinguishing superconducting metals from non-superconducting ones. In particular, the persistent absence of superconductivity in simple noble metals with well-defined Fermi surfaces poses a challenge for all traditional approaches. Here we address this problem using the Modulated Electron Lattice (MEL) Ginzburg Landau (GL) framework introduced in our previous work. In this formulation, a coarse-grained MEL charge field $ρ_{\mathrm{MEL}}(\mathbf{r})$ with momentum dependent stiffness $α(q)$ is coupled to the superconducting (SC) order parameter $ψ(\mathbf{r})$. We show that metallic superconductivity emerges only when the system satisfies a specific ``MEL enhancement window,'' characterized by a negative minimum of $α(q)$ at either a finite modulation wave vector $q^{\ast}$ or at $q=0$, together with sufficiently strong coupling between $ρ_{\mathrm{MEL}}$ and $ψ$. This unified criterion naturally partitions metallic elements into three universal classes: (i) MEL-enhanced superconductors with a finite-$q^{\ast}$ charge mode, (ii) conventional BCS superconductors as the homogeneous $q^{\ast}=0$ limit of the MEL framework, and (iii) metals for which $α(q)$ remains positive for all $q$, suppressing all MEL modes and preventing any superconducting instability. By applying this criterion to simple metallic elements, we identify why some metals develop superconductivity while others do not, possibly resolving a selection problem long open within the BCS paradigm.

Shear bands lie at the root of fracture initiation in bulk metallic glasses and amorphous polymers. For silicate glasses, in contrast, studies have largely emphasized permanent volumetric strain, commonly referred to as densification. Here we systematically investigate indentation-induced fracture in two distinct families of aluminoborosilicate glasses. The results demonstrate that plastic shear flow plays a decisive role in governing fracture initiation. In addition, molecular dynamics simulations reveal a pronounced composition dependence of softening associated with plastic shear flow, closely mirroring the experimentally observed propensity for strain localization. We conclude that silicate glasses conform to a universal pattern of rupture initiation governed by localization of shear-deformation, aligning with a broad range of amorphous materials, including bulk metallic glasses and glassy polymers.

In contrast to inert materials, living cells use molecular motors to generate forces independently of elastic strain. How do local active forces translate into large-scale shape, and what are the mechanical properties of the resulting "living matter"? Here, we address these questions within the active tension network (ATN) model for the mechanics of 2d epithelia. We represent the configuration of active forces geometrically by a tension triangulation dual to the cell tessellation. The Voronoi dual of the tension triangulation is shown to be macroscopically stress-free -- the tensions hence define an emergent reference state for the tissue. Two soft modes, curl-free and conformal deformations, map this reference to the internally stressed, but force-balanced, physical configuration of the tissue. Conformal deformations parametrize pressure gradients, leading to a generalization of von Neumann's law for the pressure in a foam. Via finite-element-like interpolation and a notion of "discrete" conformal maps, we both construct the mechanically balanced cell tessellations exactly on the microscopic level, and pass to the continuum limit. We systematically incorporate cell rearrangement into our theory by representing cell-scale topology in terms of circle packings. The results of this bottom-up coarse-graining study match a top-down continuum analysis presented in a companion paper. Thus, the geometry of force balance bridges between micro- and macro-scales, elucidating how cell-level active forces program shape. The present formalism may be useful in the study of foams, granular matter, and metamaterials, as well as in numerical simulations.

We compare two complementary theoretical approaches to compute and interpret Floquet sidebands in periodically driven quantum materials: a first-order perturbative approach (first-order perturbative Born approximation, PB1) and time-dependent nonequilibrium Green's functions (tdNEGF). Using graphene as a model Dirac system, we disentangle in pump-probe setups Floquet-dressed initial states, Volkov-dressed final states (also known as laser-assisted photoelectric effect, LAPE), and their interference. We quantify how photoemission matrix elements, polarization, incidence angle, and near-surface screening shape the momentum-resolved sideband intensity observed in tr-ARPES. PB1 yields an analytical expression for the momentum-dependent sideband intensity, and for graphene it captures the correct symmetry trends, such as the magnitude of the intensities when considering the interference between the Floquet and Volkov states and photoemission matrix elements. tdNEGF reproduces the full energy-momentum-resolved spectra, including hybridization gaps and spectral-weight redistribution. We find qualitative agreement between PB1 and tdNEGF once matrix elements are included; quantitative differences arise near hybridization regions and at specific angles where higher-order processes and self-energies are essential. Thus, for systems with simple band structures and away from these regions, the two approaches can be used in a complementary way.

We investigate the coupled dynamics of concentration and charge in asymmetric 1:1 electrolytes, focusing on the interplay between diffusion asymmetry and external electric fields. Using Brownian dynamics simulations and linearized stochastic density functional theory (SDFT), we analyze the transient response of charge and number currents to inhomogeneous electric fields, as well as the steady-state spatio-temporal fluctuations under uniform fields. Our results reveal that asymmetry in ionic diffusion coefficients introduces a non-trivial coupling between charge and number transport, which modifies the two relaxation modes already present in symmetric electrolytes -- a fast one associated with charge relaxation and a slow one linked to ambipolar diffusion. The dynamics are further modulated by the applied field, which enhances diffusion, alters screening lengths, and induces oscillatory behavior in the relaxation modes. The SDFT framework provides closed-form expressions for the intermediate scattering matrix, capturing the dynamics of density fluctuations and cross-correlations between number and charge. These predictions are validated by simulations, demonstrating excellent agreement across a wide range of wave vectors, both at equilibrium and under a finite electric field. Our findings highlight the critical role of diffusion asymmetry and external fields in tuning the transport properties of electrolytes, with implications for nanofluidic devices, energy harvesting, and iontronic circuits. This work bridges theoretical insights with practical applications, offering a robust framework for understanding and controlling electrolyte dynamics in asymmetric systems.

Deconfined quantum critical point (DQCP) describes direct, non-fine-tuned quantum phase transition between two ordered phases that break distinct and seemingly unrelated symmetries, providing a route to continuous phase transition beyond the conventional Ginzburg--Landau paradigm. In this work we extend the DQCP paradigm to systems with internal supersymmetry (SUSY), where the on-site Hilbert space furnishes a representation of a Lie superalgebra, and the Hamiltonian is invariant under the corresponding Lie supergroup. Focusing on the minimal supersymmetric generalization of spin $SU(2)$, namely $OSp(1|2)$, we propose a supersymmetric deconfined quantum critical point (sDQCP) between a phase that breaks internal $OSp(1|2)$ and a phase that instead breaks lattice rotation symmetry. We formulate a non-linear sigma model on the supersphere target space that captures the symmetry intertwinement characteristic of the sDQCP, and we further develop a gauge theory description to address its dynamical properties, including a heuristic argument for 3D XY critical behavior. Finally, we show that explicitly breaking $OSp(1|2)$ down to $SU(2)$ continuously connects our sDQCP to the conventional DQCP scenario.

Energy costs of information processing are growing exponentially. Bit erasure is a key problem in this energy-information nexus, and a number of seminal relationships have been deduced regarding the relationship between thermodynamic costs and memory storage. To continue making progress in the modern era, however, requires confronting thermodynamic costs in realistic physical systems which operate away from equilibrium. Here, we explore the thermodynamic costs of bit erasure in a complementary metal oxide semiconductor model of two types of random access memory. We find dynamic random access memory dissipates the least amount of energy when operated in the quasistatic limit, where errors are also minimized. By contrast, static random access memory is most efficiently operated in finite time due to the energy required to maintain the state of the bit. We demonstrate a numerically robust optimization scheme using mean field theory and automatic differentiation, finding optimal protocols compatible with electrical engineering insights. These results provide a framework for operating realistic circuits in thermodynamically advantageous ways.

Nonreciprocal supercurrents in Josephson junctions have recently emerged as a sensitive tool for investigating broken symmetries in superconducting quantum materials. Here, we report an even-odd Josephson diode effect (JDE) in Corbino-geometry junctions fabricated on the pristine surface of a bulk-insulating three-dimensional topological insulator (3DTI). We find that the diode polarity, which indicates the preferred direction of supercurrent flow, robustly alternates its sign depending on the parity (even or odd) of the enclosed vortex number. This behavior is absent in two key control devices: a non-topological graphene Corbino Josephson junction and a 3DTI-based linear Josephson junction. These results indicate that the polarity-tunable JDE is intrinsically linked to the unique combination of the proximitized topological superconductivity in the 3DTI surface and the Corbino device's closed-loop geometry. Our theoretical modeling attributes the observed sign change in diode polarity to the alternating sign of periodic boundary conditions in topological superconductors, supporting the interpretation that the vortex-parity-controlled JDE is a direct manifestation of the underlying Andreev bound state topology associated with the presence of non-Abelian anyons in the vortices.

We show that the Kasteleyn transition, the abrupt proliferation of infinite strings of defects in classical dimer and related models, can also be relevant for frustrated 2d quantum magnets. This is explicitly demonstrated in a phase of the spin-1/2 Heisenberg diamond-decorated honeycomb lattice where a family of exact eigenstates built as products of dimer and plaquette singlets can be mapped onto the dimer coverings of the honeycomb lattice. The low-temperature properties of this phase are accurately described by an effective dimer model with anisotropic activities and a small, tunable density of monomers, leading to an arbitrarily sharp crossover version of the Kasteleyn transition. The generalization to other geometries and the possibility to realize this model in organo-metallic compounds are briefly discussed.

We study the dynamics of circuits composed of stochastic and bistochastic controlled gates. This type of dynamics arises from quantum circuits with random controlled gates, as well as in stochastic circuits and deterministic classical cellular automata. We prove that stochastic and bistochastic controlled gates lead to two-point spatio-temporal correlation functions that vanish everywhere except when the two operators act on the same site. More generally, for multi-point correlations the two rightmost operators must act on the same site. We argue that autocorrelation, while hard to compute, typically decays exponentially towards a value that is exponentially small in the system size. Our results reveal a broad class of quantum systems that exhibit surprisingly simple correlation structures despite their complex microscopic dynamics.

Casimir torque is conventionally associated with explicit breaking of rotational symmetry, arising from material dielectric anisotropy, geometric asymmetry, or externally applied fields that themselves break rotational invariance. Here we demonstrate a fundamentally different mechanism: an axially symmetric magnetic field can generate a Casimir torque by inducing an axially asymmetric Casimir energy - and can even reverse the torque's sign. Focusing on two-dimensional altermagnets, we show that a magnetic field applied perpendicular to the plane - while preserving in-plane rotational symmetry - activates an orientation-dependent vacuum interaction through the combined crystalline symmetry $\mathrm{C_n T}$ inherent to altermagnetic order. The resulting torque emerges continuously and scales quadratically with the magnetic field strength. We further analyze its temperature and distance dependence, revealing scaling behaviors that are qualitatively different from those found in uniaxial bulk materials. Our results identify time-reversal symmetry breaking as a powerful route for engineering both the sign and strength of Casimir torque and establish altermagnets as an exciting platform for exploring phenomena driven by vacuum quantum fluctuations.

Orbital magnetism is a purely quantum phenomenon that reflects intrinsic electronic properties of solids, yet its microscopic description in interacting multiband systems remains incomplete. We develop a general quantum many-body framework for orbital magnetic responses based on the Luttinger-Ward functional. Starting from the Dyson equation, we reformulate the thermodynamic potential in a weak magnetic field and construct a controlled expansion in powers of $B$ applicable to correlated electron systems. A key technical advance is a modified ``Fourier'' representation using noncommutative coordinates, which allows the thermodynamic potential to be expressed in an effective momentum space where the magnetic field acts perturbatively. This formulation makes analytic progress possible within the Moyal algebra. As an application, we derive the spontaneous orbital magnetization and express it entirely in terms of the zero-field Hamiltonian renormalized by the self-energy. For frequency-dependent but Hermitian self-energies, we generalize the orbital magnetic moment and Berry curvature to momentum-frequency space and identify two gauge-invariant contributions built from these quantities. For frequency-independent self-energies the result reduces to the familiar geometric formula for noninteracting systems. This framework provides a unified foundation for computing orbital magnetic responses in correlated multiband materials.

Josephson junctions made of conventional superconductors display Fraunhofer-like oscillations of the critical current as a function of the threaded magnetic flux. When the superconductors are deposited on the surface of a three-dimensional topological insulator, this pattern is slightly modified due to the presence of chiral Majorana modes. Here we calculate the critical current of a Corbino Josephson junction, where the fluxoid becomes quantized and the superconducting phase has an integer winding. We discover that circular junctions exhibit similar behavior in both topologically trivial and non-trivial scenarios, while non-circular junctions demonstrate a remarkable distinction. Using a simple analytical model, we show that these non-circular junctions exhibit reentrant superconductivity with a period related to their number of corners, and numerically we find that this period is halved in the topological case. The period halving may help establish the existence of topological superconductivity in hybrid topological insulator-superconductor junctions.

Quantum many-body scars are non-thermal eigenstates embedded in the spectra of otherwise non-integrable Hamiltonians. Paradigmatic examples often appear as quasiparticle towers of states, such as the maximally ferromagnetic spin-1/2 states, also known as Dicke states. A distinguishing feature of quantum many-body scars is that they admit multiple local "parent" Hamiltonians for which they are exact eigenstates. In this work, we show that the locality of such parent Hamiltonians strongly constrains the relative placement of these states within the energy spectrum. In particular, we prove that if the full set of Dicke states are exact eigenstates of an extensive local Hamiltonian, then their energies must necessarily be equally spaced. Our proof builds on recent results concerning parent Hamiltonians of the $W$ state, together with general algebraic structures underlying such quasiparticle towers. We further demonstrate that this equal spacing property extends to local Hamiltonians defined on arbitrary bounded-degree graphs, including regular lattices in any spatial dimension and expander graphs. Hamiltonians with $k$-local interactions and a bounded number of interaction terms per site are also encompassed by our proof. On the same classes of graphs, we additionally establish equal spacing for towers constructed from multi-site quasiparticles on top of product states. For the towers considered here, an immediate corollary of the equal spacing property is that any state initialized entirely within the quantum many-body scar manifold exhibits completely frozen entanglement dynamics under any local Hamiltonian for which those scars are exact eigenstates. Overall, our results reveal a stringent interplay between locality and the structure of quantum many-body scars.

The emergence of altermagnetism, the collinear magnetic phase characterized by momentum-dependent spin-split bands but zero net magnetization, has fundamentally reshaped the classification of magnetic order. We propose an altermagnetic (AM) order in a repulsive Hubbard model on the Lieb-$5$ lattice. Considering only nearest-neighbor hoppings within the lattice, we show a phase transition from the nonmagnetic to a unique AM isolated band metal phase (AMIM), allowing clear identification of spin-split states. Additionally, the AM metallic phase (AMM) is also shown to appear as an intermediate phase during the transition from the normal metal to the AMIM in the presence of the diagonal hopping within each unit cell of the Lieb-$5$ lattice. The manifestation of distinct AM phases and the phase transitions, driven by Hubbard interaction and hopping integrals, have been explored in terms of spin-resolved band structure, spectral function, and the behavior of the AM order parameter. The stability of these AM phases against the spin-orbit coupling and temperature is also established.

We identify localizable entanglement (LE) as an order parameter for measurement-induced phase transitions (MIPT). LE exhibits universal finite-size scaling with critical exponents that match previous MIPT results and gives a nice operational interpretation connecting MIPTs to classical percolation. Remarkably, we find that LE decays exponentially with distance in the area-law phase as opposed to being essentially constant for the volume-law phase thereby, discover an intrinsic length scale $ξ_E$ that diverges at the critical measurement probability $p_c$. While classical percolation transition captures successful transport across a network, MIPT as characterized by LE can be interpreted as quantifying the amount of quantum teleportation between two given nodes in a quantum circuit. Building on this insight, we propose a two-ancilla protocol that provides an experimentally accessible readout of entanglement redistribution across the transition.

Stochastic kinetic models describe systems across biology, chemistry, and physics where discrete events and small populations render deterministic approximations inadequate. Parameter inference and inverse design in these systems require optimizing over trajectories generated by the Stochastic Simulation Algorithm, but the discrete reaction events involved are inherently non-differentiable. We present an approach based on straight-through Gumbel-Softmax estimation that maintains exact stochastic simulations in the forward pass while approximating gradients through a continuous relaxation applied only in the backward pass. We demonstrate robust performance on parameter inference in stochastic gene expression, accurately recovering kinetic rates of telegraph promoter models from both moment statistics and full steady-state distributions across diverse and challenging parameter regimes. We further demonstrate the method's applicability to inverse design problems in stochastic thermodynamics, characterizing Pareto-optimal trade-offs between non-equilibrium currents and entropy production. The ability to efficiently differentiate through exact stochastic simulations provides a foundation for systematic inference and rational design across the many domains governed by continuous-time Markov dynamics.

Estimating collision rates is of immense importance in particle-laden flows. An economical way of doing this is to directly identify incidences of caustics, or extreme clustering, by tracking particle velocity gradients in the neighborhoods of individual particles. The objective of this work is two-fold. (i) We find conditions under which caustics form, in point-vortex flow and in two-dimensional turbulence. While caustics are known to form in regions of strain, we show that the type of strain is key. Particles must remain in compressional strain throughout the process to form caustics, whereas survivor particles: which visit high strain but do not form caustics, briefly go through extensional strain during the early part of the process. This enables survivor particles to attain significantly straighter paths, and to move faster, whereas caustics particles follow paths of high curvature and move slower. As a result, caustics particles stay longer in high-strain regions than survivors. (ii) We ask about the effect of finite particle density, where the particle is denser than the background fluid. We show that finite-density particles need to sample stronger background strain than infinite-density ones to trigger caustics, but our other findings are universal across particle density.

Ferromagnetic exponential random graph models (ERGMs) are random graph models under which the presence of certain small structures (such as triangles) is encouraged; they can be constructed by tilting an Erdős--Rényi model by the exponential of a particular nonlinear Hamiltonian. These models are mixtures of metastable wells which each behave macroscopically like an Erdős--Rényi model, exhibiting the same laws of large numbers for subgraph counts [CD13]. However, on the microscopic scale these metastable wells are very different from Erdős--Rényi models, with the total variation distance between the two measures tending to 1 [MX23]. In this article we clarify this situation by providing a sharp (up to constants) bound on the Hamming-Wasserstein distance between the two models, which is the average number of edges at which they differ, under the coupling which minimizes this average. In particular, we show that this distance is $Θ(n^{3/2})$, quantifying exactly how these models differ. An upper bound of this form has appeared in the past [RR19], but this was restricted to the subcritical (high-temperature) regime of parameters. We extend this bound, using a new proof technique, to the supercritical (low-temperature) regime, and prove a matching lower bound which has only previously appeared in the subcritical regime of special cases of ERGMs satisfying a "triangle-free" condition [DF25]. To prove the lower bound in the presence of triangles, we introduce an approximation of the discrete derivative of the Hamiltonian, which controls the dynamical properties of the ERGM, in terms of local counts of triangles and wedges (two-stars) near an edge. This approximation is the main technical and conceptual contribution of the article, and we expect it will be useful in a variety of other contexts as well. Along the way, we also prove a bound on the marginal edge probability under the ERGM via a new bootstrapping argument. Such a bound has already appeared [FLSW25], but again only in the subcritical regime and using a different proof strategy.

We present a comprehensive study of quantum dot (QD) coupling to various phononic modes in a phononic waveguide, combining multiband kp and configuration-interaction (CI) QD state simulations with finite-element waveguide mode modeling. We consider self-assembled Stranski-Krastanov InGaAs/GaAs as well as local droplet-etched GaAs/AlGaAs structures. Using kp-CI calculations, we quantify the strain and piezoelectric responses of InAs and GaAs QDs. By systematically isolating volumetric/shear deformation-potential and piezoelectric channels, we demonstrate how mode symmetries dictate distinct coupling mechanisms. We identify the dominant coupling channels and characterize their observable signatures in the QD response. We predict strong linear energy shifts under volumetric strain and quadratic behavior under shear strain, especially in GaAs QDs. The piezoelectric effect is dominated by polarizability, which also leads to a quadratic response. The simulations show energy modulations up to 0.7 meV for an acoustic wave with 0.1 nm amplitude. The quadratic response to shear strain and piezoelectric field leads to frequency doubling in the QD response to a mechanical wave and to non-harmonic time traces when linear and quadratic effects contribute to a similar degree. The deep understanding of QD-acoustic couplings opens pathways to the optimal design of QD and waveguide structures, as well as to improved engineering of acousto-optic quantum interfaces.

Understanding the entanglement dynamics in quantum many-body systems under steady-state transport conditions is an actively pursued challenging topic. Hydrodynamic equations, akin to transport equations for charge or heat, would be of great interest but face severe challenges because of the inherent nonlocality of entanglement and the difficulty of identifying conservation laws. We show that progress is facilitated by using information as key quantity related to - but distinct from - entanglement. Employing the recently developed "information lattice" framework, we characterize spatially and scale-resolved information currents in nonequilibrium open quantum systems. Specifically, using Lindblad master equations, we consider noninteracting fermion chains coupled to dissipative reservoirs. By relating the information lattice to a noise lattice constructed from particle-number fluctuations, we show that information is experimentally accessible via noise easurements. Similarly, local information currents can be obtained by measuring particle currents, onsite occupations, and covariances of particle numbers and/or particle currents. Using the fermionic negativity to quantify bipartite entanglement, we also study transport-induced entanglement and its relation to information currents. For a clean particle-hole symmetric chain, we find that information currents are shielded from entering the information lattice. Impurities or particle-hole asymmetry break this effect, causing information current flow and entanglement between end segments of the chain. Our work opens the door to systematic investigations of information transport and entanglement generation in driven open quantum systems far from equilibrium.

Quantum dot - graphene hybrid materials have raised significant interest due to the unique synergy of the optical properties of colloidal quantum dots (QDs) and the transport properties of graphene. This stimulated the development of low-cost and up-scalable solution-processed strategies for hybrid materials with potential application in light harvesting and opto-electronic devices. Here we report a versatile covalent-linking based approach for the functionalization of reduced graphene oxide (rGO), to prepare a variety of QD-rGO hybrid dispersions with QDs of different size and composition (PbS, PbS/CdS and CdSe QDs), and shape (CdSe/CdS dot-in-rods). We achieved a well-controlled QD coverage of the rGO sheets by functionalizing the rGO surface with mercapto-silane linkers. A further spectroscopic investigation of near-infrared PbS QD-rGO materials demonstrates efficient electronic coupling between both materials. The QD photoluminescence emission quenching and exciton lifetime shortening up to 95%, together with subtle graphene Raman G-band shifts upon QD linking, supports electron transfer as the dominant relaxation pathway from the QD to the rGO. The use of core/shell PbS/CdS QDs allows tuning of the transfer efficiency from 94% for a 0.2 nm thin CdS shell, down to 30% for a 1.1 nm thick shell.

We show that the large $n$ limit of the $O(n)$ quantum rotor model defined on a general graph has the same critical behavior as the corresponding quantum spherical model and that the critical exponents depend solely on the spectral dimension $d_s$ of the graph. To this end, we employ a classical to quantum mapping and use known results for the large $n$ limit of the classical $O(n)$ model on graphs. Away from the critical point, we discuss the interplay between the Laplacian and the Adjacency matrix in the whole parameter plane of the quantum Hamiltonian. These results allow us to paint the full picture of the $O(n)$ quantum rotor model on graphs in the large $n$ limit.

The vibrational landscape of adsorbed molecules is central to understanding surface interactions at the atomic scale, influencing phenomena from catalysis to molecular electronics. Recent advances in atomic-scale tip-enhanced Raman spectroscopy (TERS) have enabled vibrational mapping of single molecules with sub-nanometer spatial resolution, providing unprecedented insights into molecule-surface interactions by confining light in plasmonic picocavities. Here, we exploit TERS in a cryogenic scanning tunneling microscope junction to perform Raman hyperspectral mapping of single iron phthalocyanine (FePc) molecules in three non-equivalent adsorption configurations on Ag surfaces. We explore the changes in the vibrational modes of FePc molecules adsorbed on two distinct silver crystal terminations with differing symmetry, Ag(111) and Ag(110), revealing how subtle variations in the adsorption geometry due to substrate anisotropy can strongly influence molecular vibrations, lifting the degeneracy of individual normal modes. Our findings not only demonstrate the first use of sub-nanometer TERS mapping across different symmetry configurations but also provide a deeper understanding of how site-specific vibrational properties are intimately linked to local atomic environments. This capability paves the way for precisely tailoring surface interactions and controlling chemical reactions at the atomic scale.

We propose an electrically controlled protocol to detect weak Lorentz-violating (LV) backgrounds through the second-order shift photocurrent in noncentrosymmetric crystals. Using a spinful Rice--Mele model, we show that a stationary LV background induces a momentum-odd correction to the Bloch Hamiltonian, which generates an odd-in-field contribution to the shift current. This leads to a directional asymmetry, whereby the photocurrent distinguishes opposite orientations of an applied static field. The effect originates from an LV-induced deformation of the interband phase and can be isolated experimentally by comparing field-reversed configurations, with vanishing response at transverse orientations, providing an internal consistency check. Our results demonstrate that nonlinear optical responses offer a practical and symmetry-selective route for probing LV effects in solid-state systems.

We unveil a quantum Pontus-Mpemba effect enabled by the Liouvillian skin effect in a dissipative tight-binding chain with asymmetric incoherent hopping and coherent boundary coupling. The skin effect, induced by non-reciprocal dissipation, localizes relaxation modes near the system boundaries and gives rise to non-orthogonal spectral geometry. While such non-normality is often linked to slow relaxation, we show that it can instead accelerate relaxation through a two-step protocol - realizing a quantum Pontus-Mpemba effect. Specifically, we consider a one-dimensional open chain with coherent hopping $J$, asymmetric incoherent hoppings $J_{\rm R} \neq J_{\rm L}$, and a controllable end-to-end coupling $ε$. For $ε=0$, the system exhibits the Liouvillian skin effect, with left and right eigenmodes localized at opposite edges. We compare two relaxation protocols toward the same stationary state: (i) a direct relaxation with $ε=0$, and (ii) a two-step (Pontus) protocol where a brief coherent evolution transfers the excitation across the lattice before relaxation. Although both share the same asymptotic decay rate, the two-step protocol relaxes significantly faster due to its reduced overlap with the slow boundary-localized Liouvillian mode. The effect disappears when $J_{\rm R}=J_{\rm L}$, i.e., when the skin effect vanishes. Our results reveal a clear connection between boundary-induced non-normality and protocol-dependent relaxation acceleration, suggesting new routes for controlling dissipation and transient dynamics in open quantum systems.

When two molecular species with mutual affinity are mixed together, various self-assembled phases can arise at low temperature, depending on the shape of like and unlike interactions. Among them, stripes -- where layers of one type are regularly alternated with layers of another type -- hold a prominent place in materials science, occurring e.g. in the structure of superconductive doped antiferromagnets. Stripe patterns are relevant for the design of functional materials, with applications in optoelectronics, sensing, and biomedicine. In a purely classical setting, an open question pertains to the features that spherically-symmetric particle interactions must have to foster stripe order. Here we address this challenge for a lattice-gas mixture of two particle species, whose equilibrium properties are exactly determined by Monte Carlo simulations with Wang-Landau sampling, in both planar and spherical geometry, and for equal chemical potentials of the species. Somewhat surprisingly, stripes can emerge from largely different off-core interactions, featuring various combinations of repulsive like interactions with a predominantly attractive unlike interaction. In addition to stripes, our survey also unveils crystals and crystal-like structures, cluster crystals, and networks, which considerably broaden the catalog of possible patterns. Overall, our study demonstrates that stripes are more widespread than generally thought, as they can be generated by several distinct mechanisms, thereby explaining why stripe patterns are observed in systems as diverse as cuprate materials, biomaterials, and nanoparticle films.

Clusters formed by fluctuations of two-dimensional (2D) directed interfaces around a threshold level have been extensively studied at equilibrium and in nonequilibrium steady states, but their coarsening dynamics remain poorly understood. Here, we numerically investigate this unexplored coarsening of clusters in 2D growing interfaces believed to belong to the Kardar-Parisi-Zhang universality class. Using a two-point spatial correlator, we demonstrate statistical time invariance of the evolving configurations and identify scaling forms shared across distinct models. We reveal a pronounced asymmetry in the growth of the largest clusters: one cluster emerges as a giant structure whose characteristic length exceeds the correlation length. Population-dependent scaling forms for the number densities of cluster areas are uncovered. These findings highlight new universal aspects of growing interfaces and suggest avenues for experimental verification.

The quantum anomalous Hall (QAH) insulator is uniquely characterized by the topological Chern number C. Controlling the Chern number is a key step toward functional topological electronics and enables access to exotic quantum phases beyond the traditional quantum Hall physics. Here, we report a series of QAH insulators in twisted rhombohedral graphene family, in which the Chern number can be tuned through layer configuration, in-situ electrostatic doping, and displacement field. Specifically, in twisted monolayer-rhombohedral N-layer graphene, denoted as (1+N) L, we observe QAH states with C=N at moire filling v=1, where N=3,4,5 represents the layer number of rhombohedral graphene. These results are experimentally confirmed by quantized Hall resistance and the Streda formula. In twisted monolayer-trilayer graphene, we also observe states with |C|=3 at v=3, whose sign can be switched by either electrostatic doping or displacement field. Furthermore, in twisted Bernal bilayer-rhombohedral tetralayer graphene denoted as (2+4) L, we demonstrate a displacement-field-driven topological phase transition between two distinct QAH states with C=3 and C=4 at v=1. Our work establishes twisted rhombohedral graphene as a highly versatile, layer-engineered platform for designing and dynamically controlling high-Chern-number topological matters.

The processes of simultaneous coagulation and Ostwald ripening of particles in the concluding stage of phase transformation are considered. We solve the integro-differential system of Smoluchowski-type kinetic and mass balance equations using a computationally efficient numerical algorithm based on low-rank matrices. We compare our numerical solutions for different initial particle-volume distributions with the universal distribution function for combined coagulation and Ostwald ripening. Our calculations confirm the tendency of a particulate ensemble to the universal particle-volume distribution to be approached asymptotically after a sufficiently long time, no matter what the initial particle-volume distribution might be.

Contact electrification give rise to charge accumulation at the interface when two materials are brought into contact with each other. The charge accumulation at the interface will diffuse to the interior of the conducting material if the dimensions of the contacting conducting material is of the order of an unknown critical length scale. This contact electrification induced interlayer charge transfer will modify the fundamental physical properties of both the contacting materials. This review first discusses the reported experimental evidence of flexoelectricity induced contact electrification and interlayer charge transfer in conducting thin film based heterostructures. The interlayer charge transfer creates a gradient of charge carrier in both the thin films constituting the heterostructure and also modifies the electron-electron interactions. Further, the interlayer charge transfer changes the electron-phonon coupling, spin-phonon coupling and magnetoelectronic coupling that give rise to new physical behavior, which did not exist prior to the interlayer charge transfer. The new physical behaviors from interlayer charge transfer and their mechanistic origins are reanalyzed and discussed, which include spin-Hall effect of charge carriers, topological Hall effect of magnetoelectronic electromagnon, inhomogeneous magnetoelectronic multiferroic effect, flexoelectronic proximity effect and topological spin texture. This review article presents a unified picture of current status and future directions that will provide the scientists a stepping stone for research in the field of flexoelectricity mediated contact electrification and interlayer charge transfer mediated behavior in the micro/nanoscale heterostructures of the conducting materials.

Component systems - ensembles of realizations built from a shared repertoire of modular parts - are ubiquitous in biological, ecological, technological, and socio-cultural domains. From genomes to texts, cities, and software, these systems exhibit statistical regularities that often meet the "bona fide" requirements of laws in the physical sciences. Here, we argue that the generality and simplicity of those laws are often due to basic combinatorial or sampling constraints, raising the question of whether such patterns are actually revealing system-specific mechanisms and how we might move beyond them. To this end, we first present a unifying mathematical framework, which allows us to compare modular systems in different fields and highlights the common "null" trends as well as the system-specific uniqueness, which, arguably, are signatures of the underlying generative dynamics. Next, we can exploit the framework with statistical mechanics and modern machine-learning tools for a twofold objective. (i) Explaining why the general regularities emerge, highlighting the constraints between them and the general principles at their origins, and (ii) "subtracting" them from data, which will isolate the informative features for inferring hidden system-specific generative processes, mechanistic and causal aspects.

The quantum geometric tensor (QGT), whose real and imaginary parts define the quantum metric and Berry curvature, encodes the intrinsic geometry of quantum states. While electronic QGT has been directly observed and linked to various phenomena like electron-phonon coupling, its bosonic analogue remains both theoretically and experimentally unexplored. We demonstrate that the dynamical structure factor directly encodes the full QGT throughout the Brillouin zone, establishing it as a sensitive probe of both quantum metric and Berry curvature. Applying this framework, we uncover clear geometric signatures in a twofold quadruple Weyl phonon in BaPtGe and the node-line magnon in Gd. Our results establish a general, direct route to measuring quantum geometry in bosonic systems, a crucial step toward elucidating its impact on condensed matter phenomena.

Reaching consensus on massive discussion networks is critical for reducing noise and achieving optimal collective outcomes. However, the natural tendency of humans to preserve their initial ideas constrains the emergence of global solutions. To address this, Collective Intelligence (CI) platforms facilitate the discovery of globally superior solutions. We introduce a dynamical system based on the standard $O(N)$ model to drive the aggregation of semantically similar ideas. The system consists of users represented as nodes in a $d=2$ lattice with nearest-neighbor interactions, where their ideas are represented by semantic vectors computed with a pretrained embedding model. We analyze the system's equilibrium states as a function of the coupling parameter $β$. Our results show that $β> 0$ drives the system toward a ferromagnetic-like phase (global consensus), while $β< 0$ induces an antiferromagnetic-like state (maximum dissent), where users maximize semantic distance from their neighbors. This framework offers a controllable method for managing the tradeoff between cohesion and diversity in CI platforms.

Ultrasound-induced degradation of soft polymeric colloids, like microgels, as well as a controlled drug release enabled by mechanoresponsive bonds, has recently attracted considerable attention. However, most examples in the literature focus primarily on the applications rather than examining the underlying mechanisms of the structural changes occurring in microgels due to cavitation - changes that are crucial for developing effective drug delivery systems. In this work, we provide a comprehensive view on how microgel structure governs the susceptibility to rupture and mass loss upon cavitation, investigating both conventional microgels containing mechanoresponsive disulfide bonds and more complex asymmetrically crosslinked core-shell microgels. By combining dynamic and static light scattering, small-angle X-ray scattering, and atomic force microscopy, we demonstrate that an interplay between mechanoresponsive crosslinks and the swelling degree determines the microgels susceptibility to ultrasound-induced damage. Our findings indicate that local stress from cavitation bubbles varies strongly within the microgel dispersion. The majority of microgels undergo gradual erosion at their periphery, resulting in smaller yet structurally intact particles over time, observable by light scattering and AFM. In contrast, microgels closer to a cavitation bubble can experience partial rupture or completely disintegrate, producing smaller, more polydisperse fragments, which contributes substantially to the overall mass loss observed. In the core-shell microgels with different crosslinkers in the core and shell, degradation occurs nearly uniformly across both regions, instead of selectively targeting the weaker part. These observations highlight the complexity of the degradation dynamics as well as the similarity to processes seen in linear polymers and bulk hydrogels.

This work theoretically demonstrates that a spin qubit in a parabolic quantum wire driven by a bichromatic field exhibits a confinement-tunable synthetic gauge field, leading to novel Floquet topological phenomena. The study presents the underlying mechanism for topological protection of qubit states against time-periodic perturbations. The analysis reveals a confinement-induced topological Landau-Zener transition, marked by a shift from preserved symmetries to chiral interference patterns in Landau-Zener-St$\ddot{u}$ckelberg-Majorana interferometry. Notably, the emergence of non-Abelian geometric phases under cyclic evolution in curved confinement and phase-parameter space is identified, enabling holonomic quantum computation. Additionally, the prediction of unconventional Floquet-Bloch oscillations in the quasi-energy and resonance transition probability spectra as a function of the biharmonic phase indicates exotic properties, including fractal spectra and fractional Floquet tunneling. These phenomena provide direct evidence of coherent transport in the synthetic dimension. Collectively, these findings position quantum wire materials has a versatile platform for Floquet engineering, topological quantum control, and fault-tolerant quantum information processing.

The formation of topological defects during continuous phase transitions exhibits nonequilibrium universality. While the Kibble-Zurek mechanism (KZM) predicts universal scaling of point-like defect numbers under slow driving, the statistical properties of extended defects remain largely unexplored across both slow and fast protocols. We investigate vortex string formation in a three-dimensional holographic superfluid. For slow quenches, the vortex string number follows KZM scaling, while for rapid quenches, it exhibits complementary universal scaling governed by the final temperature. Beyond the vortex string number, the loop-length distribution reveals a richer structure: individual loops follow the first-return statistics of three-dimensional random walks, $P(\ell) \sim \ell^{-5/2}$. While the total vortex length distribution remains Gaussian, its cumulants obey universal scaling laws with varying power-law exponents, and thus differ markedly from those observed in point-defect systems, indicating distinct statistical features of extended topological defects.

For quantum spin models defined on a two-dimensional lattice, we look for the best numbering of the lattice sites (a layout) that, at fixed bond dimension and other parameters of the density matrix renormalization group (DMRG) algorithm, gives the lowest value of the variational energy, maximum entropy and truncation error. We consider the conjecture that the optimal layout is a Hamiltonian path, and that it optimizes a simply computable geometric cost function. Finding the minimum of such a function, which is a variant of the minimum linear arrangement problem, provides the DMRG with an efficient layout of the lattice and improves both accuracy and convergence time. We present applications to the antiferromagnetic and spin glass spin-1/2 models on the square and triangular lattices.

We study the topological properties of a one-dimensional quasiperiodic-potential-modulated mosaic trimer lattice. To begin with, we first investigate the topological properties of the model in the clean limit free of quasiperiodic disorder based on analytical derivation and numerical calculations of the Zak phase $Z$ and the polarization $P$. Two nontrivial topological phases corresponding to the $1/3$ filling and $2/3$ filling, respectively, are revealed. Then we incorporate the mosaic modulation and investigate the influence of quasiperiodic disorder on the two existing topological phases. Interestingly, it turns out that quasiperiodic disorder gives rise to multiple distinct effects for different fillings. At $2/3$ filling, the topological phase is significantly enhanced by the quasiperiodic disorder and topological Anderson insulator emerges. Based on the calculations of polarization and energy gap, we explicitly present corresponding topological phase diagram in the $λ-J$ plane. While for the $1/3$ filling case, % the topological phase is dramatically suppressed by the same quasiperiodic disorder. the quasiperiodic disorder dramatically compresses the topological phase, and strikingly, further induces the emergence of reentrant topological phase transitions instead. Furthermore, we verify the topological phase diagrams by computing the many-body ground state fidelity susceptibility for both the $1/3$ filling and $2/3$ filling cases. Our work exemplifies the diverse roles of quasiperiodic disorder in the modulation of topological properties, and will further inspire more research on the competitive and cooperative interplay between topological properties and quasiperiodic disorder.

Monolayer transition metal dichalcogenide (TMD) field-effect transistors (FETs), with their atomically thin bodies, are promising candidates for future gate-all-around (GAA) nanoribbon architectures. While state-of-the-art Si GAA nanoribbon transistors feature channel widths in the tens of nanometers, most reported TMD-based FETs remain limited to micrometer-scale dimensions, limiting their relevance for ultra-scaled electronics. In this work, we investigate the channel width scaling in nanoribbon transistors based on monolayer MoS2 grown on 2-inch wafers, achieving widths of approximately 30-40 nm. Remarkably, nanoribbon width scaling enhances the on-current by 30-40%, reaching up to 700 uA/um for the smallest-width devices, while also improving the subthreshold slope (SS) to as low as 70 mV/dec. This enhancement is attributed to a stronger electric field at the nanoribbon edges without significant degradation from edge-related scattering. To further demonstrate the scalability of the nanoribbon device, we evaluate the variability of extremely scaled monolayer MoS2 nanoribbon transistor arrays featuring a contact pitch of 60 nm and an effective oxide thickness (EOT) of approximately 0.9 nm. Beyond MoS2, we extend the nanoribbon structure to WS2 n-type and WSe2 p-type FETs, demonstrating a viable path toward complementary monolayer TMD nanoribbon FETs for future ultra-scaled electronics.

Strontium titanate (SrTiO$_3$), famously described by Nobel laureate K. A. Müller as the "drosophila of solid-state physics", has been extensively investigated over the last seventy five years for its intricate coupling of structural, electronic, and dielectric properties and continues to serve as a foundational platform for advancing oxide electronics. In its pristine form, SrTiO$_3$ exhibits quantum paraelectric behavior below 35 K and undergoes an antiferrodistortive phase transition near 105 K. This transition generates ferroelastic twin domains separated by a dense network of domain walls, which function as nanoscale structural defects with far-reaching consequences. While the static influence of ferroelastic domain walls on carrier transport in electron-doped SrTiO$_3$ is well established, recent experimental results show that the emergence of polarity at these walls, combined with strain fields and inherent quantum fluctuations, induces correlated dynamical phenomena such as glass-like relaxations of electrons and memory effects. In this review, we highlight these recent advances, focusing on the subtle interplay between the emergence of nanoscale polar order, quantum fluctuations, and long-range strain fields. We propose that understanding charge carrier dynamics in the background of these complex ferroelastic domain wall landscapes offers a new paradigm for exploring electronic transport in the presence of local polar order and quantum fluctuations, with broad implications for correlated oxides.

We demonstrate deterministic control of branched flow of light using anisotropic nematic liquid crystals. By sculpting the director field via photoalignment, we create spatially programmable optical potentials that govern light scattering and propagation. This platform enables configurable, anisotropic branched flow of light and reveals a universal scaling law for its characteristic features, directly connecting disordered photonics with mesoscopic wave transport. Under extreme anisotropy, we observe a pronounced directional channeling effect, driven by anomalous symmetry-breaking velocity diffusion, which concentrates light propagation along preferential directions while suppressing transverse spreading. These findings establish a tunable material platform for harnessing branched flow of light, opening pathways toward on-chip photonic circuits that exploit disorder-guided transport, scattering-resilient endoscopic imaging, and adaptive optical interfaces in complex media.

Twisted bilayer systems host a wealth of emergent phenomena, such as flat-band superconductivity, ferromagnetism, and ferroelectricity, arising from moiré superlattices and unconventional interlayer coupling. Despite their central role, direct and quantitative access to the out-of-plane atomic structure in these systems has remained elusive due to their nanoscale dimensions. Here, we introduce an automated dark-field electron tomography technique that enables three-dimensional structural analysis of atomically thin materials with sub-angstrom precision. Applying this method to twisted bilayer WSe$_2$, we uncover a significant expansion of the interlayer spacing compared to the bulk configuration, exceeding 0.1 angstrom, along with a remarkable temperature-driven interlayer decoupling unique to the twisted bilayer. Ultrafast measurement further reveals optically induced interlayer separation of ~0.2 angstrom on the picosecond timescale, attributed to transient exciton formation. These findings not only establish a powerful approach for visualizing hidden out-of-plane structures in atomically thin micro-flake materials, but also uncover the intrinsic fragility and dynamical tunability of interlayer coupling in moiré-engineered 2-dimensional materials.

Topological states of matter in disordered systems without translation symmetry have attracted great interest in recent years. These states with topological characters are not only robust against certain disorders, but also can be counterintuitively induced by disorders from a topologically trivial phase in the clean limit. In this review, we summarize the current theoretical and experimental progress on disorder-induced topological phases in both condensed-matter and artificial systems. We first introduce the topological Anderson insulators (TAIs) induced by random disorders and their topological characterizations and experimental realizations. We then discuss various extensions of TAIs with unique localization phenomena in quasiperiodic and non-Hermitian systems. We also review the theoretical and experimental studies on the disorder-induced topology in dynamical and many-body systems, including topological Anderson-Thouless pumps, disordered correlated topological insulators and average-symmetry protected topological orders acting as interacting TAI phases. Finally, we conclude the review by highlighting potential directions for future explorations.

In recent years, the planar Hall effect (PHE) has become a key probe of Berry curvature and the anomalous Hall effect (AHE). Threefold-symmetric signals under in-plane fields are often attributed to such quantum mechanisms. Here, we establish a purely classical origin for a three-fold-symmetric PHE. The idea is simple yet decisive: a third-order expansion of the Boltzmann equation in the magnetic field reveals that the threefold component originates from the relative positions of the mirror planes in the crystals with respect to the measurement setups. Remarkably, the threefold contribution should be ubiquitous because this symmetry condition can be realized across a broad range of crystals. Numerical estimates based on concrete models further show that its amplitude is comparable to that expected from the AHE.

We studied transport of indirect excitons (IXs) and IX mediated spin transport in a MoSe$_2$/WSe$_2$ van der Waals heterostructure in magnetic fields up to 8 T. We observed the long-range IX transport and the long-range IX mediated spin transport in the magnetic fields. The IX transport and spin transport are characterized by the 1/e decay distances reaching $\sim$ 100 micrometers. The decay distance of the spin transport correlates with the decay distance of IX transport. These decay distances first increase and then decrease with increasing IX density for all studied magnetic fields. The long-range IX transport and the long-range spin transport in the magnetic fields are consistent with the similar long-range transport in zero magnetic field.

Although the classification of crystalline materials can be generally handled by momentum-space-based approaches, topological classification of aperiodic materials remains an outstanding challenge, as the absence of translational symmetry renders such conventional approaches inapplicable. Here, we present a numerically efficient real-space framework for classifying parity-based $\mathbb{Z}_2$ topology in aperiodic systems based on the spectral localizer framework and the direct computation of the sign of a Pfaffian associated with a large sparse skew-symmetric matrix. Unlike projector-based or momentum-space-based approaches, our method does not rely on translational symmetry, spectral gaps in the Hamiltonian's bulk, or gapped auxiliary operators such as spin projections, and instead provides a local, energy-resolved topological invariant accompanied by an intrinsic measure of topological protection. A central contribution of this work is the development of a scalable sparse factorization algorithm that enables the reliable determination of the Pfaffian's sign for large sparse matrices, making the approach practical to realistic physical materials. We apply this framework to identify the quantum spin Hall effect in quasicrystalline class AII systems, including gapless heterostructures, and to diagnose fragile topology in a large $C_2 \mathcal{T}$-symmetric photonic quasicrystal. Overall, our results demonstrate that the spectral localizer, combined with efficient sparse numerical methods, provides a unified and robust tool for diagnosing parity-based topological phases in aperiodic electronic, photonic, and acoustic materials where conventional band-theoretic indexes are inapplicable.

Stochastic renewal processes are ubiquitous across physics, biology, and the social sciences. Here, we show that continuous-time renewal dynamics can naturally produce a mixed discrete-continuous structure, with a macroscopic fraction of particles occupying a discrete state. For ensembles of continuous-time random walkers subject to Poissonian renewal resets, we develop an age-structured framework showing this discrete component corresponds to localization at the reset configuration. We next show that collective interactions can retain memory although all reset events are memoryless. Remarkably, the transition to collective memory is discontinuous, and we identify a first-order dynamical phase transition between weak collective bias, where the dynamics are stationary, to strong collective bias where the dynamics are nonstationary and display aging up to finite-size effects. We explicitly discuss ecological implications of our work, illustrating how continuous-time renewal dynamics shape macroscopic structure and collective organization with long-term memory.

Multiphase fluid flows comprising of mesoscale solid particles, liquid droplets, or gas bubbles are common in both natural and man-made systems, but quantifying the energy transfer is challenging due to complex bubble-particle interactions. In this study, we used ultrafast synchrotron X-ray imaging to study the mesoscale dynamic interactions among ultrasonic cavitation bubbles and hydrophobic particles or clusters. Critical dynamic information and data were extracted from the vast amount of X-ray images and then fed into the hybrid analytical-numerical model for calculating the energy transfer from the oscillating bubble and the imploding bubble to the nearby hydrophobic particles. Using the Ni spherical microparticles as an example, at bubble oscillation approximately 16% (80-320 nJ) of the local energy was transferred to the particle. At bubble implosion, the transferred energy increased approximately 26% (0.135-1.09 uJ). Local energy transfer occurred on timescales of 1 us to 1 ms and length scales of 1 um to 1 mm. Within each ultrasound cycle, kinetic and potential energy underwent complex exchanges, with local energy exhibiting a stepwise decay at the end of each cycle. The transferred energy was mainly consumed for enabling highly efficient particle dispersion. This research provides quantitative insights into optimizing hydrophobic nanomaterial dispersion and has broader implications for interfacial energy transfer processes such as making suspensions, composite materials and exfoliated 2D materials.

We present a unified theoretical framework for effective transport in periodic and tilted periodic potentials based on additive functionals of stochastic processes. By systematically combining the Poisson equation, corrector construction, and martingale decomposition, we show that both the long-time drift and diffusion of overdamped Brownian motion can be derived within a single and transparent scheme. In the absence of external tilt, the formalism naturally recovers the classical Lifson-Jackson formula for the effective diffusion coefficient. When a constant bias is applied, breaking detailed balance and inducing a finite stationary current, the same approach yields the Stratonovich expressions for the effective drift and diffusion in tilted periodic potentials. Beyond one dimension, we demonstrate that the same additive-functional structure extends directly to two-dimensional and general N dimensional periodic diffusions, leading to the standard homogenized drift and diffusion tensor expressed in terms of vector-valued correctors. Our derivation highlights the central role of additive functionals in separating bounded microscopic corrections from unbounded macroscopic transport and clarifies the connection between reversible and nonequilibrium steady states. This work provides a conceptually unified and mathematically controlled route to transport coefficients in periodic media, with direct relevance to stochastic transport, soft matter, and nonequilibrium statistical physics.

The coupling of spin and electronic degrees of freedom via the spin-orbit interaction (SOI) is an essential ingredient for many proposed future technologies. However, probing the strength and nature of SOI is a significant challenge, especially in heterostructures. Here, we consider the nonlinear Shubnikov-de Haas (NSdH) effect, a quantum oscillatory effect that occurs under conditions similar to those of the well-known SdH effect, but is second order in the applied electric field. We demonstrate that, unlike its linear counterpart, the NSdH effect is highly sensitive to the spin textures that arise from SOI. In particular, we show that the phase and beating of NSdH oscillations in nonlinear conductivities can clearly distinguish between different types of SOI. As a demonstration, we show how NSdH can distinguish between the linear and cubic Rashba couplings that are expected in germanium heterostructures. Our results establish the NSdH effect as a powerful and sensitive probe of SOI, offering a new framework for characterizing materials relevant to topology, spintronics, and solid-state quantum information technologies.

Phononic nanodevices offer a promising route toward quantum technologies, as phonons combine strong confinement within matter with broad coupling capabilities to various quantum systems. In particular, the piezoelectric response of materials such as lithium niobate enables coupling between superconducting qubits and gigahertz-frequency phonons. However, bulk lithium niobate phononic devices typically rely on surface acoustic waves and are therefore inherently subject to leakage from the surface into the bulk substrate. Here, we explore the acoustic behavior of resonator cavities supporting GHz-frequency Lamb waves in a 200 nm-thick suspended lithium niobate layer. We characterize the acoustic response at both room and millikelvin temperatures. We find that our resonator cavities with strong confinement reach intrinsic quality factors of approximately 6000 at the single phonon level. We use the measured parameters of the resonators to model their coupling to a superconducting transmon qubit, allowing us to evaluate their potential as quantum acoustic devices.

We investigate the stability of the many-body localized phase against quantum avalanche instabilities in a one-dimensional Heisenberg spin chain with long-range power-law interactions ($V\propto r^{-α}$). By combining exact diagonalization of static properties with Lindblad master equation simulations of open-system dynamics, we systematically map the interplay between interaction range and disorder strength. Our finite-size scaling analysis of entanglement entropy identifies a critical interaction exponent $α_c \approx 2$, which separates a fragile regime, characterized by an exponentially diverging critical disorder, from a robust short-range regime. To rigorously test the system's resistance to avalanches, we couple the boundary to an infinite-temperature bath and track the propagation of the thermalization front into the localized bulk. We find that the characteristic thermalization time follows a unified scaling law, $T_{r_{\text{th}}} \sim \exp[κ(α) LW]$ (herein, $L$ is the system size, and $W$ is the disorder intensity), which diverges exponentially with the product of system size and disorder strength. This suppression enables the derivation of a quantitative stability criterion, $W_{\text{stab}}(α)$, representing the minimum critical disorder strength required to maintain avalanche stability. Our results confirm that the MBL phase remains asymptotically stable in the thermodynamic limit when disorder exceeds an interaction-dependent threshold, bridging theoretical debates on long-range MBL and providing a roadmap for observing these dynamics in experimental platforms such as Rydberg atom arrays.

This study investigates the elastic response of a two-dimensional semi-infinite medium subjected to a moving surface load with a prescribed displacement profile. As a fundamental step, we derive analytical Green's functions for the displacement and stress fields generated by a point load traveling at a constant velocity along the surface, explicitly incorporating elastodynamic effects through Mach number dependence. These moving-load solutions serve as building blocks for constructing more general loading scenarios via linear superposition. Based on Green's functions, an inverse problem is formulated to reconstruct the unknown surface traction responsible for a given surface displacement. The inverse analysis is performed through a Fourier-domain inversion with regularization, which enables a direct and computationally efficient determination of the contact pressure without iterative forward simulations. This framework is applied to a rigid wheel-ground contact problem, where the imposed displacement is dictated by the wheel geometry. The reconstructed surface traction exhibits a smooth, symmetric distribution within the contact region, while the resulting subsurface stress fields are obtained in closed analytical form and involve dilogarithm functions. The principal stress difference reveals characteristic spatial patterns similar to photoelastic fringes, and their asymmetry increases with the Mach number, reflecting the dynamic nature of the moving contact.

We study the thermalization properties of a fully nonlinear lattice model originally derived from the two-dimensional cubic defocusing nonlinear Schrödinger equation (NLS) using analytical and numerical methods. Our analysis reveals both ergodic and nonergodic regimes; importantly, we find broad parameter ranges where the dynamics is ergodic even though it lies outside the Gibbsian parameter regime (for both $D=0.25$ and $D=2$), and a higher-energy range where ergodicity breaks down. We observe that in a certain range of parameters, the system requires non-standard statistical descriptions, indicating a breakdown of conventional thermalization. We examine the influence of the nonlinear dispersion parameter $D$ on the system's behavior, showing that increasing $D$ enhances fluctuations and speeds up the crossover of $q(T)$ toward the $\sim 1/T$ scaling. By analyzing excursion times, probability density functions, and localization patterns, we characterize transitions between ergodic and nonergodic behavior. In long-time numerical simulations within the non-ergodic regime for $D>1$, stable localization over two sites is observed, while $D<1$ favors single-site localization in the high energy density regimes. Our results provide insights into the interplay between thermalization, localization, and non-standard statistical behavior in genuinely nonlinear systems.

We investigated the electrical conductivity of platinum nanowires with heights ranging from 2 nm to 200 nm, deposited by focused electron beam induced deposition (FEBID). Post-deposition processing was employed to enhance the electrical conductivity of the platinum nanowires. Thermal annealing of as-deposited nanowires in air at 225$^{\circ}$C for 4 hours increased electrical conductance by up to five orders of magnitude. After annealing, 22.5 $\mathrm{μm}$-long nanowires with a height of 36 nm exhibited resistances of approximately 10 k$Ω$. This nanowire underwent a reduction in height to one-quarter of its original value, a reduction in width to one half, and a reduction in cross-sectional area by approximately one order of magnitude. The platinum-to-carbon weight ratio increased from 35:65 to 85:15. The electrical resistance decreased monotonically as temperature was lowered from room temperature to 100 mK, confirming that annealed FEBID platinum nanowires are promising building blocks for nanoelectronic devices operating at millikelvin temperatures.

Stochastically gated interfaces play an important role in a variety of cellular diffusion processes. Examples include intracellular transport via stochastically gated ion channels and pores in the plasma membrane of a cell, intercellular transport between cells coupled by stochastically gated gap junctions, and oxygen transport in insect respiration. Most studies of stochastically-gated interfaces are based on macroscopic models that track the particle concentration averaged with respect to different realisations of the gate dynamics. In this paper we use renewal theory to develop a probabilistic model of single-particle Brownian motion (BM) through a stochastically gated interface. We proceed by constructing a renewal equation for 1D BM with an interface at the origin, which effectively sews together a sequence of BMs on the half-line with a totally absorbing boundary at $x=0$. Each time the particle is absorbed, the stochastic process is immediately restarted according to the following rule: if the gate is closed then BM restarts on the same side of the interface, whereas if the gate is open then BM restarts on either side of the interface with equal probability. In order to ensure that diffusion restarts in a state that avoids immediate re-absorption. we assume that whenever the particle reaches the interface it is instantaneously shifted a distance $ε$ from the origin. We explicitly solve the renewal equation for $ε>0$ and show how the solution of a corresponding forward Kolmogorov equation is recovered in the limit $ε\rightarrow 0$. However, the renewal equation provides a more general mathematical framework by explicitly separating the first passage time problem of detecting the gated interface (absorption) and the subsequent rule for restarting BM. We conclude by extending the theory to higher-dimensional interfaces.

We study a family of lattice-gas systems defined on semiregular grids, obtained by projecting the vertices of three different geodesic icosahedra onto a spherical surface. By using couplings up to third neighbors we explore various interaction patterns, ranging from core-corona repulsion to square-well attraction and short-range attractive, long-range repulsive potentials. The relatively small number of sites in each grid ($\sim 100$) enables us to compute the exact statistical properties of the systems as a function of temperature and chemical potential by Wang-Landau sampling. For each case considered we highlight the existence of distinct low-temperature ``phases'', featuring, among others, regular-polyhedral, cluster-crystal, and worm-like structures. We highlight similarities and differences between these motifs and those observed on the triangular lattice under the same conditions. Finally, we discuss the relevance of our results for the bottom-up realization of spherical templates with desired functionalities.

The LaAlO3/SrTiO3 (LAO/STO) interface hosts a rich range of electronic phenomena, including unconventional electron pairing that in quantum dots gives rise to a negative effective charging energy U. Here, we show freestanding LAO/STO micro-membranes naturally hosting negative-U centers, where lateral confinement arises intrinsically, rather than from engineered nanostructures. These centers coexist with gate-tunable superconductivity and can remain stable upon thermal cycling from millikelvin temperatures to room temperature. Transport is in excellent agreement with calculations based on a negative-U Anderson model, and electrostatic simulations indicate characteristic center sizes of 20-80 nm. Our findings suggest that negative-U centers may arise from the intrinsic interfacial inhomogeneity typical of LAO/STO, and should therefore be considered a general feature of the LAO/STO interface. This could have important consequences for the microwave response of interfacial superconducting devices.

The surface half-quantum Hall effect, a hallmark consequence of axion electrodynamics, can be induced by gapping out the surface states of topological insulators through surface magnetization, and has led to a variety of topological response phenomena observed in experiments. In this work, we investigate phonon dynamics originating from an acoustic analog - the surface phonon Hall viscosity - that can also occur at the surface of magnetic topological insulators. This surface phonon Hall viscosity stems from the Nieh-Yan action in the strain response of topological insulators, where strain acts as the effective vierbein field for the bulk low-energy massive Dirac fermions. Crucially, this viscosity term entangles phonon dynamics with surface magnetization. In magnetic topological insulator films, we find that this interaction causes acoustic phonons to become chiral when the magnetization at the top and bottom surfaces is parallel, and nonreciprocal when it is anti-parallel. We further discuss potential experimental signatures of phonon dynamics induced by surface phonon Hall viscosity, specifically the phonon thermal Hall effect and magnon-polarons. Surface phonon Hall viscosity provides a mechanism to control phonon chirality and nonreciprocity via surface magnetization configurations in magnetic topological insulator films.

Two of the most iconic systems of quantum physics are the particle in a box and the Coulomb potential (the third is, of course, the harmonic oscillator). In this expository paper, we consider the quantum solution to the problem of an electron confined between the grounded planes of an infinite capacitor. The potential arises from the image charges that form in the grounded planes, along with the added condition that at x = 0, L, where L is the distance between the planes, the wavefunction must be zero. This effectively couples a hydrogen like system to a particle-in-a-box (PIB) based on L, the distance between the planes. The problem of finding the electrostatic potential of this infinite series of image charges is an old one, going back to at least 1929. Here, we give a short derivation for one of the limiting cases that yields a compact expression and show how the Kellogg infinite summation formula converges to that value. We note here that this potential is a symmetric double well potential, so there will be many familiar properties of its solutions. Then using that potential, we solve Schrödinger's equation using a spectral technique. The limiting forms of a particle in a box for small L (and high E), and that of a (degenerate) bound image charge at large L and small energy are recovered. We also discuss the tunneling level splitting that occurs in the transition from the large L to the small L regime.

Despite extensive progress in characterizing the emergent behavior of active matter, the microscopic origins of self-diffusion in interacting active systems remain poorly understood. Here, we develop a framework that quantitatively links self-diffusion to collisional forces and their temporal correlations in active fluids. We show that transport is governed by two contributions: an equal-time suppression of motion arising from anisotropic collisional forces, and a memory correction associated with the temporal persistence of these forces. Together, these effects yield an exact expression for the self-diffusivity in terms of measurable force statistics and correlation times. We apply this framework to purely repulsive active Brownian particles and find that self-diffusion is always reduced, with collisional memory acting as a strictly dissipative correction. Our results establish a direct connection between microscopic force correlations and macroscopic transport, providing a general mechanical perspective for interpreting self-diffusion in active matter.

We study the Lanczos coefficients in a quadratic model given by an impurity interacting with a multi-mode field of fermions, also known as single impurity Anderson model. We analytically derive closed expressions for the Lanczos coefficients of Majorana fermion operators of the impurity for different structures of the coupling to the hybridisation band at zero temperature. While the model remains quadratic, we find that the growth of the Lanczos coefficients structurally depends strongly on the chosen coupling. Concretely, we find $(i)$ approximately constant, $(ii)$ exactly constant, $(iii)$ square root-like as well $(iv)$ linear growth in the same model. We further argue that in fact through suitably chosen couplings, essentially arbitrary Lanczos coefficients can be obtained in this model. These altogether evince the inadequacy of the Lanczos coefficients as a reliable criterion for classifying the integrability or chaoticity of the systems. Eventually, in the wide-band limit, we find exponential decay of autocorrelation functions in all the settings $(i)-(iv)$, which demonstrates the different structures of the Lanczos coefficients not being indicative of different physical behavior.

This study explores how entanglement and quantum teleportation of magnons can be achieved in coupled spin chain systems. By utilizing different magnetic configurations, we show that parallel spin chains function like magnonic beam splitters, whereas anti-parallel chains produce two-magnon squeezing and strong entanglement. Combining these components, we design magnonic circuits capable of continuous-variable quantum entanglement and teleportation, supported by quantum Langevin simulations.

Probabilistic computers replace logic gates with networks of interacting random variables, creating bidirectional systems that can back-derive inputs from outputs. Such architectures enable efficient generation of random samples, implementations of novel algorithms, and natural solutions to classically hard problems such as prime factorization. We present a new physical implementation for these networks: ferromagnetic disks whose magnetization switching process is triggered by current pulses, skewed by external magnetic fields, and randomized by ambient thermal noise. We show that geometry-dependent magnetostatic interactions between these magnetic cells lead to system behavior that emulates deterministic logic gates. Furthermore, by chaining multiple "gates," we achieve a highly accurate bidirectional one-bit full-adder, a proof of concept for complex multi-gate logic functions with reversible information flow. This analog magnetic probabilistic computer methodology improves on other implementations in speed, tunability, and energy efficiency, thereby enabling a powerful new pathway towards practical solution of classically hard problems.

The interplay between the curved spacetimes of general relativity and quantum mechanical systems is an active field of research. However, analysis of relativistic gravitation on extended quantum systems remains understudied. To this end, we study here the effects of a general relativistic curved spacetime on the topological phases of the Su-Schrieffer-Heeger model and Kitaev superconducting wire. We find that the topological states remain robust and well localized. In the topological insulator we find that the energy level of the topological state becomes shifted away from zero according to the gravitational redshift, breaking the system's chiral symmetry. In contrast, the Majorana zero mode of the topological superconductor remains at zero energy. Furthermore, within the topological superconductor, we identify the possibility of a gravitationally induced topological phase transition leading to the formation of a domain wall, shifting one of the boundary Majorana zero modes into the bulk.

We report Beta-AlGaO/Ga2O3 tri-gate heterostructure MOSHEMTs incorporating a thin 5 nm Al2O3 gate oxide layer for improved gate control and reduced leakage. The devices were fabricated on AlGaO/GaO heterostructures grown by ozone MBE on Fe-doped Ga2O3 (010) substrates. The tri-gate MOSHEMTs, with 1 micron-wide fins and Lg=155 nm, exhibit a peak current-gain cut-off frequency fT=70 GHz and a power-gain cut-off frequency fMAX=55 GHz.The fT.L product of 10.85 GHz-micron is the highest among reported Ga2O3 FETs to date. The devices show Vth =-0.5 V, an on/off ratio 10^6 I=80 mA/mm, a peak gm=60 mS/mm, and a low gate leakage current of 10^(-10) mA/mm at Vgs=0.5 V. Passivation with a 100 nm ALD Al2O3 layer effectively removes DC/RF dispersion and maintains stable operation under pulsed IV and repeated RF measurements. These results demonstrate the potential of tri-gate AlGaO/GaO MOSHEMTs for next-generation high-frequency and high-power applications.

We show that magnetism in moiré Mott-Wigner crystals can be induced by the collective vibration of electrons around their equilibrium positions (i.e., electronic phonons), even without spin interactions between electrons. Due to a geometric valley-orbit coupling from the Berry phase effect, the zero-point energy of electronic phonons reaches minimum when electrons are fully valley polarized. This leads to a spontaneous magnetization when below a critical temperature. We also propose to engineer the magnetism through the photoexcitation of chiral electronic phonons.

Fractional Chern insulators are lattice analogs of fractional quantum Hall states that realize fractionalized quasiparticles without an external magnetic field. A key strategy to understand and design these phases is to map Chern bands onto Landau levels (LLs). Here, we introduce a universal framework that variationally decomposes Bloch bands into generalized LLs, providing a controlled and quantitative characterization of their effective LL nature. Applying this approach to twisted bilayer MoTe$_2$ modeled by first-principles-derived moiré Hamiltonians, we find that the first moiré valence band is dominated by the generalized zeroth LL across a broad range of twist angles, facilitating the formation of Abelian fractional Chern insulators in the Jain sequences. The second moiré band, renormalized via Hartree-Fock calculations at hole filling $ν_h = 2$, is dominated by the generalized first LL at twist angles $θ= 2.45^\circ$ and $2.13^\circ$. At $θ= 2.45^\circ$, we find numerical evidence for a non-Abelian Moore--Read (MR) state at $ν_h = 5/2$, with consistent signatures in both the energy spectrum and the particle entanglement spectrum. Interpolation studies further demonstrate an adiabatic connection between this state and the MR state in the conventional first LL. In contrast, at $θ= 2.13^\circ$, a charge-density-wave state prevails in the competition with the MR state due to the larger bandwidth. Our variational mapping provides a theoretical framework for exploring exotic fractionalized phases, including non-Abelian states, in realistic systems.

Unlabeled multigraphs have diverse applications across scientific fields, from transportation and social networks to polymer physics. In particular, multigraphs are essential for studying the relationship between the spatial organization and biological function of chromatin, which is often folded into complex polymer networks whose structure is closely tied to patterns of gene expression. A fundamental yet challenging aspect in applying graph theory to these areas is the enumeration of multigraphs, especially under structural constraints For example, when coupled with the statistical mechanics of polymer networks, the ability to identify traversable and connected multigraphs provides powerful tools for predicting statistically favored motifs that may arise within chromatin networks. In this work, by counting the adjacency matrices, we derive polynomial expressions that enumerate all connected, undirected, and unlabeled multigraphs with three nodes and fixed degree, and provide a method to efficiently generate them.

Spike timing offers a combinatorial address space, suggesting that timing-based spiking inference can be executed as lookup and routing rather than as dense multiply--accumulate. Yet most neuromorphic and photonic systems still digitize events into timestamps, bins, or rates and then perform selection in clocked logic. We introduce Polychronous Wave Computing (PWC), a timing-native address-selection primitive that maps relative spike latencies directly to a discrete output route in the wave domain. Spike times are phase-encoded in a rotating frame and processed by a programmable multiport interferometer that evaluates K template correlations in parallel; a driven--dissipative winner-take-all stage then performs a physical argmax, emitting a one-hot output port. We derive the operating envelope imposed by phase wrapping and mutual coherence, and collapse timing jitter, static phase mismatch, and dephasing into a single effective phase-noise budget whose induced winner--runner-up margin predicts boundary-first failures and provides an intensity-only calibration target. Simulations show that nonlinear competition improves routing fidelity compared with noisy linear intensity readout, and that hardware-in-the-loop phase tuning rescues a temporal-order gate from 55.9% to 97.2% accuracy under strong static mismatch. PWC provides a fast routing coprocessor for LUT-style spiking networks and sparse top-1 gates (e.g., mixture-of-experts routing) across polaritonic, photonic, and oscillator platforms.

Complex fluids such as water exhibits many anomalous phenomena, and research suggests these properties are closely tied to critical fluctuations near the liquid-liquid phase transition critical point (LLCP). However, whether a similar LLCP exists in metallic glass-forming liquids, which are notable for their high atomic coordination, remains an open question. Although dynamic anomalies such as the breakdown of the Stokes-Einstein (SE) relation have often been attributed to dynamic heterogeneity or structural changes, relatively few studies have analyzed these anomalies from a thermodynamic-fluctuation perspective. This gap probably stems from the challenges in detecting density-driven phase transitions in such systems. Here, we use numerical simulations to explore the thermodynamic mechanisms behind dynamic anomalies in a prototypical metallic glass-forming melt. We observe substantial thermodynamic fluctuations near a particular region, which likely corresponds to a frustration state of liquid, vapor, and glass. These fluctuations may contribute to the violation of the SE relation. Our findings offer a fresh Griffiths-like perspective on the dynamic anomalies seen in supercooled metallic liquids, and shed new light on their underlying mechanisms.

Mean-field Hartree theory is a central tool for reducing interacting many-body dynamics to an effective nonlinear one-particle evolution. This approximation has been employed also when the Hamiltonian that governs the many-body dynamics is not Hermitian. Indeed, non-Hermitian Hamiltonians model particle gain/loss or the evolution of open quantum systems between consecutive quantum jumps. Furthermore, the validity of the Hartree approximation for generic non-Hermitian Hamiltonians lies at the basis of a quantum algorithm for nonlinear differential equations. In this work, we show that this approximation can fail. We analytically solve a model of $N$ bosonic qubits with two-body interactions generated by a purely anti-Hermitian Hamiltonian, determine an analytic expression for the limit for $N\to\infty$ of the one-particle marginal state and show that such a limit does not agree with the solution of the non-Hermitian Hartree evolution equation. We further show that there exists an initial condition such that the exact one-particle marginal state undergoes a finite-time transition to a mixed state, a phenomenon that is completely absent in the case of Hermitian Hamiltonians. Our findings challenge the validity of the mean-field Hartree approximation for non-Hermitian Hamiltonians, and call for additional conditions for the validity of the mean-field regime to model the dynamics of particle gain and loss and the open-system dynamics in bosonic many-body systems.

We investigate the topological electronic properties of the $K_4$ crystal by constructing a tight-binding model. The bulk band structure hosts Weyl nodes with higher and conventional chiralities ($χ= \pm 2$ and $χ= \pm 1$) located at high-symmetry points in the Brillouin zone. Through analytical evaluation of the Berry curvature, we identify the positions and chiralities of these Weyl nodes. Furthermore, slab calculations for the (001) surface reveal Fermi arcs that connect Weyl nodes of opposite chirality, including those linking $χ= \pm 2$ nodes with pairs of $χ= \mp 1$ nodes. These results demonstrate that the $K_4$ crystal is a spinless Weyl semimetal featuring topologically protected surface states originating from multiple types of Weyl nodes.

The magnetic-field-induced correlated Chern insulator (CCI) states in magic-angle twisted bilayer graphene (MATBG) have been intensively studied in experiments, but a simple and clear understanding of their origin is still lacking. Here, we propose a unified theoretical framework for the CCI states in MATBG that successfully explains most experimental observations. The key insight of our theory is that, due to the very narrow bandwidth of MATBG, correlation-enhanced valley and spin Zeeman terms are critical for shaping the intricate Hofstadter spectrum, resulting in an interwoven, flavor-resolved (spin and valley) Hofstadter spectrum that can well describe the observed CCI states. Crucially, due to the Zeeman effect, the crossings between these flavor-polarized Hofstadter spectra are magnetic-field-dependent, causing certain CCI states to emerge only above a critical field. This is the main mechanism underlying the critical field phenomenon of the CCI states observed in experiments. Our theory provides a clear and unified physical picture for the correlated Hofstadter spectrum in MATBG.

Laser hardening of metals occurs under the influence of a shock wave, which changes the distribution and density of one-dimensional defects - dislocations. There is a relationship between the density of dislocations, the grain size and the resistance of a single crystal to shear loading. The mechanism of hardening processes continues to be intensively studied, and the dynamics of defects plays a central role here. In this paper, the dislocation entropy is analyzed from a combinatorial point of view and from the point of view of a physical oscillator with a given energy reserve. Both contributions play an important role in describing the free energy of a one-dimensional ensemble of dislocations, and are necessary to take into account the dynamic processes inside the grain of a polycrystalline structure. Keywords: Laser Shock Peening, statistical mechanics

We investigate a nonideal, thermodynamically consistent Brusselator reaction-diffusion (RD) system that explicitly incorporates molecular interactions among species in both the diffusion process and the underlying chemical reaction network. Within this framework, we systematically revisit the Cross-Hohenberg classification of instabilities to assess the feasibility and characteristics of the various types of instability arising from the interplay between entropic and energetic contributions. Our analysis demonstrates that only type I and type III instabilities (the Cross-Hohenberg classification) can occur in this system; Energetic contributions do not explicitly generate instabilities, but may implicitly control their occurrence through their influence on the fixed-point (steady-state) concentrations. In cases where instabilities of different types coexist, we show that the resulting patterns are highly sensitive to the relative strengths of the competing instabilities.

Giant Shubnikov-de Haas oscillations (SdHO) with V-shaped minima are experimentally studied in a high-mobility two-dimensional electron gas based on GaAs/AlGaAs heterostructures. A phenomenological model with two parameters (transport momentum relaxation time $τ_{\text{tr}}$ and quantum scattering time $τ_q$) is developed, accurately describing experimentally measured magnetoresistance over an unexpectedly wide range of magnetic fields (up to 3.5 T) and temperatures (from 2 K to 15 K). The model combines: (i) a quasiclassical density of states with a magnetic-field-dependent Gaussian broadening of Landau levels, (ii) a momentum relaxation time scaling with the density of states, and (iii) oscillations of the Fermi level at a fixed electron density. This model reproduces V-shaped oscillation minima with zero-resistance points, a smooth background of positive magnetoresistance, and enables the extraction of $τ_q$ and $τ_{\text{tr}}$ even in microstructures where ballistic and viscous effects dominate at low fields. As expected, the temperature dependence reveals that $τ_{\text{tr}}$ scales inversely with temperature due to acoustic phonon scattering, while $τ_q$ remains temperature-independent.

We present a multiscale modeling approach that integrates molecular dynamics simulations, machine learning, and the Elastically Collective Nonlinear Langevin Equation (ECNLE) theory to investigate the glass transition dynamics of polymer systems. The glass transition temperatures (Tg) of four representative polymers are estimated using simulations and machine learning model trained on experimental datasets. These predicted Tg values are used as inputs to the ECNLE theory to compute the temperature dependence of structural relaxation times and diffusion coefficients, and the dynamic fragility. The Tg values predicted from simulations show good quantitative agreement with experimental data. While machine learning tends to slightly overestimate Tg, the resulting dynamic fragility values remain close to experimental fragilities. Overall, ECNLE calculations using these inputs agree well with broadband dielectric spectroscopy results. Our integrated approach provides a practical and scalable tool for predicting the dynamic behavior of polymers, particularly in systems where experimental data are limited.

Fluctuation-induced forces, such as the Critical Casimir Effect (CCE), are fundamental mechanisms driving organization and self-assembly near second-order phase transitions. The existence of a comparable, universal force for systems undergoing a first-order transition has remained an unresolved fundamental question. The proposed Orderphobic Effect is one such potential mechanism. It arises from minimisation of the interfacial free energy between solutes that locally nucleate a disordered phase. Here, we report the first experimental demonstration and quantitative measurement of the Orderphobic Effect. Using a driven, non-equilibrium quasi-2D granular fluid undergoing a first-order order-disorder transition, we show that specifically designed solutes in an ordered phase nucleate a coexisting ``bubble'' of the disordered phase. By analysing its capillary fluctuations, we confirm that this phenomenon occurs due to the proximity to phase-coexistence, and we directly quantify the attractive force by measuring the interaction free energy between solutes. The observation of this general fluctuation-mediated force in a non-equilibrium steady state strongly supports its claimed universality. Our work establishes the Orderphobic Effect as the first-order equivalent to the CCE, providing a new, general route for controlling self-assembly and aggregation in soft matter and non-equilibrium systems.

Quantum dot-superconductor hybrids have been established as a suitable platform for realizing Kitaev chains hosting Majorana bound states. Implementing these structures in a qubit architecture is expected to result in coherence times that scale exponentially with the lengths of the chains. To scale to longer systems, the phase differences between all superconducting segments in the chain need to be controlled. While this control has been demonstrated by using an external magnetic flux, ideally it can be achieved with control over intrinsic system parameters. In this work, we investigate whether the relevant phase differences can be tuned through the spin degree of freedom in each QD, or the chemical potential of the discrete bound states in the hybrid sections. We confirm that both these tuning knobs allow for controlling the phase difference in the couplings between neighbouring QDs, bypassing the requirement to tune an external flux. However, we find that the amplitude of the phase shifts can deviate from a discrete $π$-shift. We introduce a spatial variation in the spin-orbit field as a possible mechanism to explain the observed behaviour and comment on the consequences for experimentally creating long Kitaev chains.

We establish a generalized work theorem for stochastic chemical reaction networks (CRNs). By using a compensated Poisson jump process, we identify a martingale structure in a generalized entropy defined relative to an auxiliary backward process and extend nonequilibrium work relations to processes stopped at bounded arbitrary times. Our results apply to discrete, mesoscopic chemical reaction networks and remain valid for singular initial conditions and state-dependent termination events. We show how martingale properties emerge directly from the structure of reaction propensities without assuming detailed balance. Stochastic simulations of a simple chemical kinetic proofreading network are used to explore the dependence of the exponentiated entropy production on initial conditions and model parameters, validating our new work theorem relationships. Our results provide new quantitative tools for analyzing biological circuits ranging from metabolic to gene regulation pathways.

Conventional Raman spectroscopy faces inherent limitations in detecting interlayer layer breathing (LB) vibrations with inherently weak electron-phonon coupling or Raman inactivity in two-dimensional materials, hindering insights into interfacial coupling and stacking dynamics. Here we demonstrate a universal plasmon-enhanced Raman spectroscopy strategy using gold or silver nanocavities to strongly enhance and detect LB modes in multilayer graphene, hBN, and their van der Waals heterostructures. Plasmonic nanocavities even modify the linear and circular polarization selection rules of the LB vibrations. By developing an electric-field-modulated interlayer bond polarizability model, we quantitatively explain the observed intensity profiles and reveal the synergistic roles of localized plasmonic field enhancement and interfacial polarizability modulation. This model successfully describes the behavior across different material systems and nanocavity geometries. This work not only overcomes traditional detection barriers but also provides a quantitative framework for probing interlayer interactions, offering a versatile platform for investigating hidden interfacial phonons and advancing the characterization of layered quantum materials.

Collective excitations are fundamental in quantum many-body physics, yet their spectra have traditionally been studied within Newtonian dynamics. In this Letter, we investigate collective excitations in Bose gases under Modified Newtonian Dynamics (MOND). We derive an anisotropic excitation spectrum in the MOND regime. This anisotropy arises directly from the intrinsic nonlinear structure of the MOND Poisson equation, forming a distinctive signature of the modified gravitational response. We then analyze the Jeans instability, obtaining analytic expressions for the direction-dependent critical wavelength and mass. These results advance our understanding of collective behavior in quantum systems under modified dynamics and establish clear theoretical signatures for testing MOND-like effects in quantum simulators.

Traditional statistical mechanics is constrained by the binary paradigms of identical/distinguishable and bosonic/fermionic particle statistics, leading to a fundamental logical gap in describing systems with partial distinguishability. We propose WH Statistics, a unified theoretical framework governed by three key parameters: continuous distinguishability λ, exclusion weight \k{appa}, and intrinsic exclusivity γ. By deriving the microstate count and entropy, we show that this framework naturally recovers the Bose-Einstein, Fermi-Dirac, and Maxwell-Boltzmann statistics, while also incorporating anyons and the classical hard-core (Langmuir) limit. We introduce a class of generalized quasiparticles, termed WHons, which exhibit exotic physical phenomena including non-monotonic degeneracy pressure peaks, Schottky-like specific heat anomalies, and tunable interference effects, driven by the interplay between fractional distinguishability and exclusion. This framework bridges the century-old discontinuity between quantum and classical exclusion principles, providing a powerful tool for investigating strongly correlated systems and programmable quantum matter.

We propose a method to prepare and certify Gaussian quantum states of the ferromagnetic resonance spin-wave modes in ferromagnets using a longitudinal drive. Contrary to quantum optics-based strategies, our approach harnesses a purely magnonic feature - the spin-wave nonlinearity - to generate magnon squeezing. This resource is used to prepare vacuum-squeezed states, as well as entangled states between modes of different magnets coupled via a microwave cavity. We propose methods to detect such states with classical methods, such as ferromagnetic resonance or local pickup coils, and quantify the required detection efficiency. We analytically solve the case of ellipsoidal yttrium iron garnet ferrimagnets, but our method applies to a vast range of shapes and sizes. Our work enables quantum magnonics experiments without single-magnon sources or detectors (qubits), thus bringing the quantum regime within reach of the wider magnonics community.

Klein tunneling stands as a fundamental probe of relativistic quantum transport in two-dimensional materials. We investigate this phenomenon in quadratic band-touching systems, where the Hilbert-Schmidt quantum distance plays a central role in the underlying mechanism. By employing a generic parabolic model, we systematically disentangle the cooperative effects of intrinsic mass asymmetry and tunable quantum geometry. We demonstrate that mass asymmetry sets the overall transmission profile, including the angular distribution and the resonance channels. In contrast, we show that quantum geometry provides a universal parameter that modulates tunneling efficiency by tuning the quantum distance, while leaving the energy dispersion unchanged. Specifically, quantum geometry plays a dual role: it governs the overall transmission amplitude through pseudospin mismatch, while its interplay with Fabry-Perot interference induces observable shifts in resonance angles. Our findings reveal that incorporating quantum geometry alongside band structure is essential for a complete description of quantum transport.

We study the thermal conductivity of amorphous solids by constructing a continuum model whose degrees of freedom are propagating vibrational modes (phonons) and extended Volterra dislocation line defects with their own vibrational degrees of freedom which do not propagate in space. Our working assumption is that these additional degrees of freedom account for the "boson peak" that is observed in glassy materials. This identification allows us to obtain the length distribution of dislocations from experimental data of the boson peak for each material, which we use as input to calculate the phonon self-energy in a quantum field theory framework using that the phonon-dislocation interaction is given by the Peach-Koehler force. The tail of the distribution for long dislocations is consistent with an $L^{-5}$ power law. Our results show that this power law yields a linear rise in the thermal conductivity, as observed in glasses at low temperatures. We then consider two approaches to describe thermal conductivity data quantitatively. In the simplest approach we only keep the low-frequency behavior of the phonon self-energy with one free parameter, plus an adjustable UV cutoff. In the more realistic approach we keep the full frequency dependence of the phonon self-energy as dictated by the phonon-dislocation interaction plus an additional contribution due to scattering with point defects, with a cutoff set by the typical interatomic spacing of the material. We obtain a satisfactory description of thermal conductivity data with both approaches. We conclude by discussing prospects to test the predictive power of this model.

Complex systems with nonreciprocal interactions are often stratified into layers. Ecosystems are a prime example, where species at one trophic level grow by consuming those at another. Yet the dynamical consequences of such stratified nonreciprocity -- where the correlation between growth and consumption differs across trophic levels -- remain unexplored. Here, using an ecological model with three trophic levels, we reveal an emergent asymmetry: nonreciprocal interactions between consumers and predators (top and middle level) destabilize ecosystems far more readily than nonreciprocity between consumers and resources (middle and bottom level). We analytically derive the phase diagram for the model and show that its stability boundary is controlled by energy flow across trophic levels. Because energy flows upward -- from resources to predators -- diversity is progressively lower at higher trophic levels, which we show explains the asymmetry. Lowering energy flow efficiency flips the asymmetry toward resources and remarkably expands the stable region of the phase diagram, suggesting that the famous "10% energy transfer" seen in natural ecosystems might promote stability. More broadly, our findings show that the location of nonreciprocity within a complex network, not merely its magnitude, determines stability.

We identify a previously unrecognized spin-lattice coupling that is allowed in the presence of broken inversion symmetry that can be considered as a lattice analogue to the electronic Rashba spin-orbit coupling. In the low-frequency regime with magnons integrated out, the interfacial spin-lattice coupling is shown to engender a kineo-elastic term in the phonon Lagrangian that couples the strain on the lattice to its velocity and thereby gives rise to a nonreciprocity in transverse phonon velocity. We further analyze the full magnon-phonon spectrum and uncover directional hybridization and absorption, leading to asymmetric phonon propagation lengths for opposite directions. Our results indicate that such interfacial spin-lattice coupling can serve as an efficient route to achieve nonreciprocal phonon propagation properties in magnetic heterostructures with strong Rashba spin-orbit coupling.

Combining high-speed photography with direct force measurements, we investigate the impact dynamics of drops of cornstarch-water mixtures -- a premier example of shear-thickening fluids -- across a wide range of impact conditions. Our study identifies three distinct impact regimes. In addition to the liquid-like and solid-like behaviors generally expected for the impact-induced response of shear-thickening fluids, we uncover a counterintuitive regime in which high-concentration cornstarch-water mixtures display a liquid-like response at the onset of impact when shear rates are high and only transition to a solid-like behavior at later times as shear rates reduce. By integrating the classic drop-impact theory with the Reynolds-Darcy mechanism for dilatancy, we develop a unified model that quantitatively describes the impact dynamics of shear-thickening drops across all regimes. Our work reveals the unexpected response of shear-thickening fluids to ultra-fast deformation and advances fundamental understanding of drop impact for complex fluids.

In finite-size systems undergoing a continuous phase transition, the passage from the symmetric phase to the broken-symmetry phase is accomplished through a hysteresis zone, up to spontaneous symmetry breaking (SSB). In the present work, we find that a resonance phenomenon takes place within this zone. This resonance is manifested as a maximization of the mean waiting time as a function of temperature inside the hysteresis region. An interesting issue concerns how this resonance is connected with the existence of particles (tachyons) or quasiparticles (kink solitons) within the hysteresis zone. Finally, we introduce the idea that this resonance delineates a continuous passage from a "classical" phase to a "quantum" phase for a binary system, such as the three-dimensional Ising model, which belongs to the same universality class as a fermion-antifermion system or, more generally, a matter-antimatter system.

In the present paper we derive a further extension of the results contained in two recent articles, both published in Open Communications in Mathematical Physics, where it was shown that the integrable version of the N-species Volterra model, introduced by V. Volterra in 1937, is in fact maximally superintegrable. Here we point out that the superintegrability property applies as well to the case of infinitely many competing species, either countable or uncountable. Analytical and numerical results are given.

Cooperative effects in oscillator networks are often associated with enhanced stability of phase-locked solutions, which increases with system size. We show that the stabilization of persistent currents in annular atomic superfluids with periodic barriers is a concrete manifestation of this phenomenon. Under the simplifying assumption of continuity of atomic flow across identical barriers, the system reduces to a ring of locally coupled Kuramoto-like oscillators. We analytically derive the stability diagram of phase-locked configurations and quantify their robustness to noise and small random initial imperfections, finding excellent agreement with experimental observations. These results are inherent to the ring topology and independent of the specific physical platform.

We obtain an exact analytic expression for the dynamical structure factor of one-dimensional quantum gas of hard rods. Our result is valid for arbitrary many-body state of the system, with finite temperature states and the ground state being important special cases that we analyse in detail. We demonstrate that the expression obeys fundamental relations such like the f-sum rule and the detailed balance. We also reveal the hidden fermionic structure behind the correlator. In the static limit we show that it can be written in terms of universal functions which, at zero temperature, coincide with the level spacing distribution function of the Gaussian Unitary Ensemble. Our work provides a full and exact characterisation of a dynamic correlation function in a strongly correlated interacting quantum many-body system.

We consider a formal (approximate) integral of motion in Hamiltonians of the form $H=\frac{1}{2}(X^2+Y^2+ω_1^2x^2+ω_2^2y^2)+ε(ηxy^2+αx^3+βx^2y+γy^3)$ generalizing previous cases with $β=γ=0$. First we give the general form of this integral when $ω_1/ω_2$ is irrational and then we consider the case of commensurable frequencies. In particular we study the integrals for the resonances $ω_1/ω_2=4/1, 5/1, 3/2, 4/3, 3/1$ and $2/1$. We also calculate the invariant curves and the orbits in the cases $ω_1/ω_2=2/1$ and $1/1$ (with $β=γ=0$) and we compare the exact-numerical and the theoretical results predicted by the formal integral when $βγ\neq0$. In the special case $ω_1/ω_2=1/1$ we find an integral when $β=γ=0$ and $ηα\neq0$ or $η=α=0$ and $βγ\neq 0$, but this is not possible when $ηαβγ\neq 0$. However, we find that the invariant curves and the orbits can be approximated by a non-resonant integral with $ω_1/ω_2=5\sqrt{2}/7=1.010\dots$.

We investigate neuronal excitability in the Hodgkin-Huxley model under temporal interference (TI) stimulation in a previously unexplored sub-Hz resonant regime and uncover a striking nonlinear response that we term 'multi-tongue frequency fractals'. Unlike single-frequency driving, which yields a smooth resonant valley, dual-frequency excitation fragments this response into a hierarchy of sharply modulated tongues whose number and structure grow with observation time, revealing clear self-similar architecture. These features emerge from transitions between non-cascaded and cascaded high-harmonic and sub-harmonic generation as detuning varies, and are maximized near the intrinsic ionic timescale at omega ~ 0.2 rad/s. Parameter sweeps show that the fractal count is higher for higher potassium conductances, lower sodium conductances and lower leak conductances. These results demonstrate that TI stimulation can elicit rich, hierarchically organized frequency responses even in classical excitable membranes, revealing fractal organization in Hodgkin-Huxley dynamics.

We consider the novel nonlinear model in (1 + 1)-dimensions for Dirac spinors recently introduced by Alexeeva, Barashenkov, and Saxena [1] (ABS model), which admits an exact explicit solitary-wave (soliton for short) solution. The charge, the momentum, and the energy of this solution are conserved. We investigate the dynamics of the soliton subjected to several potentials: a ramp, a harmonic, and a periodic potential. We develop a Collective Coordinates Theory by making an ansatz for a moving soliton where the position, rapidity, and momentum, are functions of time. We insert the ansatz into the Lagrangian density of the model, integrate over space and obtain a Lagrangian as a function of the collective coordinates. This Lagrangian differs only in the charge and mass with the Lagrangian of a collective coordinates theory for the Gross-Neveu equation. Thus the soliton dynamics in the ABS spinor model is qualitatively the same as in the Gross-Neveu equation, but quantitatively it differs. These results of the collective coordinates theory are confirmed by simulations, i.e., by numerical solutions for solitons of the ABS spinor model, subjected to the above potentials.

We investigate Hamiltonian fluid reductions of the one-dimensional Vlasov-Poisson equation. Our approach utilizes the hydrodynamic Poisson bracket framework, which allows us to systematically identify fundamental normal variables derived from the analysis of the Casimir invariants of the resulting Poisson bracket. This framework is then applied to analyze several well-established Hamiltonian closures of the onedimensional Vlasov equation, including the multi-delta distribution and the waterbag models. Our key finding is that all of these seemingly distinct closures consistently lead to the formulation of a unified form of parametric closures: When expressed in terms of the identified normal variables, the parameterization across all these closures is revealed to be polynomial and of the same degree. All these parametric closures are uniquely generated from one of the moments, called $μ$2, a cubic polynomial in the normal variables. This result establishes a structural connection between these different physical models, offering a path toward a more unified and simplified description of the one-dimensional Vlasov-Poisson dynamics through its reduced hydrodynamic forms with an arbitrary number of fluid variables.

The rarely used Hamilton-Jacobi equation has been utilized as an elegant way to find the trajectories of mechanical systems and to derive symplectic maps. Further, the exact solution in kick approximation of Hamilton's equations of motion in interaction representation is written as a generalized one-turn twist map. One can imagine that the nonlinear kick comes first, followed by the one-period rotation along the machine circumference, or a second alternative in which the one-period rotation occurs before the kick. There is a difference in the result of solving Hamilton's equations between the two cases, which is expressed in obtaining a standard forward twist map in the first case, or alternatively a backward map in the second one. This nontrivial and intuitively unclear peculiarity is usually ignored/overlooked in practically all specialized references on the topic. Finally, the statistical properties and the behavior of the density distribution of a particle beam in configuration space under the influence of an isolated sextupole have been studied.

We propose a financial liquidity policy sharing method for firm-to-firm supply networks, introducing a scalable autonomous control function for viable complex adaptive supply networks. Cooperation and competition in supply chains is reconciled through overlapping collaborative sets, making firms interdependent and enabling distributed risk governance. How cooperative range - visibility - affects viability is studied using dynamic complex adaptive systems modelling. We find that viability needs cooperation; visibility and viability grow together in scale-free supply networks; and distributed control, where firms only have limited partner information, outperforms centralised control. This suggests that policy toward network viability should implement distributed supply chain financial governance, supporting interfirm collaboration, to enable autonomous control.

Double ramification (DR) hierarchies associated to rank-$1$ F-CohFTs are important integrable perturbations of the Riemann--Hopf hierarchy. In this paper, we perform bihamiltonian tests for these DR hierarchies, and conjecture that the ones that are bihamiltonian form a $2$-parameter family. Remarkably, our computations suggest that there is a $1$-parameter subfamily of the rank-$1$ F-CohFTs, where the corresponding DR hierarchy is conjecturally Miura equivalent to the Camassa--Holm hierarchy. We also prove a conjecture regarding bihamiltonian Hodge hierarchies. Finally, we systematically study Miura invariants, and for another $1$-parameter subfamily propose a conjectural relation to the Degasperis--Procesi hierarchy.

We study the escape of particles in the lemon billiard, a two-parameter family of billiard systems defined by the intersection of two identical circles. Using numerical simulations, we explore how the survival probability depends on the position and size of the hole, as well as on the billiard shape parameter. We find that the survival probability exhibits a two-stage decay pattern: an initial exponential regime followed by a long-time power-law tail, a signature of the stickiness effect. Our results show that the short-time exponential decay rate follows a power-law dependence on the hole size, with different scaling exponents for holes placed in chaotic regions versus mixed phase space regions. For holes located in mixed phase space regions, the decay exponent of the long-time power-law tail remains approximately constant, while the amplitude follows a power-law scaling with hole size. We also examine the dependence of short-time exponential decay rate on the billiard shape parameter and observe scaling behavior for small values of this parameter, which breaks down as the parameter increases.

Long arrays of identical, self-propelling flapping flyers are inherently unstable and thus unlikely to exist without active control mechanisms. One approach to enable long in-line formations is to enforce a constant separation between the group members. The objective then becomes to determine the flapping strategies the flyers should adopt to achieve a certain separation. Using an aerodynamic model of vortex wake production and inter-flyer effects, we explore different flapping strategies for followers given the motion of the leader. The choice of tactic is dependent upon the aerodynamic, kinematic, and physical parameters of the system, and reflects an interplay between efficiency and stability. We find that whether a flyer flaps in or out of phase with its upstream neighbour, together with the target separation, strongly affect the flapping amplitude and, therefore, the energetic cost of group flight. In certain regimes, group flight is energetically favourable compared to isolated flight, while in others, flying in formation becomes less efficient. We also identify "goldilocks zones", ranges of separation in which one of the in- or out-of-phase motions can be simultaneously energetically efficient and dynamically stable. Outside these regions, energetically favourable flight is unstable and therefore unlikely to occur.

We present QDK/Chemistry, a software toolkit for quantum chemistry workflows targeting quantum computers. The toolkit addresses a key challenge in the field: while quantum algorithms for chemistry have matured considerably, the infrastructure connecting classical electronic structure calculations to quantum circuit execution remains fragmented. QDK/Chemistry provides this infrastructure through a modular architecture that separates data representations from computational methods, enabling researchers to compose workflows from interchangeable components. In addition to providing native implementations of targeted algorithms in the quantum-classical pipeline, the toolkit builds upon and integrates with widely used open-source quantum chemistry packages and quantum computing frameworks through a plugin system, allowing users to combine methods from different sources without modifying workflow logic. This paper describes the design philosophy, current capabilities, and role of QDK/Chemistry as a foundation for reproducible quantum chemistry experiments.

Single Pixel Imaging is an emerging imaging technique that employs a bucket detector (photodiode) to sample a spatially modulated light field, rather than measuring the spatial distribution with an array of detectors. This approach provides a low-cost alternative for imaging at unconventional wavelengths and enables improved signal collection in noisy measurement environments. Furthermore, it allows the application of compressive sensing to reduce the amount of acquired data and measurement time, facilitating live or in vivo imaging applications. This tutorial presents the experimental implementation of measurement bases and compressive sensing reconstruction methods, including both deterministic algorithms and deep learning approaches. Accompanying Python notebooks guide readers through the reproduction of the presented results and support the application of the methods to their own work.

Contagion processes in social systems often involve interactions that go beyond pairwise contacts. Higher-order networks, represented as hypergraphs, have been widely used to model multi-body interactions, and their presence can drastically alter contagion dynamics compared to traditional network models. However, existing analytical approaches typically assume independence between pairwise and higher-order degrees, and thus study their roles in isolation. In this paper, we develop an effective hyperdegree model (EHDM) to describe Susceptible-Infected-Susceptible (SIS) dynamics on hypergraphs that explicitly captures correlations between the distribution of groups with different sizes. Our effective hyperdegree model shows excellent agreement with stochastic simulations across different types of higher-order networks, including those with heterogeneous degree distributions. We explore the critical role of cross-order degree correlations, specifically, whether nodes that are hubs in pairwise interactions also serve as hubs in higher-order interactions. We show that positive correlation decreases the epidemic threshold and anti-correlation temporally desynchronizes infection pathways (pairwise and group interactions). Finally, we demonstrate that, depending on the level of correlation, the optimal control strategy shifts -- from one that is purely pairwise- or higher-order-focused to one in which a mixed strategy becomes optimal.

Proton minibeam (pMB) radiotherapy, delivers highly heterogeneous dose distributions alternating high-dose peaks and low-dose valleys. This aims to widen the therapeutic window by improving normal tissue sparing while maintaining the same or even better tumour control. The performance of pMB strongly depends on the collimator design and physical parameters. To better understand the physical and radiobiological drivers of this enhanced therapeutic window, we perform a detailed microdosimetric characterization of proton minibeams and assess their impact. We characterize radiation quality with microdosimetry through Monte Carlo simulations. Then we extend the Generalized Stochastic Microdosimetric Model to predict the normal tissue complication probability (NTCP) at different depths in water, 1cm, 2cm, and 4cm, for 100MeV proton minibeams realized with varying configurations of collimator. Results are compared with conventional homogeneous field (HF) irradiation after dose normalization to the tumor. The developed model is applied by considering tissues as divided into several functional subunits, connected by a seriality parameter. Microdosimetric characterization of proton minibeam irradiation shows differences between peak and valley regions in shaping lineal energy spectra, especially at low depth, while radiation quality uniforms progressively getting closer to the tumor. NTCP calculations results suggest an increased sparing effect for pMB over conventional HF. A strong dependence is found on the peak-to-valley dose ratio (PVDR), and on the seriality parameter. Predictions indicate substantial sparing from pMB, especially for PVDR > 15, including relatively serial organs with seriality around 0.7. This integrated dose-microdosimetric-biological framework elucidates how spatial fractionation, radiation quality, and organ architecture collectively shape tissue sparing in pMB.

LiDRoSIS is an automated MATLAB-Python software suite for the segmentation and quantification of lipid droplets (LDs) and reactive oxygen species (ROS) in fluorescence microscopy images of irradiated A549 and MCF7 cells exposed to gold-based nanoparticles. It combines classical image processing algorithms with statistical post-analysis through a companion Python tool, StatLysis. The platform enables reproducible, high-throughput analysis of morphological and spectral parameters linked to nanoparticle-enhanced radiobiological responses. By bridging imaging and quantitative analytics, LiDRoSIS provides a robust framework for nanomedicine and radiation biology research.

This study investigates the mathematical existence and asymptotic properties of Ulanowicz's structural resilience in complex systems such as supply chain networks. While ecological evidence suggests that sustainable systems gravitate toward an optimal state at $α= 1/\mathrm{e}$, the universality of this configuration in generalized networks remains theoretically unverified. We prove that while optimal resilience is unattainable in two-node networks due to structural over-determinacy, it exists for any directed graph with $N_\mathcal{V} \geq 3$. By constructing a symmetric network model with three types of link weights $(x, y, z)$ and uniform marginal distributions, we derive the governing equations for the optimal resilience configuration. Our analytical and numerical results reveal that as the network size $N_\mathcal{V}$ increases, the link weights required to maintain optimal resilience exhibit a power-law scaling behavior: the adjacent links scale as $O(N_\mathcal{V}^{-1})$, while the non-adjacent links scale as $O(N_\mathcal{V}^{-2})$, both accompanied by specific logarithmic corrections. This work establishes a rigorous mathematical foundation for the optimal resilience framework and provides a unified perspective on how entropy-based principles govern the robustness and evolution of large-scale complex networks, which may offer quantitative guidance for designing large-scale networked systems under robustness constraints.

Field emission coupled with molecular dynamics simulation (FEcMD) software package is a computational tool for studying atomic structure evolution, structural deformation, phase transitions, recrystallization as well as electron emission characteristics of micro- and nano-protrusions and nanowires consisting of elemental metals or multi-component alloys by means of multi-physics and multi-scale methodology. Current implementations of molecular dynamics simulation coupled with multi-scale electrodynamics (ED) and heat conduction (HC) in FEcMD program are advanced mainly in the two aspects as follows. In electrodynamics, the FEcMD program incorporates the space charge interactions (space charge potential and exchange-correlation effects) in the self-consistent solved Poisson-Schrödinger equation with Wentzel-Kramers-Brillouin-Jeffreys (WKBJ) approximation to evaluate the field emission current density and the related resistive heating process more reliably for nanowires or nano-protrusions especially for nano-gaps between two metal electrodes. Meanwhile, the two-temperature heat conduction model is implemented in electrodynamics coupled with molecular dynamics simulations (ED-MD), providing more dedicated descriptions for the hierarchical electron-phonon two-channel heat conduction mechanism and the temperature evolutions of electron and phonon subsystems under the radiofrequency (RF) or pulse electric fields. Benchmark tests are performed for some key implementations in FEcMD software to validate the numerical results, and also to demonstrate the use of program to study the atomic structure evolution of metal nano-structures under electric field and heating processes.

Cadaveric studies have provided important insights into the mechanics of the human foot arch and plantar fascia. However, repeatedly probing posture-dependent viscoelastic responses immediately after landing impact is difficult in biological specimens, leaving the contribution of skeletal architecture to landing dynamics incompletely understood. In this study, we developed an anthropomimetic foot joint structure aimed at replicating the skeletal geometry of the human foot. Using a vertical drop apparatus that simulates landing and a viscoelastic system-identification model, we investigated how skeletal structure and posture modulate the apparent post-impact viscoelastic response. The results show that the multi-jointed anthropomimetic structure exhibited a higher damping ratio than simplified flat and rigid feet. Moreover, ankle dorsiflexion and toe extension systematically shifted the identified parameters, reducing the damping ratio under the tested conditions. Taken together, these findings indicate that an arch-like, multi-jointed skeletal architecture can enhance impact attenuation in an anthropomimetic mechanical foot, and that morphology and passive posture alone can tune the trade-off between attenuation and rebound. The observed posture-dependent trends are qualitatively consistent with reported differences in human landing strategies, suggesting that skeletal architecture may partly account for the modulation. Furthermore, these results highlight the engineering advantage of anatomically informed skeletal replication for achieving human-like apparent viscoelastic behavior through postural adjustment during landing.

The Hodgkin-Huxley equations have described neuronal excitability for seventy years, yet their mathematical structure-gating exponents m3h and n4, exponential voltage dependencies, and bounded activation variables, has remained empirically justified rather than theoretically derived. Hodgkin and Huxley introduced voltage-dependent conductances controlled by gating variables. While these equations reproduce experimental observations, they were derived through curve-fitting without theoretical justification. Modern theoretical physics derives governing equations from symmetry principles through Lie group theory. We prove that the complete Hodgkin-Huxley equations necessarily follow from three fundamental symmetries: (1) compact conformational state spaces, (2) multiplicative conductance scaling, and (3) temporal translation invariance. These symmetries uniquely determine a Lie group structure isomorphic to SO(2) semidirect product with R2. From representation theory, we derive: boundedness from SO(2) compactness, exponential Boltzmann factors from scale invariance, specific integer exponents m3h and n4 from irreducible representations, and first-order kinetics from Lie algebra flows. This demonstrates that the HH equations are not empirical curve-fits but the unique mathematical structure mandated by fundamental symmetries. We reveal why gating variables must be bounded, voltage dependencies must be exponential, sodium requires three activation gates and one inactivation gate, potassium requires four activation gates, and kinetics must be first-order. This establishes that neural electrophysiology obeys the same theoretical framework as modern physics, where symmetries determine dynamics, providing a foundation for understanding channel mutations and network dynamics through group theory.

Understanding the connection between active galactic nuclei and star-formation (the AGN-SF connection) is one of the longest standing problems in modern astrophysics. In the age of large Integral Field Unit (IFU) surveys, studies of the AGN-SF connection greatly benefit from spatially resolving AGN and SF contributions to study the two processes independently. Using IFU data for 54 local active galaxies from the S7 sample, we present a new method to separate emission from AGN activity and SF using mixing sequences observed in the [NII]$λ6584$/H$α$ vs. [OIII]$λ5007$/H$β$ Baldwin-Phillips-Terlevich (BPT) diagram. We use the new decomposition method to calculate the H$α$ star-formation rate and AGN [OIII] luminosity for the galaxies. Our new method is robust to outliers in the line-ratio distribution and can be applied to large galaxy samples with little manual intervention. We infer star-formation histories (SFHs) using pPXF, conducting detailed recovery tests to determine the quantities that can be considered robust. We test the correlation between the AGN Eddington ratio, using the proxy L[OIII]/$σ_*^4$, and star-formation properties. We find a moderately strong correlation between the Eddington ratio and the star-formation rate (SFR). We also observe marginally significant correlations between the AGN Eddington ratio and the light-weighted stellar age under 100 Myr. Our results point to higher AGN accretion being associated with young nuclear star formation under 100 Myr, consistent with timelines presented in previous studies. The correlations found in this paper are relatively weak; extending our methods to larger samples, including radio-quiet galaxies, will help better constrain the physical mechanisms and timescales of the AGN-SF connection.

Developing epithelial tissues coordinate cell proliferation and mechanical forces to achieve proper size and shape. As epithelial cells tightly adhere together to form the confluent tissue, the distribution of cell areas significantly influences possible patterns of cellular packing and thereby also the mechanics of the epithelium. Therefore, it is important to understand the origin of cell area heterogeneity in developing tissues and, if possible, how to control it. Previous models of cell growth and division have been successful in accounting for experimentally observed area distributions in cultured cells and bacterial colonies, but developing tissues present additional complexity due to self-organized patterns of mechanical stresses that guide morphogenesis. Here, we address this challenge focusing on the D. melanogaster wing disc epithelium. We consider a simple model that couples cell cycle dynamics to tissue mechanics. From time-lapse imaging of the cellular network, we extract all model parameters - cell growth rates, division rates, and mechanical fluctuations - revealing that they all depend on cell size. With these independently measured parameters, our model quantitatively reproduces the observed cell area distribution without any fitting parameters and further predicts tissue pressure gradients, in quantitative agreement with previously published data. Importantly, we find that cell proliferation accounts for 85% of cell area variance, establishing it as the dominant source of packing disorder that influences tissue mechanics and organization.

The Fokker-Planck (FP) particle method accelerates rarefied-gas simulations by replacing the binary collisions of the commonly used Direct Simulation Monte Carlo (DSMC) method with a drift=diffusion process. Like all particle methods, the FP method is inherently stochastic, which leads to statistical fluctuations in macroscopic quantities and necessitates large particle numbers for accurate results. In this work, we investigate the use of quasi-random numbers, which sample distributions more evenly and thereby reduce the variance. To preserve the low-discrepancy structure across time steps, we employ the Array Randomized Quasi-Monte Carlo (Array-RQMC) technique. We combine the FP method with Array-RQMC and compare it in homogeneous and inhomogeneous problems with other commonly used variance-reduction techniques. The proposed FP-Array-RQMC approach achieves improved convergence rates compared with pseudo-random sampling and yields smaller estimator errors for sufficiently large particle numbers.

We study parameter estimation for the transport coefficients of the quark-gluon plasma by differentiating open-quantum-system-based Monte Carlo simulations of quarkonium suppression. The underlying simulator requires solving a Lindblad equation in a large Hilbert space, which makes parameter estimation computationally expensive. We approach the problem using gradient-based optimization. Specifically, we apply the score-function gradient estimator to differentiate through discrete jump sampling in the Monte Carlo wave-function algorithm used to solve the Lindblad equation. The resulting stochastic gradient estimator exhibits sufficiently low variance and can still be estimated in an embarrassingly parallel manner, enabling efficient scaling of the simulations. We implement this gradient estimator in the existing open-source quarkonium suppression code QTraj. To demonstrate its utility for parameter estimation, we infer the two transport coefficients $\hatκ$ and $\hatγ$ using gradient-based optimization on synthetic nuclear modification factor data.

Collective motion in animal groups provide examples of emergent, decentralised coordination. Here, we examine a bottom-up model of collective behavior based on Future State Maximisation (FSM). In this model agents seek to maximise the diversity of their future visual states over a finite time horizon. We further assume that a subset of agents have a directional bias, e.g. towards different destinations. We observe swarm fragmentation on increasing (i) the strength of these preferences, or (ii) the difference in preferred directions, or (iii) the number of biased agents. Depending on these factors, biased agents can leave the swarm alone, leaving behind all other agents, or they can entrain some fraction of the group to leave with them. We further study the role of a classical nearest-neighbor alignment term on cohesion. Notably, we identify the existence of an finite, optimal coupling strength that suppresses fragmentation and maximises the flock cohesion. Our results demonstrate that FSM can be successfully combined with classical flocking rules, offering a flexible framework for modeling intelligent collective systems.

In this work, we explore the possibility of using quantum computers provided for usage in cloud by big companies (such as IBM, IonQ, IQM Quantum Computers, etc.) to run our quantum neural network (QNN) developed for data analysis in the context of LISA Space Mission, developed with the Qiskit library in Python. Our previous work demonstrated that our QNN learns patterns in gravitational wave (GW) data much faster than a classical neural network, making it suitable for fast GW signal detection in future LISA data streams. Analyzing the fees from hardware providers like IBM Quantum, Amazon Braket and Microsoft Azure, we found that the fees for running the first segment of our QNN sum up to \$2000, \$60000, and \$1000000 respectively. Using free plans, we succeed to run the 3-qubit feature map of the QNN for one random data sample on {\fontfamily{qcr} \selectfont ibm\_kyoto} and {\fontfamily{qcr}\selectfont IQM Quantum Computers\_Garnet} quantum computers, obtaining a fidelity of 99\%; we could also run the first prediction segment of our QNN on {\fontfamily{qcr} \selectfont ibm\_kyoto}, implemented for 4 qubits, and obtained a prediction accuracy of 20\%. We queried providers such as IBM Quantum, Amazon Braket, Pasqal, and Munich Quantum Valley to obtain access to their plans, but, with the exception of Amazon Braket, our applications remain unanswered to this day. Other major setbacks in using the quantum computers we had access to included Qiskit library version issues (as in the cases of IBM Quantum and IQM Quantum Computers) and the frequent unavailability of the devices, as was the case with the Microsoft Azure provider. All the results presented in this paper were accumulated in 2024.

Bacteria regulate their motility through a variety of mechanisms, including quorum sensing (QS) and other density-dependent responses mediated by diffusible signals. While nonlinear density-dependent motility is well known in active-matter theory to generate nonequilibrium spatial patterns, its consequences for the coexistence of growing, interacting species remain less explored. Here we develop a minimal spatially structured model for two strongly competing species in which local demographic interactions are coupled to an escape response: each species increases its motility nonlinearly (sigmoidal) with the local abundance of its competitor. We show that this sigmoidal motility regulation promotes optimal spatial self-organization and can sustain long term coexistence via segregation, even in parameter regimes that yield competitive exclusion in well-mixed Lotka-Volterra dynamics. On two-dimensional lattices, the interplay between demographic competition and density-dependent motility generates a range of emergent patterns, including regimes in which the weaker competitor counterintuitively has higher total abundance. Overall, our results identify nonlinear, competitor-induced motility as a fundamental mechanism capable of sustaining coexistence in competing microbial populations.

Spider dragline silk supercontracts as humidity increases, displaying large axial shortening together with a reproducible macroscopic twist. The physical origin of this torsion remains debated and is often attributed to helically arranged load-bearing elements, despite the lack of direct evidence for helicity in the native fiber. Here we show that torsion can arise generically from nonlinear anisotropic elasticity: humidity-driven shortening of the amorphous matrix, mechanically constrained by stiff, axially aligned $β$-sheet--rich load-bearing segments and their experimentally induced prestretch, drives the system into a dual Poynting regime in which axial shortening couples to spontaneous twist. Coupling a diffusion-based water-uptake law to irreversible matrix remodeling and fiber plasticity, the model quantitatively reproduces monotonic and cyclic torsional measurements using parameter values consistent with available experimental material parameters. These results identify supercontraction-induced torsion in spider silk as a manifestation of a dual Poynting effect and provide a minimal, physically grounded framework for humidity-driven torsional actuation in matrix--fiber architectures.

Improved local numerical solution for the ADER-DG numerical method with a local DG predictor for solving the initial value problem for a first-order ODE system is proposed. The improved local numerical solution demonstrates convergence orders of one higher than the convergence order of the local numerical solution of the original ADER-DG numerical method and has the property of continuity at grid nodes. Rigorous proofs of the approximation orders of the local numerical solution and the improved local numerical solution are presented. Obtaining the proposed improved local numerical solution does not require significant changes to the structure of the ADER-DG numerical method. Therefore, all conclusions regarding the convergence orders of the numerical solution at grid nodes, the resulting superconvergence, and the high stability of the ADER-DG numerical method remain unchanged. A wide range of applications of the ADER-DG numerical method is presented for solving specific initial value problems for ODE systems for a wide range of polynomial degrees. The obtained results provide strong confirmation for the developed rigorous theory. The improved local numerical solution is shown to exhibit both higher accuracy and improved smoothness and point-wise comparability. Empirical convergence orders of all individual numerical solutions were calculated for a wide range of error norms, which well agree with the expected convergence orders. The rigorous proof, based on the $ε$-embedding method, of the applicability of the ADER-DG numerical method with a local DG predictor to solving DAE systems is presents.

Heat-flux boundary conditions are challenging to implement efficiently in rarefied gas flow simulations because the wall-reflected gas temperature and density must be determined dynamically during the computation. This paper aims to tackle this problem within the general synthetic iterative scheme (GSIS), where the Boltzmann kinetic equation is solved deterministically in an outer loop and macroscopic synthetic equations are solved in an inner loop. To avoid kinetic-macroscopic boundary-flux mismatch and the resulting convergence bottlenecks, for the macroscopic boundary flux at every inner iteration, the incident increment is estimated using a Maxwellian distribution, and then the reflected contribution is obtained by boundary conditions consistent with those in the kinetic solver. In addition to retaining the fast-converging and asymptotic-preserving properties of GSIS, the proposed method significantly reduces the iterations required to determine the wall-reflected gas parameters. Numerical simulations of rarefied gas flows in and around a 3D nozzle, a 2D adiabatic cylinder, and a 2D annular heat-transfer configuration show good agreement with the direct simulation Monte Carlo method, while achieving substantial efficiency gains over conventional iterative schemes.

Accurate representations of the World Air Transportation Network (WAN) are fundamental inputs to models of global mobility, epidemic risk, and infrastructure planning. However, high-resolution, real-time data on the WAN are largely commercial and proprietary, therefore often inaccessible to the research community. Here we introduce a generative model of the WAN that treats air travel as a stochastic process within a maximum-entropy framework. The model uses airport-level passenger flows to probabilistically generate connections while preserving traffic volumes across geographic regions. The resulting reconstructed networks reproduce key structural properties of the WAN and enable simulations of dynamic spreading that closely match those obtained using the real network. Our approach provides a scalable, interpretable, and computationally efficient framework for forecasting and policy design in global mobility systems.

Phase transitions are the macroscopic manifestation of microscopic processes that drive a system towards a new state. The detailed evolution of these processes, particularly in abrupt phase transitions, are currently not fully understood. Here, we introduce a theoretical framework based on internal node dependencies within a single-layer lattice. Crucially, we demonstrate that the fundamental mechanism underlying abrupt transitions is nucleation propagation preceded by a slow cascading process which scales with the range of dependencies. Our findings show that the synergy between these two distinct stages is essential for the occurrence of an abrupt transition. The first stage of a slow cascading mechanism was recently observed experimentally in superconducting layered materials, where heat acts as the dependency links, for the limit of infinite dependency range. Our model thus generalizes the framework to include finite dependency ranges, revealing previously unobserved mechanisms that could be experimentally verified through controlling the range of thermal diffusion in the material. As a universal mechanism, our model provides a robust method to test nucleation-controlled phase transitions in multiple systems, providing a path to discover and understand microscopic mechanisms in phase transitions.

Generative artificial intelligence (AI) systems can now reliably solve many standard tasks used in introductory physics courses, producing correct equations, graphs, and explanations. While this capability is often framed as an opportunity for efficiency or personalization, it also poses a subtle ethical and educational risk: students may increasingly submit correct results without engaging in the epistemic practices that define learning physics.This challenge has recently been described as the "boiling frog problem" because we may not fully recognize how rapidly AI capabilities are advancing and fail to respond with commensurate urgency. In this article, we argue that the central challenge of AI in physics education is not cheating or tool selection, but instructional design. Drawing on research on self-regulated learning, cognitive load, multiple representations, and hybrid intelligence, we propose a practical framework for cognitively activated learning activities that structures student activities before, during, and after AI use. Using an example from an introductory kinematics laboratory, we show how AI can be integrated in ways that preserve prediction, interpretation, and evaluation as core learning activities. Rather than treating AI as an answer-generating tool, the framework positions AI as an epistemic partner whose contributions are deliberately bounded and reflected upon.

Building upon previous research, this paper further explores the topological foundations for maintaining universal rules within ultra-large-scale societies. It finds that in small-scale societies, absolute egalitarianism and the rule of law can be compatible through peer monitoring within a fully connected network. However, in ultra-large-scale societies, to maintain high-dimensional rules capable of protecting innovation and property rights, a complex hierarchical structure including "high-fragility" nodes must be constructed. Through quantitative analysis of power structures, this paper proves that a flattened, two-tier structure inevitably leads to the degradation of the rule of law. Only a social topology with sufficient hierarchical depth can escape the deathly trap of the Leviathan while expanding in scale, thereby sustaining the dynamic evolution of civilization.

Accurate and efficient rumor detection is critical for information governance, particularly in the context of the rapid spread of misinformation on social networks. Traditional rumor detection relied primarily on manual analysis. With the continuous advancement of technology, machine learning and deep learning approaches for rumor identification have gradually emerged and gained prominence. However, previous approaches often struggle to simultaneously capture both the sequential and the global structural relationships among topological nodes within a social network. To tackle this issue, we introduce a hybrid model for detecting rumors that integrates a Graph Convolutional Network (GCN) with a Transformer architecture, aiming to leverage the complementary strengths of structural and semantic feature extraction. Positional encoding helps preserve the sequential order of these nodes within the propagation structure. The use of Multi-head attention mechanisms enables the model to capture features across diverse representational subspaces, thereby enhancing both the richness and depth of text comprehension. This integration allows the framework to concurrently identify the key propagation network of rumors, the textual content, the long-range dependencies, and the sequence among propagation nodes. Experimental evaluations on publicly available datasets, including Twitter 15 and Twitter 16, demonstrate that our proposed fusion model significantly outperforms both standalone models and existing mainstream methods in terms of accuracy. These results validate the effectiveness and superiority of our approach for the rumor detection task.

This paper constructs a multilayer recursive game model to demonstrate that in a rule vacuum environment, hierarchical predatory structures inevitably collapse into a monolithic political strongman system due to the conflict between exponentially growing rent dissipation and the rigidity of bottom-level survival constraints. We propose that the rise of a monolithic political strongman is essentially an "algorithmic entropy reduction" achieved through forceful means by the system to counteract the "informational entropy increase" generated by multilayer agency. However, the order gained at the expense of social complexity results in the stagnation of social evolutionary functions.

The replica-exchange Monte-Carlo (RXMC) method is a powerful Markov-chain Monte-Carlo algorithm for sampling from multi-modal distributions, which are challenging for conventional methods. The sampling efficiency of the RXMC method depends highly on the selection of the temperatures, and finding optimal temperatures remains a challenge. In this study, we propose a refined online temperature selection method by extending the gradient-based optimization framework proposed previously. Building upon the existing temperature update approach, we introduce a reparameterization technique to strictly enforce physical constraints, such as the monotonic ordering of inverse temperatures, which were not explicitly addressed in the original formulation. The proposed method defines the variance of acceptance rates between adjacent replicas as a loss function, estimates its gradient using differential information from the sampling process, and optimizes the temperatures via gradient descent. We demonstrate the effectiveness of our method through experiments on benchmark spin systems, including the two-dimensional ferromagnetic Ising model, the two-dimensional ferromagnetic XY model, and the three-dimensional Edwards-Anderson model. Our results show that the method successfully achieves uniform acceptance rates and reduces round-trip times across the temperature space. Furthermore, our proposed method offers a significant advantage over recently proposed policy gradient method that require careful hyperparameter tuning, while simultaneously preventing the constraint violations that destabilize optimization.

In a brief reflection on the principles of human society, P. A. M. Dirac articulated a structural tension between two widely affirmed norms: that it is good and natural for parents to improve the prospects of their own children, and that justice requires that all children have equal opportunities in life. These principles, each compelling on its own, cannot be fully realized together. This paper reconstructs Dirac's dilemma, connects it to the dynamics of compounding advantage and inheritance, and situates it within the broader history of political philosophy, including the work of Rawls, Dworkin, Cohen, Brighouse and Swift, Nozick, Murphy and Nagel, and others. The paper argues that attempts to eliminate the resulting injustices entirely risk damaging the non--zero--sum structures that generate general prosperity, and defends a position of "managed inequality": a robust social floor and real mobility, combined with limits on extreme dynastic accumulation and an explicit acceptance of some residual, but constrained, inherited advantage.

Generative AI has emerged as a transformative paradigm in engineering design, enabling automated synthesis and reconstruction of complex 3D geometries while preserving feasibility and performance relevance. This paper introduces a domain-specific implicit generative framework for turbine blade geometry using DeepSDF, addressing critical gaps in performance-aware modeling and manufacturable design generation. The proposed method leverages a continuous signed distance function (SDF) representation to reconstruct and generate smooth, watertight geometries with quantified accuracy. It establishes an interpretable, near-Gaussian latent space that aligns with blade-relevant parameters, such as taper and chord ratios, enabling controlled exploration and unconditional synthesis through interpolation and Gaussian sampling. In addition, a compact neural network maps engineering descriptors, such as maximum directional strains, to latent codes, facilitating the generation of performance-informed geometry. The framework achieves high reconstruction fidelity, with surface distance errors concentrated within $1\%$ of the maximum blade dimension, and demonstrates robust generalization to unseen designs. By integrating constraints, objectives, and performance metrics, this approach advances beyond traditional 2D-guided or unconstrained 3D pipelines, offering a practical and interpretable solution for data-driven turbine blade modeling and concept generation.

Collisional and stochastic wave-particle dynamics in plasmas far from equilibrium are complex, temporally evolving, stochastic processes which are challenging to model. In this work, we extend previous methods coupling differentiable kinetic simulators and plasma phase space diagnostics to learn collision operators that account for time-varying background distributions. We also introduce a more general integro-differentiable operator formulation to probe relevant terms in the collision operator. To validate the proposed methodology we use data generated by self-consistent electromagnetic Particle-in-Cell simulations. We show that both approaches recover operators that can accurately reproduce the plasma phase space dynamics while being more accurate than estimates based on particle track statistics. These results further demonstrate the potential of using differentiable simulators to infer collision operators for scenarios where no closed form solution exists or deviations from existing theory are expected.

Optimal Causation Entropy (oCSE) is a robust causal network modeling technique that reveals causal networks from dynamical systems and coupled oscillators, distinguishing direct from indirect paths. CausationEntropy is a Python package that implements oCSE and several of its significant optimizations and methodological extensions. In this paper, we introduce the version 1.1 release of CausationEntropy, which includes new synthetic data generators, plotting tools, and several advanced information-theoretical causal network discovery algorithms with criteria for estimating Gaussian, k-nearest neighbors (kNN), geometric k-nearest neighbors (geometric-kNN), kernel density (KDE) and Poisson entropic estimators. The package is easy to install from the PyPi software repository, is thoroughly documented, supplemented with extensive code examples, and is modularly structured to support future additions. The entire codebase is released under the MIT license and is available on GitHub and through PyPi Repository. We expect this package to serve as a benchmark tool for causal discovery in complex dynamical systems.

We investigate the impact of magnetic-field-induced feedback on the dynamics of a Hindmarsh-Rose neuron model exhibiting a blue-sky catastrophe. By introducing a magnetic flux variable that couples nonlinearly to the membrane potential, we demonstrate that electromagnetic effects profoundly reshape neuronal firing patterns and bifurcation structure. Interspike-interval bifurcation analysis reveals a nonmonotonic dependence on the magnetic coupling strength, with weak coupling preserving regular spiking and bursting, intermediate coupling promoting chaotic bursting, and strong coupling yielding structured irregular dynamics. These transitions are quantitatively characterized using the largest Lyapunov exponent computed via the Wolf algorithm and supported by Poincaré sections and time-series analysis. Our results establish electromagnetic feedback as a robust and tunable mechanism for controlling instability and chaos in slow-fast neuronal systems.

Large Language Models (LLMs) are rapidly reshaping scientific research. We analyze these changes in multiple, large-scale datasets with 2.1M preprints, 28K peer review reports, and 246M online accesses to scientific documents. We find: 1) scientists adopting LLMs to draft manuscripts demonstrate a large increase in paper production, ranging from 23.7-89.3% depending on scientific field and author background, 2) LLM use has reversed the relationship between writing complexity and paper quality, leading to an influx of manuscripts that are linguistically complex but substantively underwhelming, and 3) LLM adopters access and cite more diverse prior work, including books and younger, less-cited documents. These findings highlight a stunning shift in scientific production that will likely require a change in how journals, funding agencies, and tenure committees evaluate scientific works.

Accurately predicting the Curie temperature ($T_c$) of ferroelectrics from first principles remains a major challenge, as theoretical estimates often fall significantly below experimental values. In this work, we investigate the origin of these discrepancies in the prototypical ferroelectric PbTiO$_3$ by performing extensive constant-pressure ab initio molecular dynamics (AIMD) simulations and benchmarking them against classical molecular dynamics (MD) using machine learning force fields (MLFFs) derived from first-principles data. Our results show that the underestimation of $T_c$ primarily stems from the limitations of the exchange-correlation functional, rather than inaccuracies in the MLFF fitting. We uncover a critical interplay between finite-size effects and the range of interatomic interactions: although short-range MLFFs appear to yield better agreement with experimental $T_c$, this improvement results from a fortuitous cancellation of errors. Incorporating explicit long-range interactions improves accuracy for larger supercells but ultimately leads to lower predicted $T_c$ values. These findings highlight that accurate finite-temperature predictions require not only high-quality training data and sufficiently large simulation cells, but also the explicit treatment of long-range interactions and improved exchange-correlation functionals.

The rise of social media and recommendation algorithms has sparked concerns about their role in fostering opinion polarization and echo chambers. We study these phenomena using an adaptive voter model to compare two connection mechanisms: "free" global rewiring, where individuals connect with anyone sharing their opinion, and "friend-of-a-friend" local rewiring, which mimics algorithmic link recommendations on platforms like Facebook or LinkedIn. Simulations across different network topologies reveal that local rewiring increases final-state polarization of the system and fragments social networks into many disconnected components. The usual phase transition into two disconnected components turns into a fragmentation of smaller components, leading to an increase in echo chambers as well as many isolated nodes. This effect is most pronounced in clustered networks with high homophily in rewiring, illustrating how recommendation algorithms can intensify social fragmentation by changing the very structure of the network.

Pyvale is an open-source software package that aims to become an all-in-one tool for sensor simulation, sensor uncertainty quantification, sensor placement optimization, and calibration/validation. Central to this is support for image-based sensors, with a dedicated Digital Image Correlation (DIC) module designed for both standalone use and integration within broader experimental design workflows. The design philosophy behind the DIC engine in Pyvale prioritizes a user-friendly Python interface with performant compiled code under the hood. This paper covers Pyvale's 2D DIC engine design, implementation, metrological performance compared to other DIC codes, and the unique ability to handle gigapixel size scanning electron microscope (SEM) images. Finally, we compare runtimes between Pyvale and other open-source DIC codes and show strong computational performance across a range of image resolutions and thread counts.

In parts of Himachal Pradesh (Kullu and Mandi) and the Western Himalaya, village deities (\emph{devtā}) are carried through the landscape on shoulder-borne palanquins or ``raths.'' Participants often describe these raths as agents: they \emph{choose} routes, signal assent or refusal, and sometimes ``move on their own'' as if people are not moving them but are instead being moved. This paper offers : (i) a mechanistic model in which a palanquin interacts with human carriers modeled as coupled limit-cycle oscillators, and (ii) a philosophical analysis of how music and gurus/oracular specialists (\emph{gūr}/``guru'' in local English) function as couplings that stabilize collective interpretation, producing what we call \emph{distributed agency}. On the physics side we build a six-degree-of-freedom rigid-body model with unilateral handle contacts, base excitation from walking, and a Kuramoto-Adler phase description for interpersonal coupling and musical entrainment. We prove standard phase-locking conditions (Adler-type capture range) and show how unilateral contact can rectify periodic forcing and inject harmonics, creating parameter regimes in which near-perfect synchrony produces large-amplitude roll. On the simulation side we report an ensemble study (30 seeds per condition) from an archived ``Palanquin Simulator'' package: a ``baseline'' condition produces small roll (\(\mathrm{RMS}\approx 0.15^{\circ}\)) and moderate synchrony, whereas a ``music'' entrainment condition produces near-unity synchrony (\(\approx 0.99\)) but also robust roll instability (\(\mathrm{RMS}\approx 18^{\circ}\)) and frequent contact loss. We do \emph{not} claim to have validated the model against field data; we treat the simulations as a \emph{proof of plausibility} and as a generator of falsifiable predictions.

Pigmentation abnormalities, ranging from hypo- to hyperpigmentation, can serve as biomarkers of developmental disruption in fish exposed to environmental contaminants. However, the mechanistic pathways underlying these alterations remain poorly understood. Studies have shown that pattern formation in fish development requires specific pigment cell interactions. Motivated by experimental observations of pigmentation alterations following contaminant exposure, we investigate how pollutants influence pigment cell self-organization using a continuum reaction-diffusion-advection framework. The model incorporates nonlocal Morse-type kernels to describe short- and long-range interactions among melanophores and xanthophores. Our results show that perturbations to the strengths of adhesion or repulsion can drive transitions between stripes, spots, and mixed patterns, reproducing phenotypes characteristic of fish pigmentation mutants. In particular, homotypic interactions are sensitive to contamination, leading to pronounced changes in melanophore density and resulting pigmentation patterns. Time-dependent simulations indicate that pigment changes from early short-term contaminant exposure may be recoverable, whereas prolonged exposure can lead to sustained pigment loss. In a growing fish, contaminant-induced changes in cell-cell interactions directly influence stripe formation rate, stripe number, and pigmentation levels. Overall, our study provides insight into the mechanistic link between experimentally observed pigmentation alterations and the changes in spatial patterns of adult fish.

We introduce a novel spatial discretization technique for the reliable and efficient simulation of magnetization dynamics governed by the Landau-Lifshitz (LL) equation. The overall discretization error is systematically decomposed into temporal and spatial components. The spatial error analysis is conducted by formulating the LL equation within the framework of the Localized Orthogonal Decomposition (LOD) method. Numerical examples are presented to validate the accuracy and approximation properties of the proposed scheme.

Scientific attention is unevenly distributed, creating inequities in recognition and distorting access to opportunities. Using citations as a proxy, we quantify disparities in attention by gender and institutional prestige. We find that women receive systematically fewer citations than men, and that attention is increasingly concentrated among authors from elite institutions -- patterns not fully explained by underrepresentation alone. To explain these dynamics, we introduce a model of citation network growth that incorporates homophily (tendency to cite similar authors), preferential attachment (favoring highly cited authors) and group size (underrepresentation). The model shows that disparities arise not only from group size imbalances but also from cumulative advantage amplifying biased citation preferences. Importantly, increasing representation alone is often insufficient to reduce disparities. Effective strategies should also include reducing homophily, amplifying the visibility of underrepresented groups, and supporting equitable integration of newcomers. Our findings highlight the challenges of mitigating inequities in asymmetric networks like citations, where recognition flows in one direction. By making visible the mechanisms through which attention is distributed, we contribute to efforts toward a more responsible web of science that is fairer, more transparent, and more inclusive, and that better sustains innovation and knowledge production.

Active arterial mechanics, governed by vascular smooth muscle contraction, are critical to physiological regulation, cardiovascular disease progression, and clinical diagnosis. Although various in vivo methods have been developed to assess arterial stiffness, most cannot distinguish the contribution of smooth muscle tone; therefore, quantitative characterization of arterial activity remains challenging. In this study, we developed a pressure-area analysis framework integrating ultrasound imaging, blood pressure measurement, neural network-based segmentation of arterial cross-sectional area, and biomechanical model-driven inversion to infer active mechanical properties. A total of 233 volunteers (aged 18-65 year) were recruited to acquire cross-sectional ultrasound videos of the right common carotid artery for training the neural network. The segmentation results demonstrate good spatial and temporal performance of the neural network. We further recruited 10 additional volunteers (aged 25 +/- 3 year) to perform a 1-minute step test, followed by pressure-area measurements over a 30-minute recovery period. Using the proposed approach, we quantified post-exercise changes in carotid arterial active mechanics relative to baseline (i.e., the resting state). Results showed that active mechanics remained elevated for approximately 15 minutes compared to baseline (p < 0.05), whereas systolic pressure differed significantly only within the first approximately 5 minutes post-exercise (p < 0.001). These results indicate a dissociation between blood pressure and smooth muscle recovery, which may offer new insight into vascular smooth muscle regulation during physiological stress.

Accurate determination of transition states remains central to understanding reaction kinetics. Double-ended methods like the Nudged Elastic Band (NEB) ensure relevant transition states and paths, but incur high computational costs and suffer stagnation on flat or rough potential energy surfaces. Conversely, single-ended eigenmode-following techniques offer efficiency but cannot often be constrained between specific states. Here, we present the Reorienting Off-path Nudged Elastic Bands (RONEB), an adaptive hybrid algorithm that integrates the double ended nature of the NEB with the acceleration of single ended Min-Mode Following methods. RONEB provides stability based on the history of the path optimization, relative force triggering, and an alignment-based back-off penalty to dynamically decouple the climbing image from the elastic band constraints. We benchmark the method against the standard Climbing Image NEB (CI-NEB) across the Baker-Chan transition state test set using the PET-MAD machine-learned potential and the OptBench Pt(111) heptamer island surface diffusion set. A Bayesian analysis of the performance data quantifies a median reduction in gradient calls of 46.3% [95% CrI: -54.7%, -36.9%] relative to the baseline, while surface diffusion tests reveal a 28% reduction across 59 metallic rearrangement mechanisms. These results establish RONEB as a highly effective tool for high-throughput automated chemical discovery.

Many cities promote walkability through concepts such as the compact city and 15-minute city to enhance urban livability, yet few methods link spatial walkability features to empirically measured livability and account for temporal dynamics. The method developed for this study uses mobile phone data from the Helsinki Metropolitan Area (Finland) to assess whether commonly used, literature-derived livability indicators (diversity, density, proximity, accessibility) predict observed human activity patterns across different times of day. We constructed two key dimensions of livability: attractiveness and walkability with quantifiable sub-indicators that were selected based on literature. Our analysis shows that walkability, and even more so the combined livability index, correlates with activity patterns, outperforming the pure attractiveness perspective. However, this relationship is temporally unstable, significantly weakening at night and fluctuating daily. Moreover, based on Geographically Weighted Regression analysis, our results reveal significant spatial variation in the relationship between livability and the intensity of human activities. The findings suggest that traditional urban planning goals, such as functional diversity to enhance walkability, contribute to livability but have a limited impact on the 15-minute city's overall sustainable mobility objectives, necessitating a larger-scale perspective and more functionally profiled approaches for urban development.

The field of plasma physics heavily relies on simulations to model various phenomena, such as instabilities, turbulence, and nonlinear behaviors that would otherwise be difficult to study from a purely theoretical approach. Simulations are fundamental in accurately setting up experiments, which can be extremely costly and complex. As high-fidelity tools, gyrokinetic simulations play a crucial role in discovering new physics, interpreting experimental results, and improving the design of next-generation devices. However, their high computational costs necessitate the use of acceleration platforms to reduce execution time. This work revolves around the TRIangular MEsh based Gyrokinetic (TRIMEG) code, which performs high-accuracy particle-in-cell plasma simulations in tokamak geometries, leveraging a novel finite element approach. The rise of graphical processing units (GPUs) constitutes an occasion to satisfy such computational needs, by offloading the most expensive portion of the code to the accelerators. The chosen approach features GPU offloading with the OpenMP API, which grants portability of the code to different architectures, namely AMD and NVIDIA. The particle pushing as well as the grid-to-particle operations have been ported to GPU platforms. Compiler limitations had to be overcome, and portions of the code were restructured to be suitable for GPU acceleration. Kernel performance was evaluated by carrying out GPU grid size exploration, as well as scalability studies. In addition, the efficiency of hybrid MPI-OpenMP offloading parallelization was assessed. The speedup of the GPU implementation was calculated by comparing it with the pure CPU version using different rationales. The Ion Temperature Gradient (ITG) mode was simulated using the GPU-accelerated version, and its correctness was verified in terms of the energy growth rate and the two-dimensional mode structures.

The fundamental understanding of how cells physically interact with each other and their environment is key to understanding their organisation in living tissues. Over the past decades several computational methods have been developed to decipher emergent multi-cellular behaviors. In particular agent-based (or cell-based) models that consider the individual cell as basic modeling unit tracked in space and time enjoy increasing interest across scientific communities. In this article we explore a particular class of cell-based models, so-called Deformable Cell Models (DCMs), that allow to simulate the biophysics of the cell with high realism. After situating this model among other model types, We give an overview of past and recent DCM developments and discuss new simulation results of several applications covering in-vitro and in-vivo systems. Our goal is to demonstrate how such models can generate quantitative added value in biological and biotechnological problems.

Tree balance has received considerable attention in recent years, both in phylogenetics and in other areas. Numerous (im)balance indices have been proposed to quantify the (im)balance of rooted trees. A recent comprehensive survey summarized this literature and showed that many existing indices are based on similar underlying principles. To unify these approaches, three general metaconcepts were introduced, providing a framework to classify, analyze, and extend imbalance indices. In this context, a metaconcept is a function $Φ_f$ that depends on another function $f$ capturing some aspect of tree shape. In this manuscript, we extend this line of research by introducing a new metaconcept based on the heights of the pending subtrees of all inner vertices. We provide a thorough analysis of this metaconcept and use it to answer open questions concerning the well-known $B_1$ balance index. In particular, we characterize the tree shapes that maximize the $B_1$ index in two cases: (i) arbitrary rooted trees and (ii) binary rooted trees. For both cases, we also determine the corresponding maximum values of the index. Finally, while the $B_1$ index is induced by a so-called third-order metaconcept, we explicitly introduce three new (im)balance indices derived from the first- and second-order height metaconcepts, respectively, thereby demonstrating that pending subtree heights give rise to a variety of novel (im)balance indices.

Understanding the population dynamics of ecological systems is crucial for predicting shifts in biodiversity and ensuring the pro- tection of these systems. Established models often focus on pairwise species interactions, yet recent studies have highlighted the importance of higher-order interactions (HOIs) in shaping community structure and function. In this study, we investigate the effects of HOIs in an agent-based model with three species engaged in intransitive competition. We introduce an HOI where one species modifies the competition between the other two. We explore the impact of the strength, radius of influence, and speed of this interaction modification on species abundances and oscillations thereof. Our results show that these abundances are not only greatly impacted by the strength, but also by the radius and speed of the interaction modification. A deeper investigation demonstrates that the changes in the oscillations are caused by the interaction modification itself, and not the change in pairwise interaction strength caused by the HOI. These results emphasize the importance of considering the spatio-temporal scales of higher-order interactions when assessing ecosystem stability, highlighting that such interactions can introduce complex dynamical behaviors that go beyond the predictions of traditional pairwise or simpler higher-order models

Cortical gamma ($γ$)-band activity reflects local excitation-inhibition (E/I) balance. In schizophrenia (SCZ), reduced task-evoked gamma suggests altered E/I dynamics, but it is unclear whether differences stem from input properties or systematic shifts in E/I operating point and gain. We coupled a cochlear-inspired speech front end to a Wilson-Cowan E/I model to simulate gamma responses across three conditions: Healthy, SCZ-speech, and SCZ-semantics. Metrics included event-related spectral perturbation (ERSP$_γ$) and threshold-time fraction ($γ%$). A stable hierarchy emerged: Healthy(speech/semantics) $>$ SCZ(speech) $>$ SCZ(semantics), robust under equal-energy control and gain perturbations. Network dynamics coincided with single-node solutions, supporting interpretability. Pharmacological analogs showed bidirectional effects: reduced inhibition lowered $γ$, while reduced excitation increased $γ$, with no self-sustained oscillations. Findings indicate SCZ gamma deficits align more with shifts in E/I operating point and gain than input differences. This pipeline provides a testable, reusable mechanistic framework for speech-evoked gamma and a baseline for cross-population studies.

This paper investigates the classification capability of small-scale spiking neural networks based on the Leaky Integrate-and-Fire (LIF) neuron model. We analyze the relationship between classification accuracy and three factors: the number of neurons, the number of stimulus nodes, and the number of classification categories. Notably, we employ a large language model (LLM) to assist in discovering the underlying functional relationships among these variables, and compare its performance against traditional methods such as linear and polynomial fitting. Experimental results show that classification accuracy follows a power-law scaling primarily with the number of categories, while the effects of neuron count and stimulus nodes are relatively minor. A key advantage of the LLM-based approach is its ability to propose plausible functional forms beyond pre-defined equation templates, often leading to more concise or accurate mathematical descriptions of the observed scaling laws. This finding has important implications for understanding efficient computation in biological neural systems and for pioneering new paradigms in AI-aided scientific discovery.

In the wake of the SARS-CoV-2 pandemic, there has been heightened interest from applied mathematicians in infectious disease modelling. Modelling efforts often focus on predicting whether diseases are likely to be eliminated or, instead, (re-)emerge, especially as a result of control measures.This tipping point between elimination and infection waves has been successfully anticipated in the literature through the use of early warning signals and such signals often rely on the theory of critical slowing down. Recent developments have shown that these signals (increases in fluctuation variance and return time) can emerge from the system geometry in the case of non-normal dynamics rather than a change in asymptotic stability. We show how such dynamical behaviour occurs in the fluctuations from the mean-field in general stochastic systems. Using the susceptible-infectious-recovered model as an example application, we analyse how critical-like behaviour can be exploited to anticipate infection waves in the absence of an equilibrium bifurcation.

Digital traces of everyday behavior, such as e-commerce (EC) purchase logs, provide scalable signals for population-level monitoring, yet their epidemiological validity remains unclear due to weak links to clinical outcomes. We propose a behavioral proxy for disease onset based on transitions from regular to therapeutic diets observed in EC purchase histories, and evaluate its validity through large-scale cross-domain analysis. Using EC purchase data (N = 55,645 users) and independent insurance-derived clinical records, we compare ingredient-level risk patterns and seasonal disease dynamics in feline lower urinary tract disease (FLUTD) as a case study. The proxy-based estimates show strong agreement with clinical data, with correlations of r = 0.74 for ingredient-level risk patterns and r = 0.82 for seasonal variation. Both data sources consistently capture elevated disease risk during winter months. Moreover, analysis using EC data alone reproduces established domain knowledge, including the association between higher wet food consumption and lower disease risk. Our results demonstrate that behavioral signals derived from large-scale EC data can serve as validated, cost-effective complements to traditional surveillance systems, and suggest broader applicability to monitoring lifestyle-related and chronic conditions.

Constraint-based metabolic modeling is the predominant framework for simulating cellular metabolism. The central assumption of these models is that metabolism operates at a steady state, meaning that the production and consumption rates of each metabolite are balanced. This assumption imposes linear constraints on the fluxes of biochemical reactions. Flux Balance Analysis (FBA), a fundamental method in the field, is formulated as an optimization problem maximizing a cellular objective (e.g., growth) over the resulting linear subspace of steady state fluxes. Many other methods in the field are expressed either as a modification to FBA, or use FBA as a black box within an algorithm. Here, we propose a simple and general alternative to optimization that, for any flux vector, finds the closest flux distribution within the steady-state subspace. This operation corresponds to an orthogonal projection that enforces the steady-state assumption. We further introduce extensions to handle cases involving unknown or fixed fluxes through modified projections and tailored affine subspaces. The overall approach is computationally efficient, does not require a cellular objective, and is easy to implement. We validate our method and its variants on both synthetic and experimental datasets, demonstrating their speed and utility for denoising and imputing metabolic flux data, and for predicting steady-state fluxes from more readily available types of data. Code availability: The code implementing FBApro is available at https://github.com/Singh-Lab/FBApro. All code required to reproduce the figures in the paper is available, although the data used must be sourced separately. The repository also contains toy models and examples.

Probabilistic mental simulation is thought to play a key role in human reasoning, planning, and prediction, yet the demands of simulation in complex environments exceed realistic human capacity limits. A theory with growing evidence is that people simulate using simplified representations of the environment that abstract away from irrelevant details, but it is unclear how people determine these simplifications efficiently. Here, we present a "Just-in-Time" framework for simulation-based reasoning that demonstrates how such representations can be constructed online with minimal added computation. The model uses a tight interleaving of simulation, visual search, and representation modification, with the current simulation guiding where to look and visual search flagging objects that should be encoded for subsequent simulation. Despite only ever encoding a small subset of objects, the model makes high-utility predictions. We find strong empirical support for this account over alternative models in a grid-world planning task and a physical reasoning task across a range of behavioral measures. Together, these results offer a concrete algorithmic account of how people construct reduced representations to support efficient mental simulation.

The maintenance of diversity in complex ecological communities despite unpredictable dynamics and competitive exclusion is thought to require continual influx of new species or competitive advantages that accrue as species become rare. We examine isolated planktonic microbial communities under permanent ice cover in Antarctic lakes, recording prokaryotic abundance across 9 communities, 11 years, 30~m of depth, and thousands of species in the McMurdo LTER. We quantify rare species advantage by modeling community dynamics under frequency-dependent selection. We find persistent diversity and pervasive negative frequency dependence with limited immigration and turnover. While ecology and evolutionary sciences have long debated whether diversity is maintained selectively, we measure selection over a $10^4$-fold range of abundance in naturally coevolving communities and implicate rare species advantage.

This study presents a robust methodology for the 3D volumetric reconstruction of native Chileanpollen grains, specifically Gevuina avellana (hazel),Conium maculatum (hemloc) and Anthemis cotula (chamomile). Using a lens-free Digital In-line Holographic Microscopy (DLHM) system, we capture complex interference patterns that are numerically reconstructed using the Kirchhoff-Helmholtz transform. Our results demonstrate that this label-free approach provides high-fidelity morphological characterization and nanometric precision in biophysical parameter extraction, offering a scalable alternative for automated melissopalynology and environmental monitoring.

Mooney images are high-contrast, two-tone visual stimuli, created by thresholding photographic images. They allow researchers to separate image content from image understanding, making them valuable for studying visual perception. An ideal Mooney image for this purpose achieves a specific balance: it initially appears unrecognizable but becomes fully interpretable to the observer after seeing the original template. Researchers traditionally created these stimuli manually using subjective criteria, which is labor-intensive and can introduce inconsistencies across studies. Automated generation techniques now offer an alternative to this manual approach. Here, we present MooneyMaker, an open-source Python package that automates the generation of ambiguous Mooney images using several complementary approaches. Users can choose between various generation techniques that range from approaches based on image statistics to deep learning models. These models strategically alter edge information to increase initial ambiguity. The package lets users create two-tone images with multiple methods and directly compare the results visually. In an experiment, we validate MooneyMaker by generating Mooney images using different techniques and assess their recognizability for human observers before and after disambiguating them by presenting the template images. Our results reveal that techniques with lower initial recognizability are associated with higher post-template recognition (i.e. a larger disambiguation effect). To help vision scientists build effective databases of Mooney stimuli, we provide practical guidelines for technique selection. By standardizing the generation process, MooneyMaker supports more consistent and reproducible visual perception research.

Accurate, low-latency estimates of the instantaneous phase of oscillations are essential for closed-loop sensing and actuation, including (but not limited to) phase-locked neurostimulation and other real-time applications. The endpoint-corrected Hilbert transform (ecHT) reduces boundary artefacts of the Hilbert transform by applying a causal narrow-band filter to the analytic spectrum. This improves the phase estimate at the most recent sample. Despite its widespread empirical use, the systematic endpoint distortions of ecHT have lacked a principled, closed-form analysis. In this study, we derive the ecHT endpoint operator analytically and demonstrate that its output can be decomposed into a desired positive-frequency term (a deterministic complex gain that induces a calibratable amplitude/phase bias) and a residual leakage term setting an irreducible variance floor. This yields (i) an explicit characterisation and bounds for endpoint phase/amplitude error, (ii) a mean-squared-error-optimal scalar calibration (c-ecHT), and (iii) practical design rules relating window length, bandwidth/order, and centre-frequency mismatch to residual bias via an endpoint group delay. The resulting calibrated ecHT achieves near-zero mean phase error and remains computationally compatible with real-time pipelines. Code and analyses are provided at https://github.com/eosmers/cecHT.

Aortic valve opening (AO) events are crucial for detecting frequency and rhythm disorders, especially in real-world settings where seismocardiography (SCG) signals collected via consumer smartphones are subject to noise, motion artifacts, and variability caused by device heterogeneity. In this work, we present a robust deep-learning framework for SCG segmentation and rhythm analysis using accelerometer recordings obtained with consumer smartphones. We develop an enhanced U-Net v3 architecture that integrates multi-scale convolutions, residual connections, and attention gates, enabling reliable segmentation of noisy SCG signals. A dedicated post-processing pipeline converts probability masks into precise AO timestamps, whereas a novel adaptive 3D-to-1D projection method ensures robustness to arbitrary smartphone orientation. Experimental results demonstrate that the proposed method achieves consistently high accuracy and robustness across various device types and unsupervised data-collection conditions. Our approach enables practical, low-cost, and automated cardiac-rhythm monitoring using everyday mobile devices, paving the way for scalable, field-deployable cardiovascular assessment and future multimodal diagnostic systems.

Decoding visual semantic representations from human brain activity is a significant challenge. While recent zero-shot decoding approaches have improved performance by leveraging aligned image-text datasets, they overlook a fundamental aspect of human cognition: semantic understanding is inherently anchored in the auditory modality of speech, not text. To address this, our study introduces the first comparative framework for evaluating auditory versus textual semantic modalities in zero-shot visual neural decoding. We propose a novel brain-visual-auditory multimodal alignment model that directly utilizes auditory representations to encapsulate semantics, serving as a substitute for traditional textual descriptors. Our experimental results demonstrate that the auditory modality not only surpasses the textual modality in decoding accuracy but also achieves higher computational efficiency. These findings indicate that auditory semantic representations are more closely aligned with neural activity patterns during visual processing. This work reveals the critical and previously underestimated role of auditory semantics in decoding visual cognition and provides new insights for developing brain-computer interfaces that are more congruent with natural human cognitive mechanisms.

During infectious disease epidemics, pathogen transmission occurs in host populations made up of interacting subpopulations. Using stochastic simulation and analytical approximations, we examine how outbreak sizes in networked populations depend on network architecture, subpopulation sizes and the strength of coupling between subpopulations. We find, as expected, that mean outbreak sizes are frequently lower in networked populations than in homogeneous populations with the same basic reproduction number. However, after an outbreak ends, a networked population is often vulnerable to further outbreaks, and the ending of an outbreak may not imply herd immunity in any sense. Another key finding is that a relatively small amount of randomly distributed prior immunity can be more protective in a networked population than a homogeneous population, a phenomenon which can be reproduced analytically in certain cases. We also find that in networked populations, randomly distributed prior immunity is often more protective than infection-acquired immunity; but this conclusion can be reversed in populations with highly variable susceptibility. All of these conclusions have implications for designing outbreak control strategies that aim to reduce pathogen transmission during epidemics.

Hepatitis A severity increases sharply with age, while Korea is experiencing a cohort shift in which low seroprevalence adult cohorts are aging into older, higher fatality age groups. This demographic and immunological transition creates an urgent policy question regarding how adult vaccination should be prioritized under resource constraints. We evaluated three adult vaccination scenarios targeting low seroprevalence age groups (S1) 20 to 39 years, (S2) 40 to 59 years, and (S3) 20 to 59 years. Using an age structured dynamic transmission model calibrated to Korean data, we derived dynamically feasible vaccination allocation trajectories under realistic capacity constraints using an optimal control framework and linked these trajectories to long term transmission model simulations. We conducted DALY based cost effectiveness analyses over a lifetime horizon from both healthcare system and societal perspectives, and characterized uncertainty using probabilistic sensitivity analysis (PSA) and cost effectiveness acceptability curves (CEACs). Robustness was examined using one way sensitivity analyses. In the base case, S2 consistently yields the most favorable and robust cost effectiveness profile under both perspectives, with the lowest ICER. S3 achieved the largest reduction in DALYs but requires substantially higher incremental costs, resulting in a higher ICER than S2. S1 produces the smallest DALY reduction and is the least efficient strategy. PSA and CEACs confirm that S2 remains the preferred option across most willingness to pay ranges. S2 offers the most balanced and robustly cost effective strategy in Korea, capturing substantial mortality reduction while limiting additional program costs. S3 may be justified when higher budgets or willingness to pay thresholds are acceptable, but S2 provides the clearest value for money under epidemiological and economic conditions.

Understanding how the brain represents the multifaceted properties of words in context is essential for explaining the neural architecture of human language. Here, we combine large-scale psycholinguistic modeling with naturalistic fMRI to uncover the latent structure of word properties and their neural representations during narrative comprehension. By analyzing 106 psycholinguistic variables across 13,850 English words, we identified eight interpretable latent dimensions spanning lexical usage, word form, phonology orthography mapping, sublexical regularity, and semantic organization. These factors robustly predicted behavioral performance across lexical decision, naming, recognition, and semantic judgment tasks, demonstrating their cognitive relevance. Parcel-based and multivariate fMRI analyses of narrative listening revealed that these latent dimensions are encoded in overlapping yet functionally differentiated cortical systems. Multidimensional scaling and hierarchical clustering analyses further identified four interacting subsystems supporting sensorimotor grounding, controlled semantic retrieval, resolution of lexical competition, and contextual episodic integration. Together, these findings provide a unified neurocognitive framework linking fundamental lexical psycholinguistic dimensions to distributed cortical systems engaged during naturalistic language comprehension.

Many biochemical measurements define state spaces that are bounded, absorbing, and physically irreversible, yet are routinely analysed using linear and Euclidean frameworks that assume global invertibility, symmetry, and translation invariance. This mismatch can irretrievably obscure biological structure, independent of data quality, scale, or preprocessing. This work formalises the structure of bounded biochemical state spaces using cysteine redox regulation as a representative example and identify the minimal algebraic properties required for categorically correct representation. Hard boundaries, absorbing states, and irreversible ensemble dynamics render linear algebra incompatible with these objects. This work demonstrates that tropical algebra provides a natural realisation of the required properties by replacing additive linear structure with order-based, piecewise-linear operations that encode dominance, saturation, and path dependence without contradiction. By making non-invertibility and absorption explicit rather than implicit, this framework resolves a fundamental algebraic mismatch and establishes a principled foundation for the representation and analysis of bounded biochemical data.

Across neuroscience, artificial intelligence, and related fields, dominant models of intelligence typically privilege convergence: uncertainty is reduced, competing explanations are eliminated, and behaviour is governed by the optimisation of a single objective or policy. While this framing has proved powerful in many settings, it sits uneasily with biological and adaptive systems that maintain redundancy, ambiguity, and parallel explanatory processes over extended timescales. Here we propose an alternative perspective, termed polyphonic intelligence, in which coherent behaviour and meaning emerge from the coordination of multiple semi-independent inferential processes operating under shared constraints. Rather than resolving plurality through dominance or collapse, polyphonic systems sustain multiple explanatory trajectories and integrate them through soft alignment, compatibility relations, and bounded influence. We develop this perspective conceptually and formally, introducing a variational framework in which multiple coordinated approximations are maintained without winner-takes-all selection. This formulation makes explicit how plurality can remain stable, tractable, and productive, and clarifies how polyphonic inference differs from ensemble methods, mixture models, and Bayesian model averaging. Through proof-of-principle examples, we demonstrate that non-dominating, pluralistic inference can be implemented in simple computational systems without requiring centralised control or global convergence. We conclude by discussing implications for neuroscience, psychiatry, and artificial intelligence, and by arguing that intelligence may be more fruitfully understood as coordination without command rather than as the elimination of uncertainty.

Biological neural networks self-organize according to local synaptic modifications to produce stable computations. How modifications at the synaptic level give rise to such computations at the network level remains an open question. Pehlevan et al. [Neur. Comp. 27 (2015), 1461--1495] proposed a model of a self-organizing neural network with Hebbian and anti-Hebbian synaptic updates that implements an algorithm for principal subspace analysis; however, global stability of the nonlinear synaptic dynamics has not been established. Here, for the case that the feedforward and recurrent weights evolve at the same timescale, we prove global stability of the continuum limit of the synaptic dynamics and show that the dynamics evolve in two phases. In the first phase, the synaptic weights converge to an invariant manifold where the `neural filters' are orthonormal. In the second phase, the synaptic dynamics follow the gradient flow of a non-convex potential function whose minima correspond to neural filters that span the principal subspace of the input data.

Evolutionary accumulation models (EvAMs) are an emerging class of machine learning methods designed to infer the evolutionary pathways by which features are acquired. Applications include cancer evolution (accumulation of mutations), anti-microbial resistance (accumulation of drug resistances), genome evolution (organelle gene transfers), and more diverse themes in biology and beyond. Following these themes, many EvAMs assume that features are gained irreversibly -- no loss of features can occur. Reversible approaches do exist but are often computationally (much) more demanding and statistically less stable. Our goal here is to explore whether useful information about evolutionary dynamics which are in reality reversible can be obtained from modelling approaches with an assumption of irreversibility. We identify, and use simulation studies to quantify, errors involved in neglecting reversible dynamics, and show the situations in which approximate results from tractable models can be informative and reliable. In particular, EvAM inferences about the relative orderings of acquisitions, and the core dynamic structure of evolutionary pathways, are robust to reversibility in many cases, while estimations of uncertainty and feature interactions are more error-prone.

Sickle cell disease causes erythrocytes to become sickle-shaped, affecting their movement in the bloodstream and reducing oxygen delivery. It has a high global prevalence and places a significant burden on healthcare systems, especially in resource-limited regions. Automated classification of sickle cells in blood images is crucial, allowing the specialist to reduce the effort required and avoid errors when quantifying the deformed cells and assessing the severity of a crisis. Recent studies have proposed various erythrocyte representation and classification methods. Since classification depends solely on cell shape, a suitable approach models erythrocytes as closed planar curves in shape space. This approach employs elastic distances between shapes, which are invariant under rotations, translations, scaling, and reparameterizations, ensuring consistent distance measurements regardless of the curves' position, starting point, or traversal speed. While previous methods exploiting shape space distances had achieved high accuracy, we refined the model by considering the geometric characteristics of healthy and sickled erythrocytes. Our method proposes (1) to employ a fixed parameterization based on the major axis of each cell to compute distances and (2) to align each cell with two templates using this parameterization before computing distances. Aligning shapes to templates before distance computation, a concept successfully applied in areas such as molecular dynamics, and using a fixed parameterization, instead of minimizing distances across all possible parameterizations, simplifies calculations. This strategy achieves 96.03\% accuracy rate in both supervised classification and unsupervised clustering. Our method ensures efficient erythrocyte classification, maintaining or improving accuracy over shape space models while significantly reducing computational costs.

The use of multi-centric analyses is crucial for obtaining sufficient sample sizes and representative clinical populations in experimental studies. In this setting, data harmonization techniques are typically employed to address systematic biases and ensure the interoperability of the data. State-of-the-art harmonisation approaches are based on the statistical theory of random effect modeling, allowing to account for either linear of non-linear biases and batch effects. However, optimizing these statistical methods generally requires data centralization at some point during the analysis pipeline, therefore introducing the risk of exposing individual patient information while posing significant data governance issues. To overcome this challenge, in this paper we present Fed-ComBat, a federated framework for batch effect harmonization on decentralized data. Fed-ComBat enables the preservation of nonlinear covariate effects without requiring centralization of data and without prior parametric hypothesis on the variables to account for. We demonstrate the effectiveness of Fed-ComBat against a comprehensive panel of existing approaches based on the state-of-the-art ComBat, along with distributed and nonlinear variants. Our experiments are based on extensive simulated data, and on the analysis of multiple cohorts based on 7 neuroimaging studies comprising healthy controls (CI) and subjects with various disorders such as Parkinson's disease (PD), Alzheimer's disease (AD), and autism spectrum disorder (ASD). Our results show that in a federated settings, Fed-ComBat harmonization exhibits comparable results to centralized methods for both linear and nonlinear cases. On real data, harmonized trajectories of the thickness ofthe right hippocampus across lifespan measured on a set of 7 public studies show comparable results between centralized and federated models and are consistent with the literature when using a nonlinear model. The code is publicly available at: https://gitlab.inria.fr/greguig/fedcombat

Background: Being born very or extreme preterm is a major source of cerebral anomalies and neurodevelopmental disorders, whose occurrence depends on many perinatal factors. A better understanding of these factors could be provided by cerebral Magnetic Resonance Imaging (MRI) at term-equivalent age (TEA). Objective: To investigate, through cerebral TEA-MRIs, the relationship between the main perinatal factors, the occurrence of cerebral anomalies, and cerebral volumetry. Methods: We assembled a cohort of preterm babies born before 32 weeks of gestation who underwent a cerebral TEA-MRI. We assessed cerebral anomalies using a radiological scoring system -- the Kidokoro scoring -- and performed cerebral volumetry. We investigated the relationships between 9 clinical factors (birth characteristics, resuscitation treatments{\ldots}), Kidokoro scores, and brain volumes. Results: Among 110 preterms who underwent a cerebral MRI at TEA, only 6% suffered moderate-to-severe brain anomalies. We identified mechanical ventilation as a risk factor for cerebral anomalies (adjusted Odds-Ratio aOR = 4.6, 95% Confidence Interval CI [1.7-12.8]) and prolonged parenteral nutrition as a protective factor for white matter anomalies (aOR = 0.2, 95%CI [0.1-0.8]). Mechanical ventilation (p = 0.01) and being born small for gestational age (p < 0.001) were risk factors for the reduction of cerebral volumes. An increase in brain lesion severity was associated with decreased cerebral volumes (p = 0.016). Conclusion: Our study highlights the importance of treatment-related perinatal factors on the occurrence of cerebral anomalies in very and extreme preterms, and the interest in using both qualitative (Kidokoro scoring) and quantitative (volumetry) MRI-tools.

Cognitive processes are realized across an extraordinary range of natural, artificial, and hybrid systems, yet there is no unified framework for comparing their forms, limits, and unrealized possibilities. Here, we propose a cognition space approach that replaces narrow, substrate-dependent definitions with a comparative representation based on organizational and informational dimensions. Within this framework, cognition is treated as a graded capacity to sense, process, and act upon information, allowing systems as diverse as cells, brains, artificial agents, and human-AI collectives to be analyzed within a common conceptual landscape. We introduce and examine three cognition spaces -- basal aneural, neural, and human-AI hybrid -- and show that their occupation is highly uneven, with clusters of realized systems separated by large unoccupied regions. We argue that these voids are not accidental but reflect evolutionary contingencies, physical constraints, and design limitations. By focusing on the structure of cognition spaces rather than on categorical definitions, this approach clarifies the diversity of existing cognitive systems and highlights hybrid cognition as a promising frontier for exploring novel forms of complexity beyond those produced by biological evolution.

Many biological systems exhibit multiscale dynamics, where some species occur in high copy numbers while others remain rare. This heterogeneity necessitates hybrid modelling approaches: deterministic models are computationally efficient but inaccurate for low-count species, while fully stochastic simulations are accurate but prohibitively expensive. Hybrid methods like the Jump-Switch-Flow (JSF) algorithm address this by simulating low-count species stochastically and high-count species deterministically. However, selecting regime-switching thresholds to control errors for specific observables remains an open challenge. We develop a principled framework for threshold selection targeting extinction probability. We formalise JSF as a piecewise-deterministic Markov process and derive backward equations for extinction under exact and hybrid dynamics. Near extinction boundaries, complex nonlinear dynamics reduce to tractable time-inhomogeneous linear birth-death processes. This structure yields a rigorous error decomposition based on early and late excursions. Isolating the dominant error term motivates a fast, actionable heuristic. We demonstrate via Monte Carlo studies on a stochastic Lotka-Volterra model that our heuristic reliably upper-bounds empirical errors in extinction probability. This enables users to select the smallest threshold that satisfies a target error tolerance. This work paves the way for principled, efficient multiscale modelling and simulation in stochastic biological systems.

Many molecular properties depend on 3D geometry, where non-covalent interactions, stereochemical effects, and medium- to long-range forces are determined by spatial distances and angles that cannot be uniquely captured by a 2D bond graph. Yet most 3D molecular graph neural networks still rely on globally fixed neighborhood heuristics, typically defined by distance cutoffs and maximum neighbor limits, to define local message-passing neighborhoods, leading to rigid, data-agnostic interaction budgets. We propose Multiscale Interaction Mixture of Experts (MI-MoE) to adapt interaction modeling across geometric regimes. Our contributions are threefold: (1) we introduce a distance-cutoff expert ensemble that explicitly captures short-, mid-, and long-range interactions without committing to a single cutoff; (2) we design a topological gating encoder that routes inputs to experts using filtration-based descriptors, including persistent homology features, summarizing how connectivity evolves across radii; and (3) we show that MI-MoE is a plug-in module that consistently improves multiple strong 3D molecular backbones across diverse molecular and polymer property prediction benchmark datasets, covering both regression and classification tasks. These results highlight topology-aware multiscale routing as an effective principle for 3D molecular graph learning.

Understanding how the brain integrates motor suppression with motivational processes remains a fundamental question in neuroscience. The rostral Pedunculopontine nucleus, a brainstem structure involved in motor control, has been shown to induce transient motor arrest upon optogenetic or electrical stimulation. However, our current understanding of its potential role in linking motor suppression with motivational or reinforcement-related processes is still insufficient. To further explore the effects induced by PPN stimulations and infer the potential mechanism underlying its role involved in both motor and emotional regulation, we developed a fully automated, low-cost system combining real-time animal tracking with closed-loop optogenetic stimulation, using the OpenMV Cam H7 Plus and embedded neural network models. The system autonomously detects the rat's position and triggers optical stimulation upon entry into a predefined region of interest, enabling unbiased, unsupervised behavioral assays. Optogenetic activation of CaMKIIa-expressing neurons in the rostral PPN reliably induced transient motor arrest. When motor arrest was spatially paired with a defined region of interest, rats developed a robust place preference after limited training. These results suggest that rostral PPN activation can couple motor inhibition with reinforcement-related behavioral circuitry. Together, our work provides both a technical framework for scalable closed-loop neuroscience experiments and preliminary evidence that the rostral PPN may participate in coordinating motor suppression with motivational processes.

Quantum theory is widely regarded as fundamentally indeterministic, yet classical frameworks can also exhibit indeterminism once infinite information is abandoned. At the same time, relativity is usually taken to forbid superluminal signalling, yet Lorentz symmetry formally admits superluminal transformations (SpTs). Dragan and Ekert have argued that SpTs entail indeterminism analogous to the quantum one. Here, we derive a no-go theorem from natural assumptions, which can be interpreted as: superluminal transformations (SpTs) and finite information cannot coexist. Any theory accommodating SpTs must therefore allow unbounded information content, leading to a deterministic ontology akin to that of classical theories formulated over the real numbers. Thus, any apparent indeterminism arising from superluminal transformations reflects only probabilities arising from subjective ignorance, unlike the objective nature of probabilities in quantum theory, indicating that the claimed indeterminacy from superluminal extensions is not quantum.

We unveil a temperature bias of the probe in transient quantum thermometry under Markovian dynamics. Specifically, for qubit thermometers evolving under Markovian dynamics, we show that enhanced precision beyond the steady state limit can be achieved if and only if the probe is initially colder than the thermal state corresponding to the bath temperature to be estimated. In contrast, this temperature bias can be lifted when the probe dynamics is non-Markovian. In the non-Markovian regime, both hot and cold probes can simultaneously attain the same transient maximum precision, well above the steady-state value.

Recent cosmological data favour phantom-crossing dark energy, motivating models with non-minimal couplings that induce a fifth force on structure formation. Reconciling these models with local tests often requires strong screening, leading to environment-dependent clustering. We investigate such effects via a late-time structure-induced phase transition driven by a non-minimally coupled scalar field. For this purpose, we introduce norns, a fully relativistic cosmological particle-mesh code that self-consistently evolves a complex scalar field - a generalisation of the symmetron producing global U(1) strings rather than domain walls. Using simulations, we compare string and wall-forming models, quantifying impacts on the matter power spectrum, halo mass function, and defect dynamics. Strong environment-dependent effects can generate significant departures from LCDM in underdense regions while keeping the overall power spectrum changes modest (~ 4-15% at k~0.3-0.5 h Mpc^-1, sub-percent for z > 0.2). We find that an attractive fifth force can locally suppress structure growth in voids while enhancing it in surrounding overdense regions by driving outflows from the voids. These effects leave distinctive signatures in the matter density probability density function and in marked halo power spectra, which are likely detectable in low-redshift data.

We construct a natural and nonperturbative momentum space for quantum field theory on $(d+1)$-dimensional de Sitter (dS) spacetime in the Poincaré slicing, adapted to early Universe cosmology. In particular, we identify the dS frequency as the unitary-representation label of the dS isometry group $\mathrm{SO}(1, d+1)$. By diagonalizing the quadratic Casimir together with spatial translations, we provide a harmonic expansion of operators in what we call the Kontorovitch-Lebedev-Fourier (KLF) space. This momentum space shares many structural properties with its Minkowski counterpart, for instance: equations of motion reduce to algebraic equations, and the quadratic dynamics provides a simple propagator analogous to flat space. We reformulate the perturbative computation of in-in correlators in KLF momentum space, showing from first principles how time integrals turn into frequency-space integrals over meromorphic functions. We show how our construction streamlines computations, naturally accommodates the contributions from principal and complementary series in the Källén-Lehmann spectral decomposition of composite operators, and leads to an inversion formula for the spectral density.

We derive exact general solutions (as opposed to attractor particular solutions) for the evolution of a scalar field $φ$ in a universe dominated by a background fluid with equation of state parameter $w_B = -1$, extending earlier work on exact solutions with $w_B > -1$. Straightfoward exact solutions exist when the evolution is described by a linear differential equation, corresponding to constant, linear, and quadratic potentials. In the nonlinear case, exact solutions are derived for $V = V_0\ln φ$, $V = V_0 φ^{1/2}$ and $V = V_0/φ$, and the logarithmic potential also yields an exact first integral. These complicated parametric solutions are considerably less useful than those derived previously for a universe dominated by a barotropic fluid such as matter or radiation with $w_B > -1$. However, we generalize the slow-roll approximation and show that it applies to all sufficiently flat potentials in the case of a vacuum-dominated expansion, while it never applies when the universe is dominated by a background fluid with $w_B > -1$.

Massive spin-2 fields in addition to the standard massless graviton arise naturally in extensions of General Relativity, such as massive bigravity or models with extra dimensions. This work explores the observational signatures of these fields on the propagation of gravitational waves. Adopting a phenomenological framework consistent with such theories, we derive an analytical transfer function in the ultrarelativistic limit and establish detectability bounds. Finally, we provide forecasts for the accessible parameter space using current and future gravitational wave detectors.

Gravitational waves from extreme mass-ratio inspirals (EMRIs) will enable sub-percent measurements of massive black hole parameters and provide access to the demographics of compact objects in galactic nuclei. During the LISA mission, multiple EMRIs are expected to be detected, allowing statistical studies of massive black hole populations and their formation pathways. We perform hierarchical Bayesian inference on simulated EMRI catalogues to assess how well LISA could constrain the astrophysical population using parametrised population models. We test our inference framework on a variety of populations, including heterogeneous and homogeneous mixtures of parametrised subpopulations, and scenarios in which the assumed model is deliberately misspecified. Our results show that population parameters governing distributions with sharp features can be tightly constrained. Mixed populations can be disentangled with as few as $\sim20$ detections, and even with model misspecification, the inference retains sensitivity to key population features. These results demonstrate the capabilities and limitations of EMRI population inference, providing guidance for constructing realistic astrophysical population models for LISA analysis.

Efficient generation of radiation in the mid- and far- infrared relies primarily on lasers and coherent nonlinear optical phenomena driven by lasers. This wavelength range lacks of luminescent devices because the spontaneous emission rate becomes much longer than the nonradiative energy relaxation processes and therefore emitters have to count on stimulated emission produced by linear or non-linear optical gain. However, spontaneous emission is not a fundamental property of the emitter. By engineering metamaterials composed of arrays of nano-emitters into microcavities coupled to patch antennas, we have demonstrated mid-infrared electroluminescent devices emitting a collimated beam with excellent spatial properties and a factor 100 increase in the collected power, compared to standard devices. Our results illustrate that by reshaping the photonic environment around emitting dipoles, as in the Purcell effect, it is possible to enhance the spontaneous emission and conceive efficient optoelectronic light emitting devices that operate close to the thermodynamical equilibrium as LEDs in the visible range.

Leveraging a recently proposed notion of relative entropy in general probabilistic theories (GPT), we prove a finite de Finetti representation theorem for general convex bodies. We apply this result to address a fundamental question in polynomial optimization: the existence of a convergent outer hierarchy for problems with inequality constraints and analytical convergence guarantees. Our strategy generalizes a quantitative monogamy-of-entanglement argument from quantum theory to arbitrary convex bodies, establishing a uniform upper bound on mutual information in multipartite extensions. This leads to a finite de Finetti theorem and, subsequently, a convergent conic hierarchy for a wide class of polynomial optimization problems subject to both equality and inequality constraints. We further provide a constructive rounding scheme that yields certified interior points with controlled approximation error. As an application, we express the optimal GPT value of a two-player non-local game as a polynomial optimization problem, allowing our techniques to produce approximation schemes with finite convergence guarantees.

Recently, Jordan et al. (Nature, 2025) introduced a novel quantum-algorithmic technique called Decoded Quantum Interferometry (DQI) for solving specific combinatorial optimization problems associated with classical codes. They presented a constraint-satisfaction problem called Optimal Polynomial Intersection (OPI) and showed that, for this problem, a DQI algorithm running in polynomial time can satisfy a larger fraction of constraints than any known polynomial-time classical algorithm. In this work, we propose several improvements to the DQI algorithm, including sidestepping the quadratic-time Dicke state preparation. Given random access to the input, we show how these improvements result in a nearly linear-time DQI algorithm for the OPI problem. Concurrently and independently with this work, Khattar et al. (arXiv:2510:10967) also construct a nearly linear-time DQI algorithm for OPI using slightly different techniques.

We analyze the WAHUHA Floquet protocol recently applied to arrays of Rydberg atoms and derive beyond-leading-order corrections in the high-frequency expansion of the effective spin theory. We find that an appropriate choice of the pulses times can enforce an approximate symmetry corresponding to the conservation of the total magnetization. The interaction channels emerging from higher-order Floquet terms affect three-body bound states (\emph{trimers}), which gain a significant mobility. We estimate the corresponding enhancement in 1D spin chains and conclude that their dynamics is within experimental reach. Detrimental effects due to the proliferation of particles outside of the trimer magnetization sector are found to occur and spread on time-scales slower than the trimer propagation. Moreover, these can be suppressed in higher dimensional lattices, e.g. in 2D triangular lattices, as the lattice geometry brings these processes off resonance. Our results establish a concrete route to realizing mobile multiparticle bound states in Floquet-engineered Rydberg platforms.

In an entanglement summoning task, a set of distributed, co-operating parties attempt to fulfill requests to prepare entanglement between distant locations. The parties share limited communication resources: timing constraints may require the entangled state be prepared before some pairs of distant parties can communicate, and a restricted set of links in a quantum network may further constrain communication. Building on earlier work, we continue the characterization of entanglement summoning. We give an if and only if condition on entanglement summoning tasks with only bidirected causal connections, and provide a set of sufficient conditions addressing the most general case containing both oriented and bidirected causal connections. Our results rely on the recent development of entanglement sharing schemes.

Superconducting qubit networks (SQNs) embedded in a low-dissipative resonator is a promising device allowing one not only to establish the collective quantum dynamics on a macroscopic scale but also to greatly enhance the sensitivity of detectors of microwave photons. A quantum ac Stark effect provided by coupling between an SQN and microwave photons of a resonator, leads to a strong nonlinear interaction between photons. Here, we present a two-tone spectroscopy experiment in which a set of 10 superconducting flux qubits is coupled to the input R- resonator and the output T- transmission line. An external microwave pump field close to the resonance frequency populates macroscopically the resonator mode as a Bose-Einstein condensate, while a second probe beam scans the resonances referred also as Bogoliubov-like excitations. The corresponding excitation frequency measured from the transmission coefficient, |S21(f)| displays an abrupt change of the resonant dip position once the power of the pump field overcomes a critical value Pcr. This sharp shift occurs in a narrow region of pump frequencies, and can be tuned by an applied magnetic field. It is a signature of bistability of the photon number inside the resonator, in agreement with theory.

We report on the experimental observation of stimulated cooling in the non-equilibrium Bose-Einstein condensate (BEC) of weakly interacting exciton-polaritons from approximately room temperature down to 20K. By resolving the condensate in energy-momentum space and performing interferometric measurements, we distinguish the condensate from thermalized particles yet occupying excited states macroscopically. In contrast to the analytical quantum theories of non-equilibrium BEC [Shishkov et al., Phys. Rev. Lett. 128, 065301 (2022)], we observe segmentation of the particle density along the excited states into two fractions both following Bose-Einstein distribution, albeit with different effective temperatures and chemical potentials. Our results indicate that the temperature of the weakly interacting Bose gas is universally set by the density-dependent chemical potential, revealing a defining property of non-equilibrium BECs. Finally, we demonstrate that the stimulated nature of the cooling process directly governs the emergence of quantum coherence of the condensate and shapes the dissipative properties of the excited states.

Partial differential equations (PDEs) form the backbone of simulations of many natural phenom- ena, for example in climate modeling, material science, and even financial markets. The application of physics-informed neural networks to accelerate the solution of PDEs is promising, but not compet- itive with numerical solvers yet. Here, we show how quantum computing can improve the ability of physics-informed neural networks to solve partial differential equations. For this, we develop hybrid networks consisting of quantum circuits combined with classical layers and systematically test them on various non linear PDEs and boundary conditions in comparison with purely classical networks. We demonstrate that the advantage of using quantum networks lies in their ability to achieve an accurate approximation of the solution in substantially fewer training epochs, particularly for more complex problems. These findings provide the basis for targeted developments of hybrid quantum neural networks with the goal to significantly accelerate numerical modeling.

Measurement-based quantum computation (MBQC) offers a promising paradigm for photonic quantum computing, but its implementation requires the generation of specific non-Gaussian resource states. While continuous-variable encodings such as the highly complex (GKP) states have been widely studied, the much simpler binomial codes offer an experimentally accessible alternative, though they demand a distinct set of operational tools. Here, we present a toolkit for MBQC using optical binomial codes, detailing a cavity-QED protocol for conditional generation of cluster states and the implementation of Pauli measurements. Our work proposes the first steps for existing optical atom-cavity architectures to lay the groundwork for their use in quantum computation.

We propose a Floquet-engineered optical lattice clock based demodulation scheme to enhance the low-frequency performance of wound fiber-optic vibration sensors. Vibration-induced phase variations in the sensing fiber are demodulated by the Floquet-engineered Rabi spectra of the clock transition. The lattice depth with the fiber length and the Floquet-engineered Rabi spectra under the vibration from 200 Hz down to 0.5 Hz are simulated. With a fiber length of 4 km and transmission loss of 2 dB/km, a phase change sensitivity higher than 6 * 10^3 rad per g is achieved at both vibration frequencies of 200 Hz and 0.5 Hz.

We investigate the geometry of a tidally deformed, rotating black hole and timelike geodesics in its vicinity. Our framework provides a local picture of the structural evolution of a relativistic restricted three-body problem around a deformed black hole in an adiabatically evolving binary, motivated by various astrophysical settings including disk dynamics and extreme mass-ratio inspirals. As the tidal-field strength is increased, initially regular, bound geodesics undergo four stages: (i) weak chaos emerges within the bound motion; (ii) a subset of trajectories plunges into the black hole; (iii) a fraction of the remaining trajectories becomes unbound; and (iv) no bound trajectories persist. We provide semi-analytic estimates for the critical tidal amplitudes associated with each transition. Our estimates indicate that, within the frequency band of ground-based gravitational-wave detectors, the matter flow around black holes may already be depleted, whereas LISA and (B-)DECIGO could probe the earlier stages. Our results suggest that an object orbiting a tidally deformed massive BH may remain near resonances over a wide range of separations, indicating an accumulated, non-negligible impact on the gravitational-wave phase. Tidal perturbations can also introduce nonlinear couplings among epicyclic oscillations of geodesics, offering a potential avenue to resonant excitation of quasi-periodic oscillations in X-ray light curves from accreting black holes.

The space $\mathrm{Inv}(j_1,j_2,j_3,j_4)$ of SU(2)-invariant four-valent tensors, also known as intertwiners, can be understood as the quantum states of a tetrahedron in Euclidean space with fixed areas. In loop quantum gravity, they are states of the smallest "atom of space" with non-zero volume. At the same time they correspond to four-party tensor product states invariant under global rotations. We consider the multipartite entanglement of states in $\mathrm{Inv}(j_1,j_2,j_3,j_4)$ using the recently proposed entropic fill. Numerically evaluating entropic fill in the case of equal spins between $1/2$ and $11$, we find that the distributions of entanglement are very different for intertwiners as compared to generic tensors, and for coherent intertwiners as compared to generic ones. The peak in the distribution seems to be at the highest entanglement for generic intertwiners and at the lowest for generic tensors, but in terms of average entanglement, the roles are switched: average entanglement is highest in arbitrary tensors and lower in intertwiners, at least in the regime of large $j$. We also find that entanglement depends on the geometric data of coherent intertwiners in a complicated way.

Negating the Measurement Independence assumption (MI) is often referred to as the `third way' to account for the experimental violation of Bell's inequality. However, this route is generally viewed as ludicrously contrived, implying some implausible conspiracy where experimenters are denied the freedom to choose measurement settings as they like. Here, a locally realistic model of quantum physics is developed (Rational Mechanics - RaQM - based on a gravitational discretisation of Hilbert Space) which violates MI without denying free will. Crucially, RaQM distinguishes experimenters' ability to freely choose measurement settings to some nominal accuracy, from an inability to choose exact settings, which were never under their control anyway. In RaQM, Hilbert states are necessarily undefined in bases where squared amplitudes and/or complex phases are irrational numbers. Such `irrational' bases correspond to conceivable but necessarily impossible counterfactual measurements, and are shown to play a ubiquitous role in the analysis of both single- and entangled-particle quantum physics. It is concluded that violation of Bell inequalities can be understood with none of the strange processes historically associated with it. Instead, using concepts from (non-classical) $p$-adic number theory, we relate RaQM to Bohm and Hiley's concept of a holistic Machian-like Undivided Universe. If this interpretation of Bell's Theorem is correct, building more and more energetic particle accelerators to probe smaller and smaller scales, in the search for a theory which synthesises quantum and gravitational physics and hence a Theory of Everything, may be a fruitless exercise.

We experimentally demonstrate the simultaneous estimation of the three parameters characterizing a pair of incoherent optical sources in the sub-Rayleigh regime, enabling super-resolved scene characterization. Using spatial-mode demultiplexing (SPADE) with two demultiplexers--one deliberately shifted--we determine separations well below the diffraction limit and achieve sensitive joint estimation of separation, centroid, and relative brightness over a broad range of scene configurations in a single experimental setting. We benchmark our performance using Fisher-information-based Cramér-Rao bounds, and discuss the corresponding quantum limits. We investigate two complementary scenarios: a realistic case with slightly non-identical sources, and an idealized case of indistinguishable sources.

Nonlocal light-matter interactions with giant atoms in high-dimensional environments are not only fundamentally intriguing for testing quantum electrodynamics beyond the dipole approximation but also crucial for building high-dimensional quantum networks and engineering multipartite entangled states. Given the enigmatic and largely uncharted collective signatures exhibited by multiple giant atoms within two-dimensional optical baths, we delve into their nonperturbative collective dynamics within the single-excitation subspace, focusing on the case where they are coupled to a common two-dimensional photonic reservoir and employing a resolvent operator approach. We demonstrate that precisely engineered atomic arrangements lead to unconventional quantum dynamics, featuring non-Markovianity-induced beats and long-lived bound states in the continuum, thereby providing a versatile platform for implementing two-dimensional quantum memory. Phenomenologically, we observe the emergence of exotic photon emission patterns in both two- and three-dimensional (3D) baths. The emission directions are shown to be precisely controllable on demand through exact phase engineering of the coupling parameters, enabling a highly efficient chiral light-matter interface. Moreover, our generalization to a 3D bath reveals that coherent dipole-dipole interactions can survive despite the coupling to a continuum of modes, a finding that challenges conventional wisdom regarding decoherence.

The recent detection of a common-spectrum stochastic signal by multiple pulsar timing array (PTA) collaborations has provided tentative evidence for a nanohertz (nHz) stochastic gravitational-wave background (SGWB). This signal can be widely interpreted as originating from a cosmic population of inspiraling supermassive black hole binaries (SMBHBs). Current PTA analyses primarily constrain the SGWB power spectrum and its auto-angular power spectrum. However, the supermassive black holes will produce an underlying correlation with the large-scale structure of the Universe, which can help in understanding the formation and evolution of the binaries. In this work, we develop a new analysis pipeline PyGxGW-PTA for studying the cross-correlation of nHz GW signal with galaxy surveys ($C^{\rm g\, GW}_\ell$) and obtain the first constraint on the SGWB and galaxy distribution cross-correlation using the NANOGrav 15-year dataset in combination with the DESI galaxy catalog. We find no statistically significant correlation between the SGWB and the large-scale distribution of DESI galaxies and using an optimal estimator we put an upper bound on $C^{\rm g\, GW}_{\ell=8} < 0.0083$ at $95\%$ C.I. This yields the first observational upper limit on the spatial correlation between the nHz SGWB and the large-scale structure of the Universe, establishing the observational groundwork for future multi-tracer analyses that will combine PTA data with next-generation galaxy surveys to unveil the SMBHB-galaxy correlation.

We analyse the performance of different error handling methods in the quantum data centre paradigm of distributed quantum computing. We compare the impact of quantum error detection, using the three-qubit repetition code and the [[4, 1, 2]] Leung-Nielsen-Chuang-Yamamoto code, on remote gates with that of conventional entanglement distillation techniques. Detailed classical simulation is used to obtain results for realistic near-term hardware.

Robust quantum routing is essential for scalable quantum technologies. This paper investigates the resilience of routing protocols in network architectures designed for perfect, high-fidelity transfer of both classical and quantum information under ideal conditions. We encode information in the position of a quantum walker on a graph, modelling the routing of a generic qubit state from a single input to multiple (orthogonal) outputs. We analyse and assess routing performance in various regimes, evaluating their robustness against static and dynamical noise.

Understanding the thermal behavior of quantum many-body pure states is one of the most fundamental issues in quantum thermodynamics. It is widely known that typical pure states yield vanishing work, just as thermal states do, when one restricts to local operations that cannot access correlations among subsystems. However, it remains unclear whether this equivalence to thermal states persists under LOCC (local operations and classical communication), where classically accessible correlations can be exploited for work extraction. In this work, we establish criteria for determining whether many-body pure states remain equivalent to thermal states even under LOCC, and show that this thermal equivalence is governed by their multipartite quantum correlation structure. We show that states with asymptotically maximal multipartite entanglement, such as Haar-random states, cannot yield extensive work under LOCC, whereas some states with limited multipartite entanglement, such as constant-degree graph states, allow extensive work extraction despite being locally indistinguishable from thermal states. Thus, our work provides a refined operational notion of thermal equivalence beyond the traditional local regime, which is becoming increasingly important due to the recent expansion of experimentally accessible operations.

In this analysis, we investigate the polarization radiation imaging of Kerr-Newman black holes, with a particular focus on the impact of black hole charge on photon propagation and polarization characteristics. By extending the traditional Walker-Penrose method, which is limited by its reliance on specific symmetric structures and Killing tensors, we overcome these limitations by constructing an ordinary differential equations (ODEs) numerical framework that combines the photon orbit equation with the polarization parallel transport equation. This allows for the self-consistent evolution of photon trajectories and polarization states in any spacetime backgrounds without relying on specific symmetries. Using this framework, we analyze the effects of black hole spin and charge on the polarization characteristics of radiation from both prograde and retrograde accretion disks. Our results show that black hole charge can significantly modify photon trajectories and polarization patterns: increasing charge compresses and distorts the EVPA structure on photon-ring scales, inducing localized rotations and asymmetries that may provide a potential diagnostic of a nonzero black hole charge.

This thesis develops a unified framework that reconstructs the full classical content of General Relativity from the classical limit of quantum scattering amplitudes. By interpreting the analytic structure of amplitudes as the field-theoretic imprint of spacetime geometry, the work establishes a direct correspondence between quantum processes and classical gravitational observables such as metrics, deflection angles, and multipole moments. Starting from the effective-field-theory description of gravity, the thesis shows that loop amplitudes encode not only quantum corrections but also the nonlinear classical self-interaction of the gravitational field, enabling the systematic derivation of the post-Minkowskian expansion of gravitational quantities by rewriting the Einstein equations in terms of graviton scattering processes. Building upon this foundation, the framework is applied to rotating and charged sources in arbitrary spacetime dimensions. A momentum-space formulation of the energy-momentum tensor is then developed, introducing gravitational form factors and source multipoles that link, for the first time, the internal matter distribution to the external multipolar field in a completely relativistic framework. Furthermore, the thesis completes the transition from the microscopic amplitude picture to the macroscopic description of gravitational sources by engineering a multipole-based framework for black hole mimickers, then applied to build horizon-less compact objects mimicking the multipolar structure of Kerr black holes. Finally, exploiting the Kerr-Schild gauge, the Fourier transforms of rotating black hole metrics are computed in closed form, bridging perturbative and non-perturbative descriptions of gravity, and allowing to probe the multipolar structure of higher-dimensional solutions employing scattering amplitudes.

In this paper, we develop a blended dynamics framework for open quantum networks with diffusive couplings. The network consists of qubits interconnected through Hamiltonian couplings, environmental dissipation, and consensus-like diffusive interactions. Such networks commonly arise in spontaneous emission processes and non-Hermitian quantum computing, and their evolution follows a Lindblad master equation. Blended dynamics theory is well established in the classical setting as a tool for analyzing emergent behaviors in heterogeneous networks with diffusive couplings. Its key insight is to blend the local dynamics rather than the trajectories of individual nodes. Perturbation analysis then shows that, under sufficiently strong coupling, all node trajectories tend to stay close to those of the blended system over time. We first show that this theory extends naturally to the reduced-state dynamics of quantum networks, revealing classical-like clustering phenomena in which qubits converge to a shared equilibrium or a common trajectory determined by the quantum blended reduced-state dynamics. We then extend the analysis to qubit coherent states using quantum Laplacians and induced graphs, proving orbit attraction of the network density operator toward the quantum blended coherent dynamics, establishing the emergence of intrinsically quantum and dynamically clustering behaviors. Finally, numerical examples validate the theoretical results.

Sensing networks underpin applications from fundamental physics to real-world engineering. Recently, distributed quantum sensing (DQS) has been investigated to boost the sensing performance, yet current schemes typically rely on entangled probes that are fragile to noise and difficult to scale. Here, we propose a DQS protocol that incorporates a causal-order switch into a cyclic network, enabling a single probe to sequentially query N independent sensors in a coherent superposition or a probabilistic mixture of opposite causal orders. By exploiting the noncommutativity between propagation and sensing processes, our scheme achieves a 1/N^2-scaling precision limit without involving entangled probes. Importantly, our approach utilizes a classical mixture of causal orders rather than a quantum switch, making it more feasible for practical realization. We experimentally implement this scheme for distributed beam tilts sensing in a free-space quantum optical network comprising up to 9 sensors, achieving picoradian-scale precision in estimating tilt angle. Our results demonstrate a robust and scalable DQS protocol that surpasses the conventional 1/N Heisenberg scaling in precision, advancing the practical deployment of quantum sensing networks.

We examine the role of angular momentum in general relativity from both heuristic and fully covariant perspectives, with the aim of clarifying conceptual ambiguities that arise when Newtonian intuition is extrapolated into the relativistic regime. Focusing on free--fall dynamics in the Schwarzschild and Kerr spacetimes in the test--particle limit, we employ an effective--potential heuristic approach to isolate the roles of the specific energy $E$, specific angular momentum $L$, and black--hole spin $a$. Within this framework, we identify well--defined regions of parameter space in which the Kerr spacetime leads to stronger or weaker local radial infall than the Schwarzschild case at the same radius. By analysing the kinematics of infalling geodesic congruences, we show how these local regimes combine along complete trajectories to either enhance or reduce gravitational focusing. We then interpret these results within a covariant 1+3 description of general relativity, in terms of the expansion, shear and Raychaudhuri evolution of timelike congruences. We demonstrate that black--hole rotation systematically modifies the shear of infalling irrotational flows, even when the magnitude of the local expansion is reduced, and that this shear modulation governs the overall rate of focusing. Our work complements previous studies of relativistic infall by providing a unified energetic and geometric interpretation of how angular momentum and rotation can strengthen or weaken gravitational collapse relative to the non--rotating case.

Quantum superposition is often phrased as the ability to add state vectors. In practice, however, the physical quantity is a ray (a rank-one projector), so each input specifies only a projector and leaves a gauge freedom in the phases of its vector representatives. This becomes a real operational barrier when one asks for a device that, given two independently prepared unknown pure states, outputs a coherent state proportional to a prescribed linear combination. We identify the missing ingredient as not probabilistic but phase-like. One needs a physical scenario that fixes a single phase convention on the relevant set of rays, so that the overlaps become well defined complex numbers. Thus, we formalize this through phase conventions and a single notion -- dubbed as "overlap-determinability." Our main theorem gives an exact equivalence: A nonzero completely positive trace-nonincreasing map that probabilistically produces superposition on a domain exists if and only if that domain is overlap-determinable. This unifies modern no-superposition results and characterizes the exceptional yes-go protocols, which succeed precisely when side information supplies the required missing resource. We then show that granting universal access to such convention-fixed overlaps destabilizes the familiar foundational and computational constraints. It enables forbidden transformations akin to quantum cloning and yields super-luminal signaling. It would also permit reflections about unknown states, leading to exponentially fast overlap amplification and a collapse of Grover's search lower bound to a logarithmic query complexity.

We demonstrate that symmetry need not be restored in the Pati-Salam model, so that $SU(2)_L\otimes SU(2)_R\otimes SU(4)_c$ remains broken to the Standard Model group at temperatures above the Pati-Salam symmetry breaking scale. Including the leading finite-temperature corrections, suitable quartic couplings prevent a restoration transition, thereby avoiding the thermal production of 't Hooft-Polyakov monopoles after inflation, even if the reheating temperature is very high. This removes monopole-based constraints on the Pati-Salam symmetry breaking scale.

We investigate the detectability of post-Newtonian higher harmonics from Galactic double white dwarfs in the centihertz band ($\sim 0.01$ Hz). Using a synthetic population, we show that, unlike the quadrupole mode, higher harmonics remain undetectable with LISA except for rare nearby systems. In contrast, planned mid-band (decihertz) observatories such as DECIGO and BBO will be able to detect the third harmonic for about 10\% of inspiral binaries above $\sim 5$ mHz, enabling statistical constraints on mass ratios. These results highlight the successive roles of LISA and future decihertz missions in establishing a coherent strategy for space-based gravitational-wave astronomy.

The temporal degree-of-freedom of light is incredibly powerful for modern quantum technologies, enabling large-scale quantum computing architectures and record key-rates in quantum key distribution. However, the generalized measurement of large and complex quantum superpositions of the time-of-arrival of a photon remains a unique experimental challenge. Conventional methods based on unbalanced Franson-type interferometers scale poorly with dimension, requiring multiple cascaded devices and active phase stabilization. In addition, these are limited by construction to a restricted set of phase-only superposition measurements. Here we show how the coupling of spatial and temporal information inside a single multi-mode fiber can be harnessed to program completely generalized measurements for high-dimensional superpositions of photonic time-bin. Using the multi-spectral transmission matrix of the fiber, we find special sets of spatial modes that experience distinct dispersive delays through the fiber. By exciting coherent superpositions of these spatial modes, we engineer the equivalent of large, unbalanced multi-mode interferometers inside the fiber and use them to perform high-quality measurements of arbitrary time-bin superpositions in up to dimension 11. The single fiber functions as a scalable, common-path interferometer for time-bin qudits that significantly eases the experimental overheads of standard approaches based on unbalanced Franson-type interferometers, serving as an essential tool for quantum technologies that harness the temporal properties of light.

Integrated quantum photonics holds significant promise for scalable photonic quantum information processing, quantum repeaters, and quantum networks, but its development is hindered by the mismatch between materials hosting high-quality quantum emitters and those compatible with mature photonic technologies. Heterogeneous integration offers a potential solution to this challenge, yet practical implementations have been limited by inevitable insertion losses at material interfaces. Here, we present a self-aligned heterogeneous quantum photonic integration approach that can deterministically achieve near-unity coupling efficiency at the interface. To showcase our approach, we demonstrate Purcell enhancement of a silicon vacancy (SiV) center in diamond induced by a heterogeneous photonic crystal cavity defined by titanium dioxide (TiO2), as well as optical spin control and readout via a TiO2 photonic circuit. We further show that, when combined with inverse photonic design, our approach enables efficient and broadband collection of single photons from a color center into a heterogeneous waveguide. Our approach is not restricted to SiV centers or TiO2; it can be broadly applied to integrate diverse solid-state quantum emitters with thin-film photonic devices where conformal deposition is possible. Together, these results establish a practical route to scalable quantum photonic integrated circuits that combine high-quality quantum emitters with technologically mature photonic platforms.

An increase in the number of control lines between the quantum processors and the external electronics constitutes a major bottleneck in the realization of large-scale quantum computers. Frequency-division multiplexing is expected to enable multiple qubits to be controlled through a single microwave cable; however, interference from off-resonant microwave tones hinders precise qubit control. Here, we propose an active interference suppression method for frequency-division-multiplexed simultaneous gate operations. We demonstrate that deliberate incorporation of off-resonant microwave tones improves the accuracy of single-qubit gates. Specifically, we find that by incorporating off-resonant orthogonal or quasi-orthogonal microwave tones, the gate infidelity decreases proportionally to the inverse square of the number of microwave tones. Furthermore, we show that fast oscillations neglected under the rotating wave approximation degrade gate fidelity, and that this degradation can be mitigated through optimized frequency allocation. Our approach is simple yet effective for improving the performance of frequency-division-multiplexed quantum gates.

We formulate a deterministic algorithm for preparing a general $U(1)$-eigenstate of a spin-$s$ chain of length $n$. These states consist of linear combinations of computational basis states $|\vec{m}\rangle$ of $n$ qudits, each with $(2s+1)$ levels and $s= 1/2, 1, 3/2, \ldots$, whose ditstrings $\vec{m}$ have a fixed digit sum. Exploiting a Gray code for bounded integer compositions, whose consecutive ditstrings obey the Gray property, the quantum state is prepared by applying corresponding ``Gray gates.'' We use this algorithm to prepare exact eigenstates of integrable spin-$s$ XXX Hamiltonians.

Theoretical models predict that the Galaxy hosts $10^8$-$10^9$ black holes formed from the complete gravitational collapse of heavy stars and that most of these black holes are isolated, without any companion. Within 15 pc of the Solar System ($\sim 50$ ly), there may be a few black holes. If located inside one of the Local Interstellar Clouds - which occupy 5-20% of this local volume - an isolated black hole could produce detectable electromagnetic emission via accretion from the interstellar medium, given the capabilities of current or near-future observatories. However, precise predictions remain challenging due to large uncertainties in the expected accretion spectra. Outside these clouds, the accretion rate would be too low in any standard model to yield a detectable electromagnetic signal. While astrometric detection via gravitational perturbation of nearby stars is conceivable, the local stellar density is too low for this method to be realistically successful. We have searched the Gaia DR3 catalog for candidate isolated black holes accreting from the interstellar medium and identified five sources. All candidates lie close to the Galactic plane, making them likely spurious astrometric solutions, for instance caused by unmodelled background sources (crowding) and/or unmodelled binarity; nevertheless, they cannot be definitively ruled out without follow-up observations.

Super-resolution (SR) seeks to reconstruct high-resolution (HR) data from low-resolution (LR) observations. Classical deep learning methods have advanced SR substantially, but require increasingly deeper networks, large datasets, and heavy computation to capture fine-grained correlations. In this work, we present the \emph{first study} to investigate quantum circuits for SR. We propose a framework based on Variational Quantum Circuits (VQCs) with \emph{Adaptive Non-Local Observable} (ANO) measurements. Unlike conventional VQCs with fixed Pauli readouts, ANO introduces trainable multi-qubit Hermitian observables, allowing the measurement process to adapt during training. This design leverages the high-dimensional Hilbert space of quantum systems and the representational structure provided by entanglement and superposition. Experiments demonstrate that ANO-VQCs achieve up to five-fold higher resolution with a relatively small model size, suggesting a promising new direction at the intersection of quantum machine learning and super-resolution.

Ghost inflation is a well-known framework in which cosmological fluctuations can generate enhanced primordial non-Gaussianity, typically of the equilateral type. In its original form, however, it is in tension with current observational constraints. Here we instead consider a setup in which a standard inflaton drives the background evolution, while excitations of a ghost condensate act as spectator fields that interact with the inflaton. This proposal fits naturally within the cosmological collider program: the exchanged particle has a modified dispersion relation, $ω\propto k^2$. We show that this ghost-inspired dynamics weakens the usual Boltzmann suppression, similarly to models with a very small effective sound speed, yielding an enhanced bispectrum signal relative to standard cosmological collider scenarios. At the same time, the horizon-crossing scale remains a free parameter of the theory. As a result, the model shares features of both the de Sitter bootstrap and boostless frameworks. Finally, we derive the differential equations governing cosmological correlators in the ghost-collider setup. Their structure reflects the quadratic momentum dependence of the dispersion relation and distinguishes this scenario from conventional relativistic cases.

Quantum state exclusion is the task of identifying at least one state from a known set that was not used in the preparation of a quantum system. In particular, a given set of quantum states is said to admit state exclusion if there exists a measurement such that, for each state in the set, some measurement outcome rules it out with certainty. However, state exclusion is not always possible in the single-copy setting. In this paper, we investigate whether access to multiple identical copies enables state exclusion. We prove that for any set of three or more pure states, state exclusion becomes possible with a finite number of copies. We further show that the required number of copies may be arbitrarily large -- in particular, for every natural number $N$, we construct sets of states for which exclusion remains impossible with $N$ or fewer copies.

We extend the Pauli Propagation framework to simulate imaginary time evolution. By deriving explicit update rules for the propagation of Pauli operators under imaginary time evolution generated by Pauli strings, we introduce an imaginary time Pauli Propagation (ITPP) algorithm for approximating imaginary time dynamics directly in the Pauli basis. This approach enables the computation of thermal and ground-state properties while retaining the key computational advantages of Pauli Propagation. Benchmarking ITPP on the one-dimensional transverse-field Ising model demonstrates that truncation provides a controlled trade-off between accuracy and computational cost, while also revealing challenges associated with operator growth under imaginary time evolution. Finally, combining imaginary time and real-time Pauli Propagation naturally suggests a pathway toward simulating open quantum system dynamics within a unified framework.

In quantum field theory, entanglement entropy under spatial bipartitioning serves as a powerful information-theoretic probe of quantum correlations. In this work, we present the first comprehensive numerical study of the dynamical evolution and geometric scaling of entanglement entropy in a nonperturbative, strong-field QED setting -- specifically, in the context of the Sauter-Schwinger effect. While the weak-field regime is dominated by area-law states, we show that the entanglement entropy undergoes a transition from area-law to a volume-law scaling for certain strong-field regimes in the pulse-profile parameter space -- signaling a fundamental shift in the underlying correlation structure induced by nonperturbative pair production. For intermediate regimes, the scaling is a power-law that interpolates between area- and volume-law behavior. Finally, we provide interpretations based on the behavior of the low-energy pair-creation spectrum and discuss how these insights could inform future investigations of related phenomena.

Assuming all three known neutrinos are Dirac fermions, $U(1)_{B-L}$ can be an exact symmetry. We show that, if the condition of charge quantization is relaxed, the anomaly-free $B-L$ charges of two out of three right-handed neutrinos can be enhanced by arbitrarily large factors, while all other fermions retain their canonical charges. We call this setup as `enhanced $B-L$ symmetry' and promote it to be local. As long as this enhanced $B-L$ gauge symmetry remains unbroken, neutrinos stay chiral and massless at low energies. Nonzero neutrino masses then require sub-eV-scale symmetry breaking order parameters, which we associate with gravity-induced neutrino condensate. If the enhancement is large and the $B-L$ gauge boson $A'$ is lighter than the heaviest neutrino, then the neutrino decay into $A'$ directly constrains the gauge coupling, which can be significantly stronger than the baryon-based fifth-force tests. Through kinetic mixing with the photon, $A'$ can also mediate neutrino-electron and coherent neutrino-nucleus scatterings, leading to possible signatures in neutrino observatories and dark matter detectors.

Quantum Key Distribution (QKD) protocols enable two distant parties to communicate with information-theoretically proven secrecy. However, these protocols are generally vulnerable to potential mismatches between the physical modeling and the implementation of their quantum operations, thereby opening opportunities for side channel attacks. Device-Independent (DI) QKD addresses this problem by reducing the degree of device modeling to a black-box setting. The stronger security obtained in this way comes at the cost of a reduced noise tolerance, rendering experimental demonstrations more challenging: so far, only one experiment based on trapped ions was able to successfully generate a secret key. Photonic platforms have however long been preferred for QKD thanks to their suitability to optical fiber transmission, high repetition rates, readily available hardware, and potential for circuit integration. In this work, we assess the feasibility of DIQKD on a photonic circuit recently identified by machine learning techniques. For this, we introduce an efficient converging hierarchy of semi-definite programs (SDP) to bound the conditional von Neumann entropy and develop a finite-statistics analysis that takes into account full outcome statistics. Our analysis shows that the proposed optical circuit is sufficiently resistant to noise to make an experimental realization realistic.

Discretely self-similar solutions govern critical gravitational collapse and have been known only numerically since Choptuik's pioneering work. We construct, in closed analytic form, an infinite family of such solutions of the Einstein-massless-Klein-Gordon system using the large-D expansion. We characterize their structure and compare them with numerical critical solutions at finite D, identifying both universal features and distinctly large-D behavior.

Peculiar velocities are analyzed through cosmological perturbations in the Newtonian longitudinal gauge characterized by irrotational shear-free congruences in an Eulerian frame. We show that non-trivial peculiar velocity fields can be generated through Lorentzian boosts in the non-relativistic limit, where the Eulerian frame is obtained from analytic solutions of Einstein's equations sourced by an irrotational shear-free fluid with nonzero energy flux. This approach provides a physically viable interpretation of these analytic solutions, which (in general) admit no isometries, thus allowing, in principle, for modeling time and space varying 3-dimensional fields of peculiar velocities that can be contrasted with observational data on our local cosmography. As a ``proof of concept'' we examine the peculiar velocities of varying dark matter and dark energy perfect fluids with respect to the CMB frame using a simple, spherically symmetric particular solution. The resulting peculiar velocities are qualitatively compatible with observational data on the CMB dipole.

We reconsider back-reaction from large amplitude, short-scale perturbations onto a long wavelength adiabatic mode. In a loop expansion of the long-mode power spectrum, this back-reaction appears first at 1-loop. Due to the separation between the long and short scales, the separate universe method provides a simple and efficient framework for this computation. In this paper, building on our earlier work, we employ a $δN$ formula for the long mode, which captures the effect of short scales. We show that back-reaction at 1-loop is due to either (i) non-linearity of the $δN$ formula, or (ii) 1-loop corrections to the initial conditions. We argue that contributions of type (ii) cannot themselves be described within the separate universe framework, but their properties can be constrained using soft theorems and a ''multi-point propagator'' expansion. When applied to a band of enhanced short-scale perturbations that crossed the horizon during inflation, our result shows that the loop correction decouples from their detailed properties. Furthermore, the back-reaction we obtain is scale-invariant. Its magnitude is model-dependent, but is degenerate with effects from modes that were still sub-horizon at the end of inflation. In this scenario (but not necessarily in all scenarios), we conclude that the effect is not observable.

We present a systematic investigation of deep learning methods applied to quantum error mitigation of noisy output probability distributions from measured quantum circuits. We compare different architectures, from fully connected neural networks to transformers, and we test different design/training modalities, identifying sequence-to-sequence, attention-based models as the most effective on our datasets. These models consistently produce mitigated distributions that are closer to the ideal outputs when tested on both simulated and real device data obtained from IBM superconducting quantum processing units (QPU) up to five qubits. Across several different circuit depths, our approach outperforms other baseline error mitigation techniques. We perform a series of ablation studies to examine: how different input features (circuit, device properties, noisy output statistics) affect performance; cross-dataset generalization across circuit families; and transfer learning to a different IBM QPU. We observe that generalization performance across similar devices with the same architecture works effectively, without needing to fully retrain models.

Recently, it has been shown that group Fourier analysis of quantum states, i.e., decomposing them into the irreducible representations (irreps) of a symmetry group, enables new ways to characterize their resourcefulness. Given that quantum phase spaces (QPSs) provide an alternative description of quantum systems, and thus of the group's representation, one may wonder how such harmonic analysis changes. In this work we show that for general compact Lie-group quantum resource theories (QRTs), the entire family of Stratonovich-Weyl quantum phase space representations-characterized by the Cahill-Glauber parameter $s$-has a clear resource-theoretic and signal-processing meaning. Specifically, changing $s$ implements a group Fourier filter that can be continuously tuned to favor low-dimensional irreps where free states have most of their support ($s=-1$), leave the spectrum unchanged ($s=0$), or highlight resourceful, high-dimensional irreps ($s=1$). As such, distinct QPSs constitute veritable group Fourier filters for resources. Moreover, we show that the norms of the QRT's free state Fourier components completely characterize all QPSs. Finally, we uncover an $s$-duality relating the phase space spectra of free states and typical (Haar-random) highly resourceful states through a shift in $s$. Overall, our results provide a new interpretation of QPSs and promote them to a signal-processing framework for diagnosing, filtering, and visualizing quantum resources.

The Hong-Ou-Mandel test estimates the overlap between spectral functions characterizing the internal degrees of freedom of two single photons. It can be viewed as a photon-native protocol that implements the well-known quantum SWAP test. Here, we propose a native linear-optical protocol that efficiently estimates multivariate traces of quantum states called Bargmann invariants, which are ubiquitous in quantum mechanics. Our protocol may be understood as a photon-native version of the cycle test in the circuit model, which encompasses many-photon multimode quantum states. We show the protocol is sample-efficient and discuss applications, such as generalized suppression laws, efficient quantum kernel estimation for quantum machine learning, eigenspectrum estimation, and the characterization of multiphoton indistinguishability.

In this study, we explore the relation between generalised entropies and the extended uncertainty principle (EUP) models. Starting from the higher-order extended uncertainty principle (HOEUP), we obtain the modified entropy-area relation. Then, we derive the modified Friedmann equations through three different approaches: the first law of thermodynamics at the apparent horizon, the entropic gravity case, and the emergence of cosmic space. Furthermore, we check the validity of the generalised second law (GSL). Notably, HOEUP modified Friedmann equations are the limiting cases of those obtained from a recently proposed novel entropy, which is derived from Modified Newtonian Dynamics (MOND) [{\it Phys. Dark Universe} {\bf 49} (2025) 101967]. Motivated by this connection, we derive a novel EUP, referred to as MOND EUP, from a reverse procedure. This novel EUP reproduces to EUP relations associated with Rényi and dual Kaniadakis entropies in the limiting cases. Moreover, we show that HOEUP corresponds to perturbative limit of MOND entropy. The main new result of this paper is a reverse procedure beginning from a recently proposed novel MOND entropy to construct a unified EUP. This reverse procedure is not limited with the present case. In principle, the method can be applied to other generalised entropy formalisms, suggesting that our findings may establish a unified framework that bridges the generalised entropies, cutoff mechanisms, and EUP models. In particular, the corresponding modified uncertainty principles may have effective cutoff mechanisms for the entropy forms, which do not explicitly display cutoff mechanisms. Thus, these entropies may have cutoff mechanism due to their corresponding modified uncertainty principles.

Quantum memories are key components of emerging quantum technologies. They are designed to store quantum states and retrieve them on demand without losing features such as superposition and entanglement. Verifying that a memory preserves these features is indispensable for applications such as quantum computation, cryptography and networks, yet no general and assumption-free method has been available. Here, we present a device-independent approach for certifying black-box quantum memories, requiring no trust in any part of the experimental setup. We do so by probing quantum systems at two points in time and then confronting the observed temporal correlations against classical causal models through violations of causal inequalities. We perform a proof-of-principle experiment in a trapped-ion quantum processor, where we certify 35 ms of a qubit memory. Our method establishes temporal correlations and causal modelling as practical and powerful tool for benchmarking key ingredients of quantum technologies, such as quantum gates or implementations of algorithms.

While many bounds have been proved for partial trace inequalities over the last decades for a large variety of quantities, recent problems in quantum information theory demand sharper bounds. In this work, we study optimal bounds for partial trace quantities in terms of the spectrum; equivalently, we determine the best bounds attainable over unitary orbits of matrices. We solve this question for Schur-convex functionals acting on a single partial trace in terms of eigenvalues for self-adjoint matrices and then we extend these results to singular values of general matrices. We subsequently extend the study to Schur-convex functionals that act on several partial traces simultaneously and present sufficient conditions for sharpness. In cases where closed-form maximizers cannot be identified, we present quadratic programs that yield new computable upper bounds for any Schur-convex functional. We additionally present examples demonstrating improvements over previously known bounds. Finally, we conclude with the study of optimal bounds for an $n$-qubit system and its subsystems of dimension $2$.

Ensembles of random fuzzy non-commutative geometries may be described in terms of finite (\(N^2\)-dimensional) Dirac operators and a probability measure. Dirac operators of type \((p,q)\) are defined in terms of commutators and anti-commutators of \(2^{p+q-1}\) hermitian matrices \(H_k\) and tensor products with a representation of a Clifford algebra. Ensembles based on this idea have recently been used as a toy model for quantum gravity, and they are interesting random-matrix ensembles in their own right. We provide a complete theoretical picture of crossovers, phase transitions, and symmetry breaking in the \(N \to \infty \) limit of 1-parameter families of quartic Barrett-Glaser ensembles in the one-matrix cases \((1,0)\) and \((0,1)\) that depend on one coupling constant \(g\). Our theoretical results are in full agreement with previous and new Monte-Carlo simulations.

This paper studies the effect of extra dimensions on the arrow of time within the framework of $f(R)$ gravity. We demonstrate that the observed irreversibility of physical processes can be explained by the monotonic growth of the extra-dimensional space. Unlike traditional cosmological approaches, our model does not link the arrow of time to the entropy of matter or radiation; rather, it identifies it with the Bekenstein-Hawking-Wald entropy of the geometric background. We establish a formal relation between the volume of the multidimensional manifold and the statistical weights of its geometric states. This leads to a fundamental relationship where the flow of time is intrinsically linked to the growth of multidimensional entropy. A key consequence of our framework is that the arrow of time remains a persistent feature for a 4D observer situated on a brane, even in the complete absence of matter or radiation. This directionality is driven by the dominant entropy production in the higher-dimensional bulk, which dominates local statistical fluctuations and ensures a stable causal direction.

We investigate accretion disk dynamics and the formation of quasi-periodic oscillations (QPOs) in the infrared limit around Kerr-like black holes in asymptotically safe gravity. Relativistic hydrodynamic solutions of Bondi-Hoyle-Lyttleton (BHL) accretion reveal that quantum corrections significantly modify the structure of the shock cone formed around the black hole. The black hole spin controls the asymmetric of the shock cone through frame-dragging effects, whereas the quantum correction parameter softens the effective gravitational potential, resulting in a wider shock opening angle, weaker post-shock compression, and reduced density concentration within the cone. Time-dependent mass accretion rates reveal oscillation modes trapped within the shock cone. The power spectral density (PSD) investigations suggest that these modes naturally generate low-frequency QPOs, whose amplitudes, coherence, and harmonic structure depend on both the spin and the quantum correction parameter. The PSD analyses performed at different radial locations reveal that identical QPO frequencies are obtained in all cases. The numerically detected frequencies result from the excitation of global oscillation modes trapped within the post-shock region. The resulting global modes are found to consist of fundamental frequencies, their associated harmonic overtones, and near-commensurate frequency ratios such as 2:1 and 3:2. Coherent oscillations are enhanced and near-commensurate frequency ratios are produced when moderate rotation and moderate quantum corrections are coupled. Large quantum correction parameters, on the other hand, wash out unique spectral peaks and suppress oscillation amplitudes.

Generalized contextuality is a possible indicator of non-classical behaviour in quantum information theory. In finite-dimensional systems, this is justified by the fact that noncontextual theories can be embedded into some simplex, i.e. into a classical theory. We show that a direct application of the standard definition of generalized contextuality to continuous variable systems does not envelope the statistics of some basic measurements, such as the position observable. In other words, we construct families of fully classical, i.e. commuting, measurements that nevertheless can be used to show contextuality of quantum theory. To overcome the apparent disagreement between the two notions of classicality, that is commutativity and noncontextuality, we propose a modified definition of generalised contextuality for continuous-variable systems. The modified definition is based on a physically-motivated approximation procedure, that uses only finite sets of measurement effects. We prove that in the limiting case this definition corresponds exactly to an extension of noncontextual models that benefits from non-constructive response functions. In the process, we discuss the extension of a known connection between contextuality and no-broadcasting to the continuous-variable scenario, and prove structural results regarding fixed points of infinite-dimensional entanglement breaking channels.

Causality is one of the fundamental structures of spacetimes, it determines the possible behaviour and propagation of physical information through different relations. Causal structure can be analysed through the various modal logics it induces. The modal logics for the standard chronological and causal relations of the archetypal Minkowski spacetime have been classified. However only partial results have been achieved for the strict variant of the causal relation, also known as the after relation. The present work continues this analysis towards arbitrary spacetimes. By utilizing the definition of the causal relations through causal paths, we can lift known results about the modal logics of Minkowski spacetime to general spacetimes. In particular, for the after relation, we show that a previously studied formula within the logics of Minkowski spacetime holds in arbitrary spacetimes. We introduce a related modal formula that demonstrates that the logic of two-dimensional spacetimes are more expressive than higher dimensional ones. Lastly, we study the interrelation between the logical properties and physical properties along the causal ladder, a classification of causal structures according to a hierarchy of physically relevant properties.

Mixed-integer linear programming problems are extensively used in industry for a wide range of optimization tasks. However, as they get larger, they present computational challenges for classical solvers within practical time limits. Quantum annealers can, in principle, accelerate the solution of problems formulated as quadratic unconstrained binary optimization instances, but their limited scale currently prevents achieving practical speedups. Quantum-classical algorithms have been proposed to take advantage of both paradigms and to allow current quantum computers to be used in larger problems. In this work, a hardware-agnostic Benders' decomposition algorithm and a series of enhancements with the goal of taking the most advantage of quantum computing are presented. The decomposition consists of a master problem with integer variables, which is reformulated as a quadratic unconstrained binary optimization problem and solved with a quantum annealer, and a linear subproblem solved by a classical computer. The enhancements consist, among others, of different embedding processes that substantially reduce the pre-processing time of the embedding computation without compromising solution quality, a conservative handling of cut constraints, and a stopping criterion that accounts for the limited size of current quantum computers and their heuristic nature. The proposed algorithm is benchmarked against classical approaches using a D-Wave quantum annealer for a scalable family of transmission network expansion planning problems.

We investigate the behavior of quantum trajectories conditioned on measurement outcomes. Under a condition related to the absence of so-called dark subspaces, Kümmerer and Maassen had shown that such trajectories almost surely purify in the long run. In this article, we first present a simple alternative proof of this result using Lyapunov methods. We then strengthen the conclusion by proving that purification actually occurs at an exponential rate in expectation, again using a Lyapunov approach. Furthermore, we address the quantum state estimation problem by propagating two trajectories under the same measurement record--one from the true initial state and the other from an arbitrary initial guess--and show that the estimated trajectory converges exponentially fast to the true one, thus quantifying the rate at which information is progressively revealed through the measurement process.

Spatially-resolved photon counting of a twin beam performed by an iCCD camera allows for versatile tailoring the properties of the beams formed by parts of the original twin beam. Dividing the idler beam of the twin beam into three equally-intense parts and post-selecting by detecting a given number of photocounts in the whole signal beam we arrive at the idler fields exhibiting high degrees of nonclassicality and being endowed with tripartite quantum correlations. Nonclassicality is analyzed with the help of suitable nonclassicality witnesses and their corresponding nonclassicality depths. Suitable parameters are introduced to quantify quantum correlations. These parameters are analyzed as they depend on the field intensity. The experimental photocount histograms are reconstructed by the maximum-likelihood approach and the obtained photon-number distributions are compared with a suitable model in which the original twin beam is approximated by an appropriate multi-mode Gaussian field and undergoes the corresponding beams' transformations.

We study a cosmological scenario featuring an early phase of null energy condition (NEC) violation. Within this framework, we show that perturbative unitarity bounds place strong constraints on both the amplitude and the spectral tilt of primordial gravitational waves. Our analysis is largely insensitive to the detailed realization of the transition between the NEC-violating phase and subsequent cosmological phases, allowing our results to be extended to a broader class of models. Finally, the perturbative unitarity approach employed here is applicable to a wide range of cosmological scenarios.

Two applications of gates from the B gate equivalence class can generate all two-qubit gates. This local equivalence class is invariant under the mirror (multiplication with the SWAP gate) operation, inverse (Hermitian conjugate) operation, and the combined inverse and mirror operations. The last two symmetries are associated with the ability of a two-qubit gate to generate the two-qubit local gates and the SWAP gate in two applications. No single local equivalence class of two-qubit gates, except the B gate equivalence class, has these two symmetries. Only the planar regions of the Weyl chamber, describing the mirror operation, contain the local equivalence classes with either one of the two symmetries. We show that there exist one-parameter families of local equivalence classes on these planes, with and without the B gate equivalence class, such that each of them can be used to construct a parameterized universal two-qubit quantum circuit that involves only two nonlocal two-qubit gates. We also discuss the implementation of the gates from a few families of local equivalence classes on superconducting quantum computers for optimal generation of all two-qubit gates.

We propose a Bayesian evidence-based inference framework based on relative belief ratios and apply it to discriminating between one and two incoherent optical point sources using spatial-mode demultiplexing (SPADE). Unlike the Helstrom measurement, SPADE require no collective detection and its optimal for asymptotically large samples. Our method avoids ad hoc statistical constructs and relies solely on the information contained in the data, with all assumptions entering only through the likelihood model and prior beliefs. Using experimental evidence, we demonstrate the superior resolving performance of SPADE over direct imaging from a new and extensible perspective; one that naturally generalizes to multiple sources and offers a practical robust approach to analyzing quantum-enhanced superresolution.

Quantum hypothesis testing concerns the discrimination between quantum states. This paper introduces a novel lower bound for asymmetric quantum hypothesis testing that is based on the Nussbaum-Szkoła mapping. The lower bound provides a unified recovery of converse results across all major asymptotic regimes, including large-, moderate-, and small-deviations. Unlike existing bounds, which either rely on technically involved information-spectrum arguments or suffer from fixed prefactors and limited applicability in the non-asymptotic regime, the proposed bound arises from a single expression and enables, in some cases, the direct use of classical results. It is further demonstrated that the proposed bound provides accurate approximations to the optimal quantum error trade-off function at small blocklengths. Numerical comparisons with existing bounds, including those based on fidelity and information spectrum methods, highlight its improved tightness and practical relevance.

Although quantum computing holds promise for solving Combinatorial Optimization Problems (COPs), the limited qubit capacity of NISQ hardware makes large-scale instances intractable. Conventional methods attempt to bridge this gap through decomposition or compression, yet they frequently fail to capture global correlations of subsystems, leading to solutions of limited quality. We propose the Distributed Variational Quantum Algorithm (DVQA) to overcome these limitations, enabling the solution of 1,000-variable instances on constrained hardware. A key innovation of DVQA is its use of the truncated higher-order singular value decomposition to preserve inter-variable dependencies without relying on complex long-range entanglement, leading to a natural form of noise localization where errors scale with subsystem size rather than total qubit count, thus reconciling scalability with accuracy. Theoretical bounds confirm the algorithm's robustness for p-local Hamiltonians. Empirically, DVQA achieves state-of-the-art performance in simulations and has been experimentally validated on the Wu Kong quantum computer for portfolio optimization. This work provides a scalable, noise-resilient framework that advances the timeline for practical quantum optimization algorithms.

Quantum-correlated states of light, such as squeezed states, constitute a fundamental resource for quantum technologies, enabling enhanced performance in quantum metrology, quantum information processing, and quantum communications. The practical deployment of such technologies requires squeezed-light sources that are compact, efficient, low-cost, and robust. Here we report a compact narrowband source of two-mode squeezed light at 795 nm based on four-wave mixing in hot 85Rb atomic vapor. The source is implemented in a small, modular architecture featuring a single fiber-coupled input, an electro-optic phase modulator combined with a single Fabry-Perot etalon for probe generation, and two free-space output modes corresponding to the signal and conjugate fields. Optimized for low pump power, the system achieves up to -8 dB of intensity-difference squeezing at an analysis frequency of 0.8 MHz with a pump power of only 300 mW. The intrinsic narrowband character of the generated quantum states makes this source particularly well suited for atomic-based quantum sensing and quantum networking, including interfaces with atomic quantum memories. Our results establish a versatile and portable platform for low-SWaP squeezed-light generation, paving the way toward deployable quantum-enhanced technologies.

We present a systematic method for constructing universal composite phase gates with a continuously tunable target phase. Using a general Cayley--Klein parametrization of the single-pulse propagator, we design gates from an even number of nominal $π$ pulses and derive analytic phase families by canceling, order by order in a small deviation parameter, the leading contributions to the undesired off-diagonal element of the composite propagator, independently of the dynamical phase. The resulting sequences provide intrinsic robustness against generic control imperfections and parameter fluctuations and remain valid for arbitrary pulse shapes. Numerical simulations in a standard two-level model confirm high-order error suppression and demonstrate broad, flat high-fidelity plateaus over wide ranges of simultaneous pulse-area and detuning errors, highlighting the efficiency of the proposed universal composite phase gates for resilient phase control in quantum information processing.

In a gravitational lensing system, the relative transverse velocities of the lens, source, and observer induce a frequency shift in the observed radiation. While this shift is typically negligible in most astrophysical contexts, strategies for its detection have been proposed for both electromagnetic and gravitational waves. This paper provides a rigorous theoretical treatment of the effect, deriving general expressions for the frequency shift within a lensing system embedded in a cosmological spacetime. Our formulation remains valid for arbitrary distances and velocities - including highly relativistic regimes - under any Friedmann-Lemaître-Robertson-Walker metric. Expanding upon previous papers on moving lenses, we provide a detailed analysis of frequency effects induced by lenses moving at relativistic speeds. Furthermore, we extend standard lensing theory by deriving an exact formula for the variation in the source's viewing direction. This result is of interest for strongly anisotropic emitters, such as compact binary systems emitting gravitational waves. Finally, we quantify the apparent misalignment between the lens and the source's two images produced by time-delay effects in lens systems moving with high velocity.

Superradiance is a hallmark of cooperative quantum emission, where radiative decay is collectively enhanced by coherence among emitters. Here, extending superradiant effects to photon pair generation from multi-level atoms, two-photon process offers a pathway to novel quantum light sources and a useful case for practical superradiance. We report bright heralded single-photon superradiance via spontaneous four-wave mixing in a 1-mm-long, high-density cesium vapor cell. By reducing the average distance between atoms in the atomic vapor to 0.29 times the idler photon wavelength, we observe a dramatic narrowing of the temporal two-photon wavefunction. This compression of temporal two-photon wavefunction evidences the superradiance of heralded photons in the collective two-photon emission dynamics. Furthermore, our heralded single-photon superradiance is accompanied by a coincidence-to-accidental ratio of 200 and the detected photon-pair counting exceeding 10^6 pairs/s. These findings establish dense thin atomic vapors as a practical, robust medium for realizing superradiant photon sources, with immediate relevance for quantum optics and the development of efficient photonic quantum technologies.

Estimating the energy spectra of quantum many-body systems is a fundamental task in quantum physics, with applications ranging from chemistry to condensed matter. Algorithmic shadow spectroscopy is a recent method that leverages randomized measurements on time-evolved quantum states to extract spectral information. However, implementing accurate time evolution with low-depth circuits remains a key challenge for near-term quantum hardware. In this work, we propose a hybrid quantum-classical protocol that integrates Time Evolution via Probabilistic Angle Interpolation (TE-PAI) into the shadow spectroscopy framework. TE-PAI enables the simulation of time evolution using shallow stochastic circuits while preserving unbiased estimates through quasiprobability sampling. We construct the combined estimator and derive its theoretical properties. Through numerical simulations, we demonstrate that our method accurately resolves energy gaps and exhibits enhanced robustness to gate noise compared to standard Trotter-based shadow spectroscopy. We further validate the protocol experimentally on up to 20 qubits using IBM quantum hardware. This makes TE-PAI shadow spectroscopy a promising tool for spectral analysis on noisy intermediate-scale quantum (NISQ) devices.

The aim of this exposé is to make explicit the analogy between the classical notion of non-independent probability distribution and the quantum notion of entangled state. To bring that analogy forth, we consider a classical systems with two dependent random variables and a quantum system with two components. In the classical case, afet observing one of the random variables, the underlying sample space and the probability distribution change. In the quantum case, when and event pertaining to one of the components is observed, the post-measurement state captures, both, the change in the state of the system and implicitly the new probability distribution. The predictions after a measurement in the classical case and in the quantum case, have to be computed with the conditional distribution given the value of the observed variable.

In the Wigner-Weyl phase space formulation of quantum mechanics, we analyse the problem of the spreading of an initial state or an initial operator under time evolution when described in terms of the Krylov basis. After constructing the phase space representations of the Krylov basis states generated by a Hamiltonian from a given initial state by using the Weyl transformation, we subsequently use them to cast the Krylov state complexity as an integral over the phase space in terms of the Wigner function of the time-evolved initial state, so that the contribution of the classical Liouville equation and higher-order quantum corrections to the Wigner function time evolution equation towards the Krylov state complexity can be identified. Next, we construct the double phase space functions associated with the Krylov basis for the operators by using a suitable generalisation of the Weyl transformation applicable for superoperators, and use them to rewrite the Krylov operator complexity as an integral over the double phase space in terms of a generalisation of the usual Wigner function. These results, in particular, show that the complexity measures based on the expansion of a time-evolved state (or an operator) in the Krylov basis can be thought to belong to a general class of complexity measures constructed from the expansion coefficients of the time-dependent Wigner function in an orthonormal basis in the phase space, and help us to connect these complexity measures with measures of complexity of time-evolved state based on harmonic expansion of the time-dependent Wigner function.

Any single on-off photocounter, which can only detect the presence or absence of photons without discriminating their number, is not capable of identifying nonclassical nature of light. This limitation arises because any photocounting statistics obtained with such a detector can be easily reproduced with coherent states of a light mode. We show that a simple modification of an on-off detector -- introducing controlled attenuation as a tunable setting -- enables such detectors to reveal nonclassical properties of radiation fields.

We consider corrections to the Bekenstein Hawking Area Formula for black hole entropy, which have inverse powers of the horizon area for very large horizon areas, for classical spherically symmetric black hole solutions of F(R) modified gravity theory, using the Wald formula for the entropy function with modifications suggested by Jacobson, Kang and Myers. Requiring that the coefficient of such corrections be absolutely consistent with gravitational wave observational results validating the Hawking Area Theorem for binary black hole coalescences, implies constraints on parameters of F(R) gravity. For the sake of comparison, we present a computation of inverse area corrections for quantum black holes in quantum general relativity, using the It from Bit approach of Wheeler modified by some tenets of Loop Quantum Gravity.

We present a method for performing a full graph expansion for light-matter systems, utilizing the linked-cluster theorem. This method enables us to explore $1/N$ corrections to the thermodynamic limit $N\to \infty$ in the number of particles, giving us access to the mesoscopic regime. While this regime is yet largely unexplored due to the challenges of studying it with established approaches, it incorporates intriguing features, such as entanglement between light and matter that vanishes in the thermodynamic limit. As a representative application, we calculate physical quantities of the low-energy regime for the paradigmatic Dicke-Ising chain in the paramagnetic normal phase by accompanying the graph expansion with both exact diagonalization (NLCE) and perturbation theory (\pcst), benchmarking our approach against other techniques. We investigate the ground-state energy density and photon density, showing a smooth transition from the microscopic to the macroscopic regime up to the thermodynamic limit. Around the quantum critical point, we extract the $1/N$ corrections to the ground-state energy density to obtain the critical point and critical exponent using extrapolation techniques.

Simulation-based inference provides a powerful framework for Bayesian inference when the likelihood is analytically intractable or computationally prohibitive. By leveraging machine-learning techniques and neural density estimators, it enables flexible likelihood or posterior modeling directly from simulations. We introduce Residual Neural Likelihood Estimation (RNLE), a modification of Neural Likelihood Estimation (NLE) that learns the likelihood of non-Gaussian noise in gravitational-wave detector data. Exploiting the additive structure of the signal and noise generation processes, RNLE directly models the noise distribution, substantially reducing the number of simulations required for accurate parameter estimation and improving robustness to realistic noise artifacts. The performance of RNLE is demonstrated using a toy model, simulated gravitational-wave signals, and real detector noise from ground based interferometers. Even in the presence of loud non-Gaussian transients, glitches, we show that RNLE can achieve reliable parameter recovery when trained on appropriately constructed datasets. We further assess the stability of the method by quantifying the variability introduced by retraining the conditional density estimator on statistically identical datasets with different optimization seeds, referred to as training noise. This variability can be mitigated through an ensemble approach that combines multiple RNLE models using evidence-based weighting. An implementation of RNLE is publicly available in the sbilby package, enabling its deployment within gravitational-wave astronomy and a broad range of scientific applications requiring flexible, simulation-based likelihood estimation.

We investigate dimensional constraints arising from representation theory when abstract graph edges possess internal degrees of freedom but lack geometric properties. We prove that such internal degrees of freedom can only encode directional information, necessitating quantum states in $\mathbb{C}^2$ (qubits) as the minimal representation. Any geometrically consistent projection of these states maps necessarily to $\mathbb{R}^3$ via the Bloch sphere. This dimensional constraint $d=3$ emerges through self-consistency: edges without intrinsic geometry force directional encoding ($\mathbb{C}^2$), whose natural symmetry group $SU(2)$ has three-dimensional Lie algebra, yielding emergent geometry that validates the hypothesis via Bloch sphere correspondence ($S^2 \subset \mathbb{R}^3$). We establish uniqueness (SU($N>2$) yields $d>3$) and robustness (dimensional saturation under graph topology changes). The Euclidean metric emerges canonically from the Killing form on $\mathfrak{su}(2)$. A global gauge consistency axiom is justified via principal bundle trivialization for finite graphs. Numerical simulations verify theoretical predictions. This result demonstrates how dimensional structure can be derived from information-theoretic constraints, with potential relevance to quantum information theory, discrete geometry, and quantum foundations.

Improving multiparameter quantum estimation in magnonic systems via quantum noise suppression is a well-established and critical research objective. In this work, we propose an experimentally realistic scheme to improve the precision of simultaneously estimating different parameters in a cavity-magnon system by utilizing a degenerate optical parametric amplifier (OPA). The OPA enhances the estimation precision by decreasing the most informative quantum Cramér-Rao bound, calculated employing the symmetric logarithmic derivative (SLD) and the right logarithmic derivative (RLD). We show that when nonlinearity is introduced into the system, quantum noise is significantly suppressed. Our results show how different physical parameters influence multiparameter estimation precision and provide a detailed discussion of the associated physical mechanisms in the steady state. Our results focus on exploring practical Gaussian measurement schemes that can be realized experimentally. Besides, we further analyze the system's dynamics, comparing both the SLD quantum Fisher information (QFI) and the classical Fisher information (CFI) for both homodyne and heterodyne detection. This approach provides a robust foundation for multiparameter quantum estimation, offering significant potential for application in hybrid magnomechanical and optomechanical systems.

We introduce equations of motion for a parity-violating fermionic mean-field theory (PV-FMFT): a numerically efficient fermionic mean-field theory based on parity-violating fermionic Gaussian states (PV-FGS). This work provides explicit equations of motion for studying the real- and imaginary-time evolution of spin-1/2 Hamiltonians with arbitrary geometries and interactions. We extend previous formulations of parity-preserving fermionic mean-field theory (PP-FMFT) by including fermionic displacement operators in the variational Ansatz. Unlike PP-FMFT, PV-FMFT can be applied to general spin-1/2 Hamiltonians, describe quenches from arbitrary initial spin-1/2 product states, and compute local and non-local observables in a straight-forward manner at the same modest computational cost as PP-FMFT -- scaling as $O(N^3)$ in the worst case for a system of $N$ spins or fermionic modes. We demonstrate that PV-FMFT can exactly capture the imaginary- and real-time dynamics of non-interacting spin-1/2 Hamiltonians. We then study the post quench-dynamics of the one- and two-dimensional Ising model in presence of longitudinal and transversal fields with PV-FMFT and compute the single site magnetization and correlation functions, and compare them against results from other state-of-the-art numerical approaches. In two-dimensional spin systems, we show that the employed spin-to-fermion mapping can break rotational symmetry within the PV-FMFT description, and we discuss the resulting consequences for the calculated correlation functions. Our work introduces PV-FMFT as a benchmark for other numerical techniques and quantum simulators, and it outlines both its capabilities and its limitations.

In the context of $p$-adic quantum mechanics, we investigate composite systems of $p$-adic qubits and $p$-adically controlled quantum logic gates. We build on the notion of a single $p$-adic qubit as a two-dimensional irreducible representation of the compact $p$-adic special orthogonal group SO(3)$_p$. We show that the classification of these representations reduces to the finite case, as they all factorise through some finite quotient SO(3)$_p$ mod $p^k$. Then, we tackle the problem of $p$-adic qubit composition and entanglement, fundamental for a $p$-adic formulation of quantum information processing. We classify the representations of SO(3)$_p$ mod $p$, and analyse tensor products of two $p$-adic qubit representations lifted from SO(3)$_p$ mod $p$. We solve the Clebsch-Gordan problem for such systems, revealing that the coupled bases decompose into singlet and doublet states. We further study entanglement arising from those stable subsystems. For $p=3$, we construct a set of gates from $4$-dimensional irreducible representations of SO(3)$_p$ mod $p$ that we prove to be universal for quantum computation.

The multimode bunching probability is expected to provide a useful criterion for validating boson sampling experiments. Its applicability, however, is challenged by the existence of anomalous bunching, namely paradoxical situations in which partially distinguishable particles exhibit a higher bunching probability in two or more modes than perfectly indistinguishable ones. Using multimode bunching as a reliable criterion of genuine indistinguishability, therefore, requires a clear identification of the interferometric configurations in which anomalous bunching can or cannot occur. In particular, since uncontrolled small time delays between single-photon pulses constitute a common source of mode mismatch in current photonic platforms, it is essential to determine whether the resulting photon distinguishability might lead to anomalous bunching. Here, we first identify a broad class of interferometric configurations in which anomalous bunching is rigorously excluded, thereby establishing regimes where multimode bunching-based validation remains valid. Then, we find that, quite unexpectedly, temporal mode mismatch does not belong to this class. We exhibit a specific interferometric setup in which temporal distinguishability enhances multimode bunching, demonstrating that time delays can induce an anomalous behavior. These results help clarify the conditions under which multimode bunching remains a reliable validation tool.

Non-Hermitian physics has emerged as a rapidly advancing field of research, revealing a range of novel phenomena and potential applications. Traditional non-Hermitian Hamiltonians are typically simulated by constructing asymmetric couplings or by introducing dissipation and gain to realize non-Hermitian systems. The quadratic bosonic system (QBS) with squeezing interaction is intrinsically Hermitian; however, its dynamical evolution matrix in both real and momentum spaces is non-Hermitian. Based on this, applying a field-operator transformation xp to the dynamical evolution matrix yields quadrature nonreciprocal transmission between the x and p operators. This nonreciprocal characteristic can be utilized in signal amplifiers. On the other hand, within the Bogoliubov-de Gennes framework in momentum space, one can observe non-Hermitian topological phenomena such as point-gap topology and the non-Hermitian skin effect, both induced by spectra with nonzero winding numbers. Additionally, QBS can be employed to realize non-Hermitian Aharonov-Bohm cages and to extend non-Bloch band theory. Previous studies in non-Hermitian physics have largely concentrated on classical systems. The influence of non-Hermitian properties on quantum effects remains a key issue awaiting exploration and has evolved into a research direction at the interface of non-Hermitian and quantum physics.

We formulate pure-state entanglement in families as a geometric obstruction. In standard quantum information, entanglement is defined relative to a chosen tensor-product factorization of a fixed Hilbert space. In contrast, for a twisted family of pure-state spaces, which can be described by Azumaya algebras $A$ of degree $n$ on $X$ and their Severi-Brauer schemes \[ SB(A)=P\times^{PGL_n}\mathbb{P}^{n-1}\to X, \] such a subsystem choice may fail to globalize. We formalize this algebro-geometrically: fixing a factorization type $\mathbf d=(d_1,\dots,d_s)$ with $n=\prod_i d_i$, the existence of a global product-state locus of type $\mathbf d$ is equivalent to a reduction of the underlying $PGL_n$-torsor $P\to X$ to the stabilizer $G_{\mathbf d}\subset PGL_n$. Thus, entanglement is the obstruction to the existence of a relative Segre subscheme inside $SB(A)$. Writing $Σ_{\mathbf d}\subset \mathbb{P}^{n-1}$ for the Segre variety, we call a reduction to $G_{\mathbf d}$ a $\mathbf d$-subsystem structure. Our first main result identifies the moduli of $\mathbf d$-subsystem structures with the quotient $P/G_{\mathbf d}$. Moreover, we realize naturally $P/G_{\mathbf d}$ as a locally closed subscheme of the relative Hilbert scheme, \[ \text{Hilb}^{Σ_{\mathbf d}}\!\bigl(SB(A)/X\bigr)\ \subset\ \text{Hilb}\bigl(SB(A)/X\bigr), \] parametrizing relative closed subschemes fppf-locally isomorphic to $Σ_{\mathbf d}\times X$.

On-chip generation of high-purity single photons is essential for scalable photonic quantum technologies. Spontaneous parametric down-conversion (SPDC) is widely used to generate photon pairs for heralded single-photon sources, but intrinsic spectral correlations of the pairs often limit the purity and interference visibility of the heralded photons. Existing approaches to suppress these correlations rely on narrowband spectral filtering, which introduces loss, or exploiting different polarizations, which complicates on-chip integration. Here, we demonstrate a new strategy for generating spectrally separable photon pairs in thin-film lithium niobate nanophotonic circuits by harnessing higher-order spatial modes, with all interacting fields residing in the same polarization. Spectral separability is achieved by engineering group-velocity matching using higher-order transverse-electric modes, combined with a Gaussian-apodized poling profile to further suppress residual correlations inherent to standard periodic poling. Subsequent on-chip mode conversion with efficiency exceeding 95\% maps the higher-order mode to the fundamental mode and routes the photons into distinct output channels. The resulting heralded photons exhibit spectral purities exceeding 94\% inferred from joint-spectral intensity and 89\% from unheralded $g^{(2)}$ measurement. This approach enables flexible spectral and temporal engineering of on-chip quantum light sources for quantum computing and quantum networking.

We investigate the phenomenological implications of quantum gravity on rotating black holes within the framework of the Generalized Extended Uncertainty Principle (GEUP), which incorporates both minimal length (ultraviolet) and large-scale (infrared) corrections. Lacking a full non-perturbative formulation of quantum gravity, we adopt a metric-based approach. We construct a stationary, axisymmetric ansatz via the Newman-Janis algorithm to model the kinematic features of a rotating black hole subject to Generalized Extended Uncertainty Principle (GEUP) corrections. The thermodynamic analysis reveals that in the infrared-dominated regime, the Hawking temperature scales as $T_H \sim M^{-3}$, leading to a rapid cooling phase that significantly prolongs the lifetime of supermassive black holes. We derive the modified Teukolsky Master Equation for gravitational perturbations and demonstrate that the background geometry preserves the isospectrality between axial and polar modes. In the eikonal limit, the quasinormal mode (QNM) spectrum exhibits orthogonal shifts: the minimal length parameter $β$ induces a spectral blueshift and enhanced damping, while the large-scale parameter $α$ induces a spectral redshift and suppressed damping. Finally, we constrain the theory using observational data from LIGO/Virgo and the Event Horizon Telescope. We establish that the shadow of M87* is approximately $10^6$ times more sensitive to large-scale corrections than Sgr A*, placing stringent bounds on the EUP parameter, while gravitational wave spectroscopy provides complementary constraints on the GUP sector.

We introduce $\texttt{WaveletNet}$, a wavelet-based neural network architecture to identify and reduce non-Gaussian noise in gravitational wave data. Traditionally, convolutional neural networks (CNNs) have been widely used as a flexible machine learning method to mitigate non-Gaussian noise. However, training CNNs requires many data samples, especially when the input data segments are long. Glitches that mimic high-mass black hole signals are empirically known to have a wavelet-like structure. We exploit this property in $\texttt{WaveletNet}$ by using simple neural networks to learn the best family of wavelets to model glitches in the LIGO-Virgo-KAGRA O3 data. Due to its simplicity, our framework is significantly more sample-efficient than CNNs. As a use case, we build upon the $\texttt{TIER}$ method and show how $\texttt{WaveletNet}$ can improve the performance of any search pipeline. We take potential GW candidates from the pipeline, and then downweight the candidates having noisy strain regions in their vicinity. We use our framework in a modular way: we provide an output score which can be added to the pipeline's existing detection statistic score for the candidates. We test our method using candidates from the $\texttt{IAS-HM}$ search pipeline and show that it improves the search sensitive volume by up to 15% for high-mass, asymmetric binaries.

The Box-Muller transform is a widely used method to generate Gaussian samples from uniform samples. Quantum amplitude encoding methods encode the multi-variate normal distribution in the amplitudes of a quantum state. This work presents the Quantum Box-Muller transform which creates a superposition of binary-encoded grid points representing the multi-variate normal distribution. The gate complexity of our method depends on quantum arithmetic operations and, using a specific set of known implementations, the complexity is quadratic in the number of qubits. We apply our method to Monte-Carlo integration, in particular to the estimation of the expectation value of a function of Gaussian random variables. Our method implies that the state preparation circuit used multiple times in amplitude estimation requires only quantum arithmetic circuits for the grid points and the function, in addition to a single controlled rotation. We show how to provide the expectation value estimate with an error that is exponentially small in the number of qubits, similar to the amplitude-encoding setting with error-free encoding.

We introduce a physics-informed Generative Adversarial Network framework that recasts quantum resource-state generation as an inverse-design task. By embedding task-specific utility functions into training, the model learns to generate valid two-qubit states optimized for teleportation and entanglement broadcasting. Comparing decomposition-based and direct-generation architectures reveals that structural enforcement of Hermiticity, trace-one, and positivity yields higher fidelity and training stability than loss-only approaches. The framework reproduces theoretical resource boundaries for Werner-like and Bell-diagonal states with fidelities exceeding ~98%, establishing adversarial learning as a lightweight yet effective method for constraint-driven quantum-state discovery. This approach provides a scalable foundation for automated design of tailored quantum resources for information-processing applications, exemplified with teleportation and broadcasting of entanglement, and it opens up the possibility of using such states in efficient quantum network design.

The unrivaled maturity of its nanofabrication makes silicon a promising hardware platform for quantum information processing. To this end, efficient single-photon sources and spin-photon interfaces have been implemented by integrating color centers or erbium dopants into nanophotonic resonators. However, the optical emission frequencies in this approach are subject to temporal fluctuations on both long and short timescales, which hinders the development of quantum applications. Here, we investigate this limitation and demonstrate that it can be alleviated by integrating the emitters into Fabry-Perot instead of nanophotonic resonators. Their larger optical mode volume enables both increasing the distance to crystal surfaces and operating at a lower dopant concentration, which reduces implantation-induced crystal damage and interactions between emitters. As a result, we observe a fivefold reduction of the spectral diffusion linewidth down to 4.0(2) MHz. Calculations and experimental investigations of isotopically purified 28-Si crystals suggest that the remaining spectral instability is caused by laser-induced electric-field fluctuations. In direct comparison with a nanophotonic device, the instability is significantly reduced at the same intracavity power, enabling a tenfold increase of the optical coherence time up to 20(1) microseconds. These findings represent a key step towards spectrally stable spin-photon interfaces in silicon and their potential applications in quantum networking and distributed quantum information processing.

Mid-circuit measurements are a major bottleneck for superconducting quantum processors because they are slower and noisier than gates. Measurement-free quantum error correction (mfec) replaces repeated measurements and classical feed-forward by coherent quantum feedback, but existing mfec protocols suffer from severe connectivity overhead when mapped to planar surface-code architectures: transversal interactions between logical patches require SWAP chains of length $O(d)$ in the code distance, which increase depth and generate hook errors. This work introduces a 3D stacked surface-code architecture for measurement-free fault-tolerant quantum error correction that removes this connectivity bottleneck. Vertical transversal couplers between aligned surface-code patches enable coherent parity mapping and feedback with zero SWAP overhead, realizing constant-depth $O(1)$ inter-layer operations in d while preserving local 2D stabilizer checks. A fault-tolerant mfec protocol for the surface code is constructed that suppresses hook errors under realistic noise. An analytical performance model shows that the 3D architecture overcomes the readout error floor and achieves logical error rates orders of magnitude below both standard measurement-based surface codes and 2D mfec variants in regimes with slow, noisy measurements, identifying 3D integration as a key enabler for scalable measurement-free fault tolerance.

Quantum entanglement offers an incredible resource for enhancing the sensing and spectroscopic probes. Here we develop a microscopic theory for the stimulated Raman scattering (SRS) using entangled photons. We demonstrate that the time-energy correlation of the photon pairs can optimize the signal for polyatomic molecules. Our results show that the spectral-line intensity of the entangled-photon SRS (ESRS) is of the same order of magnitude as the one for the entangled two-photon absorption (ETPA); the parameter window is thus identified to do so. Moreover, the vibrational coherence is found to play an important role for enhancing the ESRS against the ETPA intensity. Our work paves a firm road for extending the schemes of molecular spectroscopy with quantum light, based on the observation of the ETPA in experiments.

With the increasing number of high-precision astronomical observations, physical quantities that were previously inaccessible to accurate calculations, such as cosmic birefringence, have once again become a focal point of interest. Such phenomena induce a nonvanishing cross-correlation between the $E$- and $B$-mode polarizations of the cosmic microwave background (CMB), thereby providing a direct observational signature of parity violation. The Chern-Simons coupling between the scalar field in early dark energy (EDE) models and CMB photons is regarded as a plausible mechanism for generating cosmic birefringence. Recent data from the Atacama Cosmology Telescope (ACT) deliver $EB$ measurements at higher multipole moments than those previously achieved by {Planck}, while DESI and PantheonPlus datasets provide new and stringent constraints on the late-time expansion history. Using a joint analysis of {Planck}, DESI DR1, Pantheon+, and ACT data, we perform a full-parameter constraint on the cosmic birefringence effects induced by the EDE-CMB photon coupling. Our results favor a higher Hubble constant, $H_0 = 76.9^{+2.9}_{-2.5}\,\rm km\,s^{-1}\,Mpc^{-1}$, and a relatively large EDE fraction, $f_{\mathrm{EDE}} = 0.232^{+0.074}_{-0.047}$. By comparing the cosmological evolution of this model across different data combinations, we find that the ACT-$EB$ data combined with {Planck} + DESI + PantheonPlus provide good constraints to both early- and late-Universe observations.

Black holes contain more entropy than any other component of the observable universe. Gravitational-wave observations from LIGO and Virgo have shown evidence of a previously unknown black hole mass range, which provides new information to update the entropy budget. Increases in entropy due to binary black hole mergers, as implied in the second law of thermodynamics, should also be added to the budget. In this study, we update the cosmological entropy budget for black holes in the stellar to lite-intermediate-mass range $(5-300~M_\odot)$, originating from either supernovae or binary mergers, by utilizing a suite of population synthesis models and phenomenological fits derived from numerical relativity. We report three new insights: Firstly, the cumulative entropy from merging black holes surpasses the total entropy from cosmic microwave background photons around the onset of the Over-massive Black Hole Galaxy phase at $z\sim 12$, suggesting that mergers played a more significant role in shaping the thermodynamic state of the early universe than relic radiation. Secondly, if primordial black holes constitute a nonzero fraction of dark matter, their early binary mergers establish an ``entropy floor" in the Dark Ages and can dominate the cumulative merger-generated entropy history even for small abundances. Thirdly, by computing the cosmological density parameters, we highlight the thermodynamic asymmetry in black hole mergers, where the production of gravitational-wave energy is inefficient compared to the immense generation of Bekenstein-Hawking entropy.

Quantum state distribution over vast distances is essential for global-scale quantum networks and fundamental test of quantum physics at space scale. While satellite platforms have demonstrated thousand-kilometer entanglement distribution, quantum key distribution and quantum teleportation with ground, future constellations and deep-space missions demand photon sources that are robust, compact, and power-efficient. Integrated photonics offers a scalable solution, yet a critical spectral gap persists. Although telecom-band integrated photon-pair sources are well established, near-visible photons offer distinct advantages for satellite-to-ground links by mitigating diffraction loss and maximizing the collection efficiency of optical telescopes. Scalable integrated sources in this regime have remained elusive due to the fundamental challenge of achieving anomalous dispersion in materials transparent at visible wavelengths. Here we bridge this gap by demonstrating an integrated near-visible photon-pair source based on a wide-bandgap, ultralow-loss, silicon nitride (Si$_3$N$_4$) microresonator. By engineering the dispersion of higher-order waveguide modes, we overcome the intrinsic normal dispersion limit to achieve efficient phase matching. The device exhibits a spectral brightness of 4.87$\times$10$^7$ pairs/s/mW$^2$/GHz and a narrow photon linewidth of 357 MHz. We report high-purity heralded single-photon generation with a heralding rate up to 2.3 MHz and a second-order correlation function as low as 0.0041. Furthermore, we observe energy-time entanglement with 98.4% interference visibility, violating the CHSH limit even at flux exceeding 40.6 million pairs/s. Combined with the proven radiation hardness of Si$_3$N$_4$, this source constitutes a flight-ready hardware foundation for daylight quantum communications and protocols requiring on-orbit multiphoton interference.

Electron paramagnetic resonance (EPR) is a valuable tool for physics, chemistry, biology and medicine, providing complementary spectroscopic information to NMR. It has long been known that EPR at high magnetic fields offers greater spectral resolution, but limitations in the THz instrumentation have prevented the full realization of these opportunities. Here we describe an EPR spectrometer at the high magnetic field of 14 T using 396 GHz excitation, which adapts techniques from liquid-state NMR to obtain sharp EPR resonances with a width of 210 ppb (full-width half-maximum). We use this to measure resonance positions, and hence g-factors, with a precision that reaches $\pm$16 ppb. Our use of in-situ liquid-state NMR of our solvent within the same sample improves the accuracy of these measurements: it allows us to reference our EPR measurement back to dilute gas 3He NMR for which quantum calculations are accurate. We measure the g-factor of N@C60 in deuterated toluene as g = 2.002 099 09 (3), where the 3 in brackets means that the uncertainty on the last digit is $\pm$3.

Rotation and quantum tunneling are fundamental concepts in physics, and their interplay in the ultracold atomic systems is of particular interest. In this theoretical work, we explore how tunneling dynamics in a bosonic Josephson junction are modified when the system is placed in a rotating, non-inertial frame. We show that the tunneling dynamics of ultracold bosons in a two-dimensional double-well potential offer an alternative pathway for inferring the rotation frequency. Using the mean-field and many-body analyses, we demonstrate that rotation strongly modifies the tunneling time period as well as the momentum and angular momentum dynamics. When the rotation axis passes through the center of the double well, the observables show distinct dynamical responses with increasing rotation frequency, enabling the rotation frequency to be assessed from changes in the tunneling dynamics. When the potential is displaced from the rotation axis, the rotation induces asymmetric tunneling and partial self-trapping, allowing both the rotation frequency and the displacement to be inferred. We further show that for an off-centered double well, the tunneling dynamics exhibit a pronounced orientation dependence, enabling the orientation of the double well to be inferred from the observed dynamics. The many-body analysis further shows that the depletion dynamics are strongly influenced by rotation, providing an additional tool for assessing the rotation frequency. Finally, we study the effect of time-dependent rotation in which the double well is gradually set into motion in the laboratory frame and identify distinct dynamical signatures that depend sensitively on the switching time. Together, these results establish a comprehensive framework for inferring the rotation frequency, radial displacement, and orientation directly from the tunneling dynamics.

Computation of Uhlmann fidelity between many-body mixed states generally involves full diagonalization of exponentially large matrices. In this work, we introduce a polynomial-time algorithm to compute certified lower and upper bounds for the fidelity between matrix product density operators (MPDOs). Our method maps the fidelity estimation problem to a variational optimization of sequential quantum circuits, allowing for systematic improvement of the lower bounds by increasing the circuit depth. Complementarily, we obtain certified upper bounds on fidelity by variational lower bounds on the trace distance through the same framework. We demonstrate the power of this approach with two examples: fidelity correlators in critical mixed states, and codeword distinguishability in an approximate quantum error-correcting code. Remarkably, the variational lower bound accurately track the universal scaling behavior of the fidelity with a size-consistent relative error, allowing for the extraction of previously unknown critical exponents. Our results offer an exponential improvement in precision over known moment-based bounds and establish a scalable framework for the verification of many-body quantum systems.

This work presents a routing-aware pruning strategy for quantum circuits executed on Noisy Intermediate-Scale Quantum (NISQ) devices. We propose a method to remove parametric controlled rotations whose small rotation angles do not justify the routing overhead required for their implementation. By selectively pruning such gates, the method mitigates fidelity loss arising from additional SWAP operations introduced during compilation. Our approach evaluates whether executing a gate leads to greater fidelity loss than omitting it. Simulations on benchmark circuits with realistic noise models show that the method reduces two-qubit gate counts (up to 48.6%) while improving final state fidelity (up to 47.7%), especially for larger circuits where routing costs dominate.

Quantifying non-stabilizerness (``magic'') in interacting fermionic systems remains a formidable challenge, particularly for extracting high order correlations from quantum Monte Carlo simulations. In this Letter, we establish the two-point stabilizer Rényi entropy (SRE) and its mutual counterpart as robust, computationally accessible probes for detecting magic in diverse fermionic phases. By deriving local estimators suitable for advanced numerical methods, we demonstrate that these metrics effectively characterize quantum phase transitions: in the one-dimensional spinless $t$-$V$ model, they sharply identify the Luttinger liquid to charge density wave transition, while in the two-dimensional honeycomb lattice via determinant quantum Monte Carlo, they faithfully capture the critical exponents of the Gross-Neveu-Ising universality class. Furthermore, extending our analysis to the fractional quantum Hall regime, we unveil a non-trivial spatial texture of magic in the Laughlin state, revealing signatures of short-range exclusion correlations. Our results validate the two-point SRE as a versatile and sensitive diagnostic, forging a novel link between quantum resource theory, critical phenomena, and topological order in strongly correlated matter.

Fault-tolerant state preparation is essential for reliable quantum error correction, particularly in Steane-type error correction, which relies on robust ancilla states for syndrome readout. One method of fault-tolerant state preparation is to initialize multiple ancilla states and check them against each other to detect problematic errors. In the worst case, the number of states required for successful initialization grows polynomially with the code distance, but it has been shown that this can be reduced to constant ancilla overhead-in the best case, only four states are required. However, existing techniques for finding low-overhead initialization schemes are limited to codes with large symmetry groups, such as the Golay code. In this work, we propose a general, automated synthesis methodology for Steane-type fault-tolerant state preparation circuits that applies to arbitrary Calderbank-Shor-Steane (CSS) codes and does not rely on code symmetries. We apply the proposed methods to various CSS codes up to a distance of seven and simulate the successful fault-tolerant initialization of logical basis states under circuit-level depolarizing noise. The circuits synthesized using the proposed methodology provide an important step towards experimental realizations of high-fidelity ancilla states for near-term demonstration of fault-tolerant quantum computation.

We show that quantum search can be used to compute the hardness to round an elementary function, that is, to determine the minimum working precision required to compute the values of an elementary function correctly rounded to a target precision of $n$ digits for all possible precision-$n$ floating-point inputs in a given interval. For elementary functions $f$ related to the exponential function, quantum search takes time $\tilde O(2^{n/2} \log (1/δ))$ to return, with probability $1-δ$, the hardness to round $f$ over all $n$-bit floating-point inputs in a given binade. For periodic elementary functions in large binades, standalone quantum search yields an asymptotic speedup over the best known classical algorithms and heuristics.

Quantum noise is conventionally viewed as a fundamental obstacle in near-term quantum computing, motivating extensive error correction and mitigation strategies. We present numerical evidence that challenges this consensus. Through experiments on quantum graph neural networks for molecular property prediction, we discover that quantum noise induces heterogeneous, initialization-dependent responses. Among randomly initialized models with identical architecture, approximately one-third show performance improvement under moderate noise, while a smaller fraction deteriorate and the remainder are marginally affected. We identify a strong negative correlation ($r = -0.62$) between baseline model performance and noise benefit, suggesting that noise acts as an implicit regularizer for under-optimized models while disrupting well-converged ones. The observed optimal noise level falls below theoretical predictions, indicating error cancellation in structured quantum circuits. These findings demonstrate that quantum noise effects depend critically on initialization quality and need not be uniformly detrimental, suggesting a shift from universal noise mitigation toward structure- and noise-aware optimization strategies.

Quantum walks represent pillars of quantum dynamics and information processing. They provide a powerful framework for simulating quantum transport, designing search algorithms, and achieving universal quantum computation. Several physical platforms have been employed to implement QWs, such as trapped atoms, trapped ions, nuclear magnetic resonance systems and photonic quantum systems either in bulk optics or waveguide structures and fiber-loop networks. Here we focus on the most promising approach, that is photonic integrated circuits. We will review how the employment of this versatile experimental platform has allowed to explore several phenomena related to QW-based protocols, e.g. the evolution in presence of different kinds of noise. In this landscape, to the best of our knowledge, few examples report on the introduction of absorbing centers and their effects on the coherence of the dynamics. Here we present and discuss the results related to absorbing boundaries in QWs obtained through theoretical simulations and experiments conducted with the universal photonic quantum processors realized by Quix Quantum.

We report on a microscopic theory of quantum friction. Our approach investigates the interplay between the dispersive response and the relative center-of-mass motion of two ground-state atoms. This coupling yields a quantum force, which can be expressed as a power series in the velocity. The significance of each contribution depends on its order parity: while even-order terms are reversible, odd-order terms are irreversible and only survive in the presence of an internal dissipation mechanism. In addition, we obtain general, model-independent properties for the work performed by these contributions for arbitrary scattering trajectories. These results enable an unambiguous identification of odd-parity terms with microscopic quantum friction. At room temperature, the dominant microscopic quantum friction is of first order in the velocity and presents a strong quantum character. Our microscopic theory reveals that several properties of quantum friction obtained in specific settings -- such as the cubic dependence on velocity at zero temperature -- are indeed universal features already present at the atomic scale.

We construct simulation-secure one-time memories (OTM) in the random oracle model, and present a plausible argument for their security against quantum adversaries with bounded and adaptive depth. Our contributions include: (1) A simple scheme where we use only single-qubit Wiesner states and conjunction obfuscation (constructible from LPN): no complex entanglement or quantum cryptography is required. (2) A new POVM bound where e prove that any measurement achieving $(1 - ε)$ success on one basis has conjugate-basis guessing probability at most $\frac{1}{2m} + O(ε^\frac{1}{4})$. (3) Simultation-secure OTMs in the quantum random oracle model where an adversary can only query the random oracle classically. (4) Adaptive depth security where, via an informal application of a lifting theorem from Arora et al., we conjecture security against adversaries with polynomial quantum circuit depth between random oracle queries. Security against adaptive, depth-bounded, quantum adversaries captures many realistic attacks on OTMs built from single-qubit states; our work thus paves the way for practical and truly secure one-time programs. Moreover, depth bounded adaptive adversarial models may allow for encoding one-time memories into error corrected memory states, opening the door to implementations of one-time programs which persist for long periods of time.

The concept of random walk, in which particles or waves undergo multiple collisions with the microscopic constituents of a surrounding medium, is central to understanding diffusive transport across many research areas. However, this paradigm may break down in complex systems, where quantum interference and memory effects render the particle propagation anomalous, often fostering localization. Here we report on the observation of such anomalous dynamics in a minimal setting: an ultracold mass-imbalanced mixture of two fermionic gases in three dimensions. We release light impurities into a gas of heavier atoms and follow their evolution across different collisional regimes. Under strong interspecies interactions, by lowering the temperature we unveil a crossover from normal diffusion to subdiffusion. Simultaneously, a localized fraction of the light gas emerges, displaying no discernible dynamics over hundreds of collisions. Our findings, incompatible with the conventional Fermi-liquid picture, are instead captured by a model of an atom propagating through a (quasi-)static disordered landscape of point-like scatterers. These results highlight the key role of quantum interference in our mixture, which emerges as a versatile platform for exploring disorder-free localization phenomena.

We present a resonant electro-optic transducer for efficient conversion between microwave and telecom wavelength photons. Our platform employs a bulk lithium niobate crystal whose large dielectric constant creates wavelength-scale confinement of microwave photons. By incorporating this crystal within a high-finesse Fabry - Perot optical cavity, microwave photons couple to optical photons through the electro-optic effect. We demonstrate the ability to tune our system into triply resonant operation, where microwave photons, optical pump photons, and upconverted optical photons are simultaneously resonant with high quality factor electromagnetic modes of the system. The device achieves photon number conversion efficiency at the percent level, comparable to state-of-the-art devices at room temperature -- sufficient to resolve the thermal occupation of the microwave mode -- while avoiding the noise and loss associated with metal electrodes. These results establish our all-dielectric devices as a promising platform for high-precision sensing of optically detected microwave fields and as a viable route toward single-photon-level microwave - optical quantum transduction.

Given an array $a[1..n]$, the Range Minimum Query (RMQ) problem is to maintain a data structure that supports RMQ queries: given a range $[l, r]$, find the index of the minimum element among $a[l..r]$, i.e., $\operatorname{argmin}_{i \in [l, r]} a[i]$. In this paper, we propose a quantum data structure that supports RMQ queries and range updates, with an optimal time complexity $\widetilde Θ(\sqrt{nq})$ for performing $q = O(n)$ operations without preprocessing, compared to the classical $\widetildeΘ(n+q)$. As an application, we obtain a time-efficient quantum algorithm for $k$-minimum finding without the use of quantum random access memory.

The binary black hole signal GW250114, the clearest gravitational wave detected to date, offers a unique opportunity to test general relativity in the relativistic strong-gravity regime. How well does GW250114 agree with Einstein's predictions in the plunge-merger-ringdown stage? To address this point, we constrain deviations from general relativity across the plunge-merger-ringdown stage of spin-precessing binaries with a parametrized waveform model within the effective-one-body formalism. We find that deviations from the peak gravitational-wave amplitude and instantaneous frequency of the $(\ell, |m|)=(2,2)$ mode are constrained to about $10\%$ and $4\%$, respectively, at $90\%$ credible level. These constraints are, respectively, two and four times more stringent than those obtained by analyzing GW150914. We also constrain, for the first time, the instantaneous frequency of the $(\ell, |m|)=(4,4)$ mode at merger to about $6\%$, and the time at which the gravitational-wave amplitude peaks to about $5~\mathrm{ms}$. These results are the most precise tests of general relativity in the nonlinear regime to date, and can be employed to constrain extensions of Einsten's theory.

This work aims to present the quantum aspects of a non-commutative gauge gravity formulation of a Schwarzschild-like black hole constructed via the Moyal twist $\partial_t \wedge \partial_θ$. Particle creation is estimated for bosonic and fermionic fields using the quantum tunneling method, with divergent integrals treated through the residue prescription. Since the surface gravity is well defined for this configuration, the corresponding emission rates and evaporation lifetimes are also computed. In addition, previously reported results in the literature on gauge gravity Schwarzschild black holes are revisited. Finally, we infer constraints on the non-commutative parameter $Θ$ from solar-system tests.

Feynman famously said that single-particle interference is ``a phenomenon which is impossible to explain in any classical way, and which has in it the heart of quantum mechanics.'' In this paper we show that some of the phenomenology of interference can be reproduced in a ``classical'' way, by reproducing the Elitzur-Vaidman Bomb Tester (including their improved version) using an extension of the quantum simulation logic (QSL) formalism. Our result improves and simplifies a previous result by Catani \emph{et al}, which relies on a much more complicated extension involving a ``toy field theory.'' We also show that not all single-particle interference can be explained by such a simple extension (including that of Catani et al), by showing that Hofmann's three-path interferometer is ``nonclassical'' in a very specific sense: it violates a Kochen-Specker-noncontextual inequality. Given that both our extension of QSL and Catani et al's extension are \emph{noncontextual} -- so do not reproduce the contextual behaviour of Hofmann's three-path interferometer -- the behaviour of that interferometer is a proper example of a phenomenon that has in it the heart of quantum mechanics, according to Feynman.

We study classical polynomial-time approximation algorithms for the transverse-field Ising model (TFIM) Hamiltonian, allowing a mixture of ferromagnetic and anti-ferromagnetic interactions between pairs of qbits, alongside transverse field terms with arbitrary non-negative weights. Our main results are a series of approximation algorithms (all approximation ratios with respect to the true quantum optimum): (i) a simple maximum of two product state rounding algorithm achieving an approximation ratio $γ\approx 0.71$ , (ii) a strengthened rounding, inspired by the anticommutation property of the two $X_i, Z_iZ_j$ observables achieving ratio $γ\approx 0.7860$, and (iii) a further improvement by interpolation achieving ratio $γ\approx 0.8156$. We also give an explicit (purely ferromagnetic) TFIM instance on three qbits for which every product state achieves at most $169/180\approx 0.9389$ of the true optimum, yielding an upper bound for all algorithms producing product state approximations, even in the purely ferromagnetic case.

We investigate strategies for simulating open quantum systems coupled to dissipative baths by comparing explicit wave function-based discretization [via multi-layer multi-configuration time-dependent Hartree (ML-MCTDH)] and the implicit density matrix-based master equation method [via tree tensor network hierarchical equations of motion (TTN-HEOM)]. For dissipative baths characterized by exponentially decaying bath correlation functions, the implicit discretization approach of HEOM -- rooted in bath correlation function decompositions -- proves significantly more efficient than explicit discretization of the bath into discrete harmonic modes. Explicit methods, like ML-MCTDH, require extensive mode discretization to approximate continuum baths, leading to computational bottlenecks. Case studies for two-level systems and a Fenna--Matthews--Olson complex model highlight TTN-HEOM's superiority in capturing dissipative dynamics with relaxations with a minimal number of auxiliary modes, while the explicit methods are as exact as the HEOM in pure dephasing regimes. This comparison is enabled by the TENSO package, which has both ML-MCTDH and TTN-HEOM implemented using the same computational structure and propagation strategy.

A quantum system (quanton) traverses an interferometer with $N$ equally probable paths and interacts with another quantum system (detector) that stores path information in a set of symmetric states. In this interferometric framework, we present entropic wave-particle duality relations between quantum coherence, characterized by the relative entropy of coherence of the quanton state, and which-path knowledge, quantified by the mutual information obtained through detector-state discrimination. By applying a general optimal discrimination measurement, which has a closed-form solution and encompasses other fundamental strategies as special cases, we provide an exact quantification of which-path knowledge in a variety of scenarios. This measurement is carried out in two steps. First, an optimal separation map with a prescribed separation level $ξ\in [0,1]$ probabilistically reduces the overlaps between the input detector states with maximum success rate, or increases them in case of failure. Then, a minimum-error (ME) measurement discriminates either only the successful outputs (standard approach) or both the successful and failure outputs (concatenated approach). We show that the duality relation is tighter at $ξ=0$, where both approaches reduce to the ME measurement. For $ξ>0$, each approach yields a distinct relation that becomes less tight as $ξ$ increases, with the concatenated one providing the tighter bound. Finally, by using the discrete uncertainty principle, we determine the sets of detector states that lead to saturation of the duality relation, showing that they span $n$-dimensional subspaces of the detector space, where $n$ divides $N$. As a result, nontrivial saturation occurs only for interferometers with a nonprime number of paths. From the identified saturating sets, we highlight how the quanton-detector correlations underlie this phenomenon.

We analyze the propagation and polarization properties of high-frequency gravitational waves in Einstein-Æther theory on vorticity-free and slowly-varying backgrounds at both leading and next-to-leading orders within the geometric optics approximation. The linear perturbation analysis is performed in the background Æther-orthogonal frame, in which the axes of the gravitational wave sound cones remain perpendicular to these hypersurfaces, thereby simplifying the analysis. The leading-order results show that Einstein-Æther theory admits two tensor modes, two vector modes, and one scalar mode, consistent with the findings in the flat spacetime background. We further derive the dispersion relations and linear stability conditions for these modes in curved backgrounds. At next-to-leading order, we obtain the amplitude evolution equations, finding that the graviton number is conserved for the tensor modes but not for the vector and scalar modes. Next-to-leading-order effects also induce mixing among polarization modes. Our study demonstrates that, after imposing the GW170817 constraint on the propagation speed of gravitational waves, the vector modes mixed with the leading-order tensor modes cannot be used to distinguish between general relativity and Einstein-Æther theory. On the other hand, the mixing between scalar modes and the leading-order tensor modes leads to distinct predictions in the two theories, providing a promising avenue to test Einstein-Æther gravity through the detection of polarization mixing in gravitational waves.

This paper describes the reduced phase space of $N=1$, $D=4$ supergravity in the fully off-shell Palatini--Cartan formalism. This is achieved through the KT construction, allowing an explicit description of first-class constraints on the boundary. The corresponding BFV description is obtained, and its relation with the BV one in the bulk is described by employing the BV pushforward in the particular example of a cylindrical spacetime.

We investigate whether the inner product structure of quantum mechanics can be modified without violating fundamental physical principles. We consider a generalized inner product defined by a positive operator and assume local unitary dynamics, existence of entangled states and the no-signalling principle. We show that any nontrivial choice of inner product different from standard one inevitably leads to superluminal signalling, in contradiction with relativistic causality. Therefore, the standard Hilbert-space inner product is uniquely enforced by no-signalling.

In this paper we approach the problem of superdeterminism from a novel point of view, highlighting its character as a more metaphysical than scientific proposition. First, we introduce a distinction between two types of superdeterministic theories, naïve (NSD) and metaphysical (MSD), and argue how NSD presents significant epistemic flaws. We show how NSD justifies itself through claims to fundamentality, thus connoting itself as a metaphysical theory rather than a scientific one. We finally illustrate that the most developed MSD model so far, Invariant Set Theory, implicitly proposes a confused form of priority monism. Our paper thus reinforces the thesis that theories should demonstrate rather than assume fundamentality and that it is methodologically flawed for a theory to assume its own fundamentality for the sole purpose of defending against criticisms.

Electrovac pp--waves in Brinkmann form provide exact Einstein--Maxwell solutions for co--propagating null radiation. Motivated by lensing or scattering, one often ``modulates'' a plane electromagnetic wave by a weak transverse envelope $1+γf(x,y)$. We show that, within the aligned null pp--wave ansatz ($A_v=0$, no $v$--dependence, $F_{xy}=0$) and enforcing the source--free Maxwell equations to $\mathcal O(γ)$, a generic profile $f(x,y)$ is incompatible with Maxwell: the transverse field $F_{ui}$ must be both divergence--free and curl--free on the transverse plane, hence $F_{ui}=\partial_iΦ$ with $Δ_\perpΦ=0$. We give a minimal, polarization--agnostic gauge completion of the modulated potential and prove a cancellation theorem: under standard decay/regularity (or zero--mode) conditions that exclude additional harmonic transverse modes, all $\mathcal O(γ)$ dependence on $f$ drops out of $F_{ui}$ and therefore out of the electrovac source $T_{uu}$. Consequently, the electromagnetic contribution to the Brinkmann profile is universal at $\mathcal O(γ)$: the familiar cycle--averaged isotropic $r^2$ term plus an isotropic oscillatory correction at frequency $2ω$, present only for non-circular polarisation. We isolate the residual Maxwell--admissible freedom as harmonic (holomorphic) transverse data and, by Kerr--Schild linearity, superpose an arbitrary co--propagating vacuum gravitational pp--wave, relating TT--gauge strain to Brinkmann amplitudes. Modelling genuinely localised beams, therefore, requires currents, non-null components, or more general Kundt/gyraton geometries.

In this paper, we present a remarkably simple expression for the exact solution to the $D = 5$, $\mathcal{N} = 2$ supergravity coupled to three vector multiplets with the prepotential $\mathcal{V} = STU -W^2U \equiv 1$, which represents a five-dimensional static Kaluza-Klein black hole with squashed $S^3$ horizons and four independent electric charges. It is asymptotically locally flat and has a spatial infinity $R \times S^1 \hookrightarrow S^2$. We compute its conserved charges via the counterterm method and demonstrate that the thermodynamic quantities satisfy both the first law and Bekenstein-Smarr mass formula, provided the length of the compact extra dimension is treated as a thermodynamic variable.

This paper is a sequel in which we further analyze the recently derived quantum gravity equations which apply in accelerating cosmological spacetimes and whose solutions should be equivalent to all order re-summations of the perturbative leading logarithms that appear. In particular we study their implications concerning the primordial tensor power spectrum and the gravitational force due to a test source.

We demonstrate a robust scheme for quantum information storage based on bound states in a two-dimensional coupled-cavity array. When a target cavity is tuned to resonance with the array, a bound state in the continuum (BIC) emerges, coexisting with two conventional bound states outside the band. The resulting dynamics reflects a delicate interplay between these bound states, which can be fully captured through exact analytical solutions. In the weak-coupling regime, the BIC dominates, enabling perfect and persistent information storage. At stronger coupling, all bound states contribute, leading to oscillatory behavior and reduced storage fidelity. These results, valid at both zero and finite reservoir temperatures and further supported by a single-particle framework, reveal distinctive non-Markovian features in continuous-variable systems and highlight the potential of photonic lattices for scalable all-optical decoherence-free quantum memory platforms.

We initiate the study of quantum Interactive Oracle Proofs (qIOPs), a generalization of both quantum Probabilistically Checkable Proofs and quantum Interactive Proofs, as well as a quantum analogue of classical Interactive Oracle Proofs. In the model of quantum Interactive Oracle Proofs, we allow multiple rounds of quantum interaction between the quantum prover and the quantum verifier, but the verifier has limited access to quantum resources. This includes both queries to the prover's messages and the complexity of the quantum circuits applied by the verifier. The question of whether QMA admits a quantum interactive oracle proof system is a relaxation of the quantum PCP Conjecture. We show the following two main constructions of qIOPs, both of which are unconditional: - We construct a qIOP for QMA in which the verifier shares polynomially many EPR pairs with the prover at the start of the protocol and reads only a constant number of qubits from the prover's messages. - We provide a stronger construction of qIOP for QMA in which the verifier not only reads a constant number of qubits but also operates on a constant number of qubits overall, including those in their private registers. However, in this stronger setting, the communication complexity becomes exponential. This leaves open the question of whether strong qIOPs for QMA, with polynomial communication complexity, exist. As a key component of our construction, we introduce a novel single prover many-qubits test, which may be of independent interest.

We demonstrate nonreciprocal optical transmission for both intense classical fields and weak quantum signals within a reconfigurable optomechanical platform driven by three-mode parametric interactions. The platform is modular, where each three-mode optomechanical system serves as a fundamental building block. Operating independently, a single block achieves nonreciprocity for classical fields. Specifically, asymmetric radiation pressure from intrinsic optomechanical nonlinearity induces nonreciprocal mechanical displacement, modulating the cavity intensity through optomechanical feedback. This enables full isolation of backward transmission without requiring parameter initialization. Alternatively, for quantum signals, the platform is reconfigured by activating photonic and phononic exchange channels between the two blocks. In this configuration, nonreciprocity arises from quantum interference between direct photon hopping and indirect conversion pathways. Constructive interference enables unidirectional low-loss transmission, while destructive interference completely suppresses the reverse direction. After adiabatically eliminating the auxiliary modes, the optimal nonreciprocal frequency and the trade-off between insertion loss and nonreciprocal bandwidth can be controlled by engineering optomechanically induced mechanical dissipation. Additionally, the three-mode-based device requires less control-field power than two-mode systems under resolved-sideband conditions, demonstrating versatile potential for optical nonreciprocity applications across classical and quantum domains.

Molecular nanomagnets are compounds characterized by a high-spin magnetic core that is protected by organic ligands. They have recently gained attention as potential quantum information carriers in solid-state quantum computing platforms, simultaneously exhibiting classical macroscopic properties and quantum features in light of their complex nature and configuration. Addressing the condition when they manifest unquestionable quantum behavior is key to guarantee their effectiveness as resources for quantum information processing. We address the quantumness of molecular nanomagnets using a recently formulated criterion [cf. Krisnanda et al., Phys. Rev. Lett. 119, 120402 (2017)] demonstrating that these systems exhibit an intrinsic quantum nature, as evidenced by their ability to generate and enhance quantum correlations between two non-interacting probes. Our analysis, which is performed addressing various dynamical regimes, paves the way to the design of experimentally viable tests of non-classicality in multipartite registers consisting of ensembles of molecular nanomagnets.

Under unitary evolution, chaotic quantum systems initialized in simple states rapidly develop high complexity, precluding any efficient classical description. Quantum chaos is traditionally characterized by spectral properties of the Hamiltonian, most notably through the spectral form factor, while the hardness of classical simulation within the stabilizer formalism, commonly referred to as quantum magic, can be quantified by the stabilizer Rényi entropy. In this Letter, we propose a relation between the dynamics of the stabilizer Rényi entropy for thermofield double states and the spectral form factor, based on general arguments for chaotic systems with all-to-all interactions. This relation implies that the saturation of the stabilizer Rényi entropy is governed by a first-order dynamical transition. We then demonstrate this relation explicitly in the Sachdev-Ye-Kitaev model, using an auxiliary-spin representation of the stabilizer Rényi entropy that exhibits an emergent $Z_2$ symmetry. We further find that, in the high-temperature regime of the SYK model, the transition occurs at a finite time, with the long-time phase marked by spontaneous $Z_2$ symmetry breaking. In contrast, at low temperatures, the transition is pushed to times exponentially long in the system size. Our results reveal an intriguing interplay between quantum chaos and quantum magic.

We analyze the problem of a free 1D Klein-Fock-Gordon-Majorana (KFGM) particle in an interval. By free, we mean that there is no potential within the interval and that its walls are penetrable; hence, the pertinent energy current density does not vanish at the walls. Certainly, quantization in an interval is not trivial because certain restrictions imposed by the domains of the operators involved arise. Here, our objective is to obtain the Hamiltonian for these particles. In practice, the Feshbach-Villars (FV)--free Hamiltonian is the proper operator for characterizing them and is a function of the momentum operator. Additionally, a Majorana condition must also be imposed on the wavefunctions on which these two operators can act. Thus, we start by calculating the pseudo self-adjoint momentum operator. A three-parameter set of boundary conditions (BCs) constitutes its domain. Up to this point, the domain of the Hamiltonian is induced by the domain of the momentum operator; however, we ensure that only the BCs for which the energy current density has the same value at each end of the interval are in its domain. All these BCs essentially belong to a one-parameter set of BCs. Moreover, because the FV equation is invariant under the operation of parity, the parity-transformed wavefunction is also a solution of this equation, which further restricts the domain of the free FV Hamiltonian. Finally, knowing the most general three-parameter set of BCs for the pseudo self-adjoint FV Hamiltonian for a 1D KFGM particle in an interval, we find that only two BCs can remain within the domain of the FV--free Hamiltonian: the periodic BC and the antiperiodic BC. These BCs are satisfied by both the two-component FV wavefunction, with these components being related, and the one-component KFG wavefunction, which can be real or imaginary.

What ultimately fixes the sample cost of quantum learning -- algorithmic ingenuity or physical law? We study this question in an arena where computation, learning, and causality collide. A twist on Grover's search that reflects about an a priori unknown state can collapse the query complexity from $O(\sqrt{N})$ to $O(\log N)$ over a search space $N$, i.e., an exponential speedup. Yet, standard quantum theory forbids such a unknown-state reflection (no-reflection theorem). We therefore build a state-learning-assisted architecture, called ``amplify-learn,'' which alternates the coherent amplitude amplification with state learning. Embedding this amplify-learn into the Bao-Bouland-Jordan no-signaling framework, we show that the logarithmic-round dream would open a super-luminal communication channel unless each round expends the learning-sample and reflection-circuit budgets scaling at least as $Ω(\sqrt{N}/\log N)$. In parallel, we derive tight computational learning-theoretic sample bounds for learning circuit-generated pure states, revealing a state-universal ansatz ``lock'' at order $N$ in the worst case. The dramatic closure is that no-signaling does not merely veto the unphysical primitive, but it fixes the only consistent reflection-circuit complexity, and feeding this causality-enforced complexity into the computational learning bound makes it collapse onto the very same $\sqrt{N}/\log N$ scaling demanded by no-signaling alone. No-signaling thus acts as a regulator of learnability: a constraint that mediates between physics and computation, welding query, gate, and sample complexities into a single causality-compatible triangle.

Modeling the dynamics of open quantum systems on noisy intermediate-scale quantum (NISQ) devices constitutes a major challenge, as high noise levels and environmental degradations lead to the decay of pure quantum states (decoherence) and energy losses. This situation represents one of the most important problems in the field of quantum information technologies. While existing data-driven methods struggle to generalize beyond the training data (extrapolation), physics-informed neural networks (PINNs) require predefined governing equations, which limit their discovery capability when the underlying physics is incomplete or unknown. In this work, we present the ParaQNN (ParaQuantum neural network) architecture, an equation-free framework for physical discovery. ParaQNN disentangles multi-scale dynamics without relying on a priori laws by employing a dialetheist logic layer that models coherent signal and decoherent noise as independent yet interacting channels. Through extensive benchmark tests performed on Rabi oscillations, Lindblad dynamics, and particularly complex ``mixed regimes'' where relaxation and dephasing processes compete, we show that ParaQNN exhibits a consistent performance advantage compared to Random Forest, XGBoost, and PINN models with incomplete physical information. Unlike its competitors, ParaQNN succeeds in maintaining oscillatory and damping dynamics with high accuracy even in extrapolation regions where training data are unavailable, by ``discovering'' the underlying structural invariants from noisy measurements. These results demonstrate that paraconsistent logic provides a structurally more stable epistemic foundation than classical methods for learning quantum behavior in situations where mathematical equations prove insufficient.

We investigate scalarization of charged quantum Oppenheimer-Snyder extremal (cqOSe)-black hole in the Einstein-Gauss-Bonnet-scalar theory with a nonlinear electrodynamics term. This black hole is described by quantum parameter $α$ and magnetic charge $P$. It is equivalent to the qOS-extremal black hole whose action is still unknown when imposing a relation of $(3αP^2)^{1/4}\to 3M/2$. Focusing on the onset of scalarization, we find the single branch of scalarized cqOS extremal (scqOSe)-black holes. To obtain a scalar cloud (seed) for the single branch, however, we have to consider its near-horizon geometry of the Bertotti-Bobinson (BR) spacetime. In this case, two scalar clouds for positive and negative coupling constant $λ$ are found to represent two branches. Applying Sen's entropy function approach to this theory, we obtain the entropy which is the only physical quantity to describe the scqOSe-black holes. We find that the positive branch is preferred than the negative branch.

Quantum systems governed by time-dependent Hamiltonians pose significant challenges for the accurate computation of unitary time-evolution operators, which are essential for predicting quantum state dynamics. In this work, we introduce a physics-informed deep learning approach based on Physics-Informed Neural Networks to estimate these operators over the full time domain. By incorporating physical constraints such as unitarity and leveraging the second-order Magnus expansion on the evolution operator, the proposed framework enables the estimation of unitary matrices at different time intervals. The model is trained using simulated unitary operators and evaluated on quantum systems ranging from 2 to 6 qubits. For larger many-body systems, specifically those with 7 and 8 qubits, the same methodology is employed to reconstruct an effective time-dependent Hamiltonian, from which the corresponding time-evolution operator is computed over the entire temporal domain. The proposed framework achieves fidelities exceeding 0.92 using a limited number of unitary samples, indicating a potential reduction in measurement and data acquisition costs. These results highlight the effectiveness of the approach for data-driven simulation and identification of quantum dynamical systems, with direct relevance to quantum computing and quantum simulation applications.

We investigate the excitation of quasi-normal modes (QNMs) in gravitational lensing by a Schwarzschild black hole using a scalar field model. By employing a time-domain mode-sum method, we analyze the complex interplay between an incident burst signal and the black hole spacetime. We find that the incident waves can non-resonantly excite a substantial number of high-$l$ modes, with amplitudes for modes as high as l=20 remaining significant compared to the fundamental l=0 mode. We confirm through QNM template fitting that the late-time behaviors of these excited modes are indeed QNMs. After passing through the black hole, we find that the lensed waves form a highly directional and coherent Gaussian beam whose cross-sectional intensity profile is well-described by a Gaussian profile. Unlike spherical waves, this beam's amplitude does not decrease with distance from the black hole but remains nearly constant in the near-field region. Moreover, due to the superposition of numerous QNMs, oscillations largely cancel each other out. The lensed temporal waves do not exhibit typical oscillatory patterns.

In this paper, we briefly present the Monte Carlo event generators BHLUMI and BHWIDE for small and large angle Bhabha scattering, respectively, and discuss possible ways of their improvements in order to satisfy precision needs of future electron-positron colliders.

The one-loop Euclidean partition function on the sphere is known to exhibit a nontrivial phase for massless fields of spin greater than one. Such a phase appears to be in tension with a state counting interpretation of the partition function and its relation to the de Sitter entropy. It has been recently argued that the phase associated with the gravitational path integral can be cancelled by including the contribution of an observer. In this note, we compute the total phase of Vasiliev higher spin gravity on the sphere by summing over the contributions of all spins. We evaluate the resulting infinite sum using two different regularization schemes, obtaining consistent results. We find that for the non-minimal Vasiliev theory, which includes massless fields of all integer spins, the total phase vanishes in all dimensions. This result suggests that the sphere partition function of these theories may be consistent with a counting interpretation, without explicitly including an observer.

DAMSA (DArk Messenger Searches at an Accelerator) is a novel short-baseline accelerator experiment aimed at probing short-lived physics processes, including searches for evidence of a dark sector of particle physics and well-motivated Standard Model signals. Motivated by open questions in neutrino physics and the absence of conclusive evidence for conventional weakly interacting massive particles, DAMSA targets MeV-to-sub-GeV dark-sector messengers with feeble couplings that can be produced in abundance at the PIP-II LINAC. By employing an ultra-short baseline of order one meter, DAMSA is uniquely positioned to overcome the beam-dump "ceiling" that limits sensitivity to promptly decaying particles in longer-baseline experiments. The conceptual design emphasizes a beam-dump production scheme combined with a compact detector optimized for rare decays while mitigating intense neutron-induced backgrounds inherent to high-power proton beams. To validate the experimental strategy and detector technologies, the Little DAMSA Path-Finder (LDPF) proof-of-concept experiment is proposed, focusing on axion-like particles decaying to two photons and operating with 300 MeV electron beams at FAST. Successful realization of LDPF will establish the feasibility of the DAMSA approach, enabling a broad and powerful program to explore short-lived new physics and precision Standard Model processes in a previously inaccessible regime. This conceptual design document outlines the technical details of DAMSA's physics goals, the beam facility proposals, key experimental challenges and how to overcome them, and the proposed experimental staging campaigns.

We provide a study of the parameter space of the complex 2-Higgs Doublet Model (C2HDM), focusing on signs of large CP-violating couplings of the 125 GeV Higgs boson with the fermions. The study is performed utilizing Machine Learning (ML) techniques developed recently for parameter space exploration, including an Evolutionary Strategy Algorithm and Novelty Reward. We give particular attention to the electron electric dipole moment (eEDM). We confirm that the recently found kite diagrams are crucial for the outcome of the analysis. Moreover, their use also mitigates the dependence of the results on the scale and scheme choice of the masses in the loop diagrams. We furthermore point out that, already at the current level of experimental precision, the Barr-Zee diagrams with charm quark loops must be taken into account. The combined use of kite diagrams and ML techniques allows for the resurrection of large fermion CP-odd couplings for Type-II and Flipped C2HDM when the 125 GeV Higgs coincides with the second lightest neutral scalar. This arises due to cancellations, typically of the per-mil order, which, moreover, will still be possible for a foreseeable eEDM precision down to $10^{-33}$ e.cm. For these cases, the constraints on the CP-odd couplings arises from the precision LHC measurements.

According to modern physics, our Universe is baryon-asymmetric. That phenomenon can not be described in the frameworks of the Standard Model of particle physics. Globally, the Universe consists of baryon matter. However, some scenarios can lead to the existence of local antimatter domains. In the research, the chemical evolution of such an isolated antimatter domain, surrounded by baryonic matter, is studied. The size of the domain is estimated according to the conditions of its survival in baryon surrounding, and the process of annihilation at its border is taken into account.

The BESIII Collaboration has collected large data samples from $e^+e^-$ collisions at center-of-mass energies ranging from 1.84 to 4.95 GeV, which include the world's largest charmonium sample, consisting of 10 billion $J/ψ$ and 3 billion $ψ(3686)$ events. These high-statistics datasets enable BESIII to carry out a wide range of studies in hyperon physics. In this article, we review the major achievements of the BESIII Collaboration in this field, which can be broadly categorized into four areas: hyperon polarization and $CP$ violation, rare hyperon decays, hyperon pair production, and hyperon-nucleon interactions.

It is not understood whether scale-separated AdS vacua in string theory admit a holo- graphic dual. A well-known class of such vacua is provided by the DGKT solutions of massive type IIA string theory, where scale separation arises from large fluxes. In this work, we construct a ten-dimensional brane geometry whose near-horizon limit repro- duces the DGKT vacua, using a flux-backtracking approach combined with intersecting D4-brane stacks dual to the unbounded flux sector. We then use this setup to test whether the brane worldvolume theory decouples from the bulk. Modes localised near the branes, deep in the AdS throat, are found to be infinitely redshifted with respect to asymptotic observers. Moreover, an analysis of graviton fluctuations shows the presence of an infinite potential barrier near the branes, providing a direct indication of decoupling. We conclude by comparing these results with recent arguments against decoupling in scale-separated AdS vacua, which focus on the asymptotic region where modes are blueshifted.

We study the partial-wave expansion of residues of five-point tree-amplitude involving identical scalar particles in the external legs. We check the construction using massive spinor-helicity building blocks and by matching to the tree-level five-point Veneziano amplitude at fixed mass levels. As an application, we express five-point splitting constraints - the reduction of the five-point amplitude to products of four-point amplitudes on special kinematic loci - as linear relations among the five-point partial-wave coefficients. At low mass levels these constraints, together with spin truncation, fix the full five-point partial-wave data in terms of the four-point coefficients and imply simple compatibility conditions; remarkably, imposing two independent splitting loci also forces the residue to vanish on their intersection, making the associated hidden zero manifest in partial-wave space. We also show that once both channels allow spin-2 exchange a genuine kernel can remain, indicating the need for additional higher-point input to achieve complete rigidity.

Quantum Chromodynamics (QCD) exhibits complementary descriptions of hadrons: a rest-frame picture based on confinement, chiral symmetry breaking and interquark forces, and a high-energy light-front picture expressed through parton distributions (PDFs,TMDs,GPDs) and form factors. This review develops a unified framework that connects these two domains. It is based mostly on multiple studies by the authors in the past few years. Using the Instanton Liquid Model (ILM) to capture essential nonperturbative features of the QCD vacuum, we derive effective interactions for mesons, baryons, and multiquark states, construct their wave functions in hyperspherical coordinates, and boost them to the light front. The resulting light-front Hamiltonians, incorporating both perturbative and instanton-induced dynamics in the Wilsonian spirit, provide realistic nonperturbative inputs for computing PDFs, DAs, GPDs, quasi-distributions, and gravitational form factors at a well-defined low scale. The connection to perturbative QCD is then established by matching gradient-flow-renormalized operators and LF wave functions to the standard $\overline{\rm MS}$ scheme. Perturbative DGLAP and ERBL evolution then connects these predictions to experimentally accessible regimes. % This approach is applied to quarkonia, glueballs, light mesons, baryons, tetraquarks, pentaquarks, and higher multiquark hadrons, yielding consistent descriptions of both their spectra and partonic structure. Special emphasis is placed on the energy-momentum tensor and the mechanical properties of hadrons, which emerge naturally from the same dynamical ingredients. Overall, the framework demonstrates a clear continuity between hadronic spectroscopy and partonic observables, offering a coherent multiscale picture of hadron structure rooted in the underlying dynamics of QCD.

The measurement of three-dimensional femtoscopic correlations between identical charged kaons (K$^\pm$K$^\pm$) produced in p$-$Pb collisions at center-of-mass energy per nucleon pair $\sqrt{s{_{\rm NN}}} = 5.02$ TeV with ALICE at the LHC is presented for the first time. This measurement, supplementary to those in pp and Pb$-$Pb collisions, allows understanding the particle-production mechanisms at different charged-particle multiplicities and provides information on the dynamics of the source of particles created in p$-$Pb collisions, for which a general consensus does not yet exist. It is shown that the measured source sizes increase with charged-particle multiplicity and decrease with increasing pair transverse momentum. These trends for K$^\pm$K$^\pm$ are similar to the ones observed earlier in identical charged-pion and K$_{\rm s}^{0}$K$_{\rm s}^{0}$ correlations in Pb$-$Pb collisions at various energies and in $π^\pm π^\pm$ correlations in p$-$Pb collisions at $\sqrt{s{_{\rm NN}}} = 5.02$ TeV. At comparable multiplicity, the source sizes measured in p$-$Pb collisions agree within uncertainties with those observed in pp collisions, and there is an indication that they are smaller than those observed in Pb$-$Pb collisions. The obtained results are also compared with predictions from the hadronic interaction model EPOS~3, which tends to underestimate the source size for the most central collisions and agrees with the data for semicentral and peripheral events. Furthermore, the time of maximal emission for kaons is extracted. It turns out to be comparable with the value obtained in highly peripheral Pb$-$Pb collisions at the same energy, indicating that the kaon emission evolution is similar to that in p$-$Pb collisions.

We investigate a dark sector coupled to the Standard Model (SM) through a kinetically mixed dark photon $U$ associated with a new $U(1)'$ gauge symmetry. Kinetic mixing $\varepsilon$ induces an effective coupling to the electromagnetic current, while $U$ interacts with stable dark matter (DM) $χ$ via a dark gauge coupling $g_χ$. Our analysis is based on the parton-hadron-string dynamics (PHSD) transport approach, extended to include dark photon production and decay into dileptons ($U\!\to e^+e^-$). In PHSD, dark photons are produced in high-energy collisions through Dalitz decays of light mesons ($π^0,η,η',ω$), Delta-resonances ($Δ\!\to N U$), direct vector meson decays ($ρ,ω,φ\!\to U$), kaon decays, and $q\bar q\!\to U$ annihilation. Building on previous PHSD benchmarks against dilepton data, we extract upper limits on $\varepsilon^2(m_U,m_χ,α_χ)$ in both the visible regime ($m_U<2m_χ$), where $U\!\to e^+e^-$ dominates, and the invisible regime ($m_U>2m_χ$), where $U\!\toχ\barχ$ is kinematically open. Cosmological and astrophysical constraints are incorporated in two complementary ways. First, we compute the velocity-dependent self-interaction cross section $σ/m_χ$ for Yukawa-mediated SIDM and confront it with bounds from dwarf galaxies, galaxy groups, and clusters. Second, we determine thermal relic target curves by computing the relic abundance and requiring $Ω_{\rm DM}h^2\simeq 0.12$, consistent with \textit{Planck} measurements of the cosmic microwave background. Combining PHSD limits on $\varepsilon^2$ with relic density and self-interaction requirements, we exclude regions of the $(m_χ,m_U)$ plane for each DM realization (Dirac, Majorana, or complex scalar) and identify benchmark scenarios in which heavy-ion, cosmological, and astrophysical constraints are simultaneously satisfied.

The identical charged-kaon correlations induced by quantum-statistics effects and final-state interactions are measured in Pb$-$Pb collisions at $\sqrt{s_{\rm NN}} = 5.02$ TeV. The results of one- (1D) and three-dimensional (3D) analyses show that the obtained system-size parameters (radii) are smaller for more peripheral collisions and decrease with increasing pair transverse momentum $k_{\rm T}$. The 1D parameters agree within uncertainties with those obtained in Pb$-$Pb collisions at $\sqrt{s_{\rm NN}}=2.76$ TeV. The observed power-law dependence of the extracted 3D radii as a function of the pair transverse momentum is a signature of the collective flow in the particle-emitting system created in Pb$-$Pb collisions. This dependence is well reproduced by the integrated hydrokinetic model calculations except for the outward projection of the radius (measured in the longitudinally co-moving system) for the most central collisions. The time of maximal emission for kaons is extracted from the 3D analysis in a wide collision centrality range from 0 to 90%. Its reduction with decreasing charged-particle multiplicity is well reproduced by the hydrokinetic model predictions, and means that kaons are emitted earlier in more peripheral events.

In this paper we have developed general algorithm for finding all orbifolds of Berglund-Hubsch-type Calabi-Yau manifolds and their mirrors. An explicit construction is formulated for finding all admissible deformations and groups defining mirror pairs of orbifolds. Then using our algorithm for one of the Calabi-Yau manifolds, defined by a Fermat-type polynomial, we found all mirror pairs of orbifolds. For this model, for each pair of orbifolds, the number of generations and the number of singlets i.e. particles participating only in gravitational interactions (dark matter particles) were found.

We formulate Carroll hydrodynamics with the inclusion of a spin current. Our strategy relies on the fact that the $c\to 0$ limit of relativistic hydrodynamics yields the equations of Carroll hydrodynamics. Starting with the pre-ultralocal parametrization of the background geometry and the hydrodynamic degrees of freedom for a relativistic fluid endowed with a spin current, the $c\to 0$ limit produces Carroll hydrodynamics with spin. It is known that boost-invariant hydrodynamic models for ultrarelativistic fluids relevant for the physics of quark-gluon plasma, such as Bjorken and Gubser flow, are manifestations of Carroll hydrodynamics under appropriate geometric choices for the underlying Carrollian structure. In this work, we further this mapping between such boost-invariant models and Carroll hydrodynamics, now with the inclusion of a spin current.

We investigate the temperature dependence of the shear viscosity ($η$) and bulk viscosity ($ζ$) of the gluon plasma using lattice QCD over the range 0.76--2.25 $T_c$, extending from below the transition temperature $T_c$ across the transition region and into the deconfined phase. At each temperature, we employ three large, fine lattices, which enables controlled continuum extrapolations of the energy-momentum tensor correlators. Using gradient flow together with a recently developed blocking technique, we achieve percent level precision for these correlators, providing strong constraints for a model-based spectral analysis. Since the inversion to real-time information is intrinsically ill posed, we extract viscosities by fitting spectral functions whose ultraviolet behavior is matched to the best available perturbative result, while the infrared region is described by a Lorentzian transport peak. The dominant modeling uncertainty associated with the transport-peak width is bracketed by varying it over a physically motivated range set by thermal scales. We find that the shear-viscosity-to-entropy-density ratio, $η/s$, exhibits a minimum near the transition temperature $T_c$ and increases for $T>T_c$, whereas the bulk-viscosity-to-entropy-density ratio, $ζ/s$, decreases monotonically over the entire temperature range studied.

We present the sensitivity to Flavor Changing Neutral Currents (FCNC) in interactions involving the top quark at future $e^+e^-$ and $μ^+μ^-$ machines. We consider the $Ztc$ vertex as well as four-fermion contact interactions involving top and charm quarks. To incorporate limits from (HL-)LHC we consider FCNC from Randall-Sundrum models and we recast LHC searches for the resonances that at the microscopic level give rise to the FCNC effects. We determine the maximal strength of the effective FCNC couplings $Ztc$ coupling allowed by LHC. We find that the LHC currently improves on the limit set by previous machines, e.g. LEP indirect sensitivity to heavy vectors. Future improvements of direct searches at HL-LHC may reach a level equivalent to $BR(t\to c Z)\simeq 10^{-6}$. We explore the possibility to probe even smaller FCNC coupling strength using an $e^+e^-$ machine at center-of-mass energy suitable for a Higgs factory $E_{cm}\in$ [200,240] GeV or to probe contact interactions involving top and charm flavors at a high-energy muon collider at $E_{cm}=10$ TeV.

We compute the gluonic contribution to the gravitational form factors of the proton using the improved holographic QCD model, in which the proton is described in terms of bulk Dirac fermions. Model parameters are constrained using lattice and phenomenological input, allowing us to obtain estimates for the gravitational form factors and to compare them with results from other approaches. The resulting $\mathcal{D}(t)$ form factor is found to exhibit an infrared pole in our framework. Using the extracted form factors, we analyze mechanical properties of the proton, including pressure and shear distributions. We obtain estimates of $ρ_\text{mech} = 0.95$ fm and $ρ_\text{mass} = 0.61$ fm for the mechanical and the mass radii of the proton, respectively, which are similar to values in other nonperturbative studies.

This paper presents experimental and simulation data to characterize the Ring Oscillators (RO) produced in 65-nm CMOS technology for the next promising generation of readout chips for the pixel detector in the Inner Tracker (ITk) at the ATLAS experiment at CERN. To enable a better understanding of the RO block embedded in ITkPixV1.1 single chip card (SCC), tests at various temperatures, voltages, accumulated total ionizing dose (TID) with X-ray irradiation, and high-temperature annealing will be presented. The objective of this study is to examine the RO output dependency based on different variable conditions and provide simulation data using Cadence, an electronic design automation (EDA) software to validate the experimental outcomes.

We constrain the $S$-wave baryon-baryon interaction using SU(6) symmetry within a nonrelativistic effective field theory. The most general leading-order Lagrangian contains two independent parameters, which we determine using physical $NN$ and lattice QCD $ΩΩ$ scattering lengths. This framework allows for parameter-free predictions in the strangeness $S=-2$ sector relevant to the $H$ dibaryon. Solving the coupled-channel scattering problem, we identify two bound states below the $ΛΛ$ threshold, one deeply bound and one shallow, along with resonances near the $ΣΣ$ and $Σ^*Σ^*$ thresholds. We demonstrate that these poles result in distinct enhancements in $ΛΛ$ invariant mass distributions, suggesting that the $H$ dibaryon exists as a multichannel bound state and providing clear signatures for experimental verification.

In light of recent precision measurements from BESIII, we reanalyze the $D, D_s \to η^{(\prime)}$ transition form factors using QCD light-cone sum rules, incorporating high-twist contributions and well-established next-to-leading-order QCD corrections. Our analysis confirms the chiral enhancement effect arising from twist-3 light-cone distribution amplitudes of the pseudoscalar mesons, and demonstrates a rapid convergence of the operator product expansion. The resulting high-accuracy form factors enable us to determine the optimal $η$-$η^\prime$ mixing parameters from the precise experimental data for the $D, D_s \to η^{(\prime)} \ell^+ ν_\ell$ (with $\ell = e, μ$) differential decay rates. We find that the BESIII data strongly favor a set of mixing parameters, characterized by small decay constants and a large mixing angle, in the quark flavor basis. Notably, the light-cone-sum-rule predictions for the decays $D \to η^{(\prime)} \ell^+ ν_\ell$, induced by weak $c\to d$ current, reach a precision comparable to the BESIII experimental results. Nevertheless, further refined measurements and more accurate form-factor determinations will be essential to scrutinize the potential role of gluonic components in charmed meson semileptonic decays.

A precision measurement of the $K^{*0}$ meson yield is reported in Au+Au collisions at $\sqrt{s_{NN}} = 7.7,\; 11.5,\; 14.6,\; 19.6,$ and $27~\mathrm{GeV}$ using the high-statistics data sample collected by the STAR experiment during the Beam Energy Scan II (BES-II) program at RHIC. The transeverse momentum ($p_{T}$)-integrated yield ratios $(K^{*0} + \overline{K^{*0}})/(K^{+} + K^{-})$ in central collisions show a suppression relative to peripheral collisions at the $(1.7\text{-}3.6)\,σ$ level, while a thermal model without final-stage rescattering overpredicts this ratio with a deviation of $(6.9\text{-}8.2)\,σ$. These results indicate a loss of the measured $K^{*0}$ signal in central collisions due to re-scattering of its hadronic decay products in the hadronic phase. The $p_{T}$-integrated yield of charged kaons exhibits an approximate scaling with charged-particle multiplicity, independent of collision energy and system size. A similar trend is observed for the short-lived $K^{*0}$ resonance, although significant deviations emerge at lower energies. At BES energies, the $K^{*0}/K$ ratio shows stronger suppression than at the highest RHIC and LHC energies within a given multiplicity bin, particularly in central and mid-central collisions. This behavior is consistent with changes in the effective hadronic interaction cross section and is supported by transport model calculations, which indicate dominant meson-baryon interactions at lower energies and meson-meson interactions at higher energies.

Central exclusive production (CEP) of $η$ and $η'$ mesons in proton-proton collisions at high-energies is discussed. At the LHC the main mechanism for the production of these mesons should be double-pomeron ($\rm I\!P$-$\rm I\!P$) exchange, that is, the fusion reactions ${\rm I\!P I\!P} \to η, η'$. We show that for a scalar pomeron these fusion reactions are not possible. In contrast, in the tensor-pomeron model CEP of $η$ and $η'$ mesons via double-pomeron exchange is allowed. We discuss these reactions for the c.m. energy $\sqrt{s} = 29.1$ GeV realised at the WA102 experiment and for $\sqrt{s} = 13$ TeV corresponding to the LHC experiments. Cross sections and distributions are presented and discussed.

Neutrino physics offers unique insights into phenomena beyond the Standard Model (BSM). This thesis presents phenomenological investigations organized around three pillars: consolidation of the three-flavor oscillation paradigm, exploration of new physics viability, and precise determination of solar neutrino fluxes. The theoretical framework introduces massive neutrinos, leptonic mixing, and flavor transitions, followed by experimental results emphasizing Borexino and NOvA data analyses. The first pillar establishes the three-flavor framework through global analysis of solar, atmospheric, reactor, and accelerator data, providing updated determinations of mixing angles ($θ_{12}$, $θ_{13}$, $θ_{23}$) and mass-squared differences ($Δm^2_{21}$, $Δm^2_{31}$), while quantifying ambiguities in mass ordering and $θ_{23}$ octant. The second pillar investigates Non-Standard Interactions (NSI) with electrons and quarks, combining Borexino data with COHERENT's CE$ν$NS measurements to establish bounds on propagation and detection couplings, excluding viable NSI parameter regions including potential LMA-D solutions. The third pillar advances solar neutrino physics through precision flux determinations, integrating pp-chain and CNO-cycle measurements. Results show preference for high-metallicity Standard Solar Models and incompatibility between $3+1$ mixing parameters favored by Gallium experiments and solar observations. This synthesis guides future experiments toward resolving mass ordering, CP violation, and dark sector interactions.

We examine three exclusive nuclear ground state shell models implemented in the NEUT neutrino event generator and benchmark them against the recent JSNS$^2$ measurement of missing energy using a monoenergetic neutrino source. The nature of the measurement allows a detailed investigation of nuclear ground-state modeling using a neutrino source, and gives access to a direct measurement of the neutron spectral function in a $^{12}$C nucleus. The NEUT intranuclear cascade and nuclear deexcitation \textsc{NucDeEx} are used to simulate inelastic final-state interactions and nuclear deexcitations respectively. We find that the spectral function (SF) models perform better than relativistic mean field models in modeling both the ground state and the tail of the missing energy distribution when the NEUT cascade and nuclear excitation channels are turned on. We also find that taking into account the missing energy threshold for single nucleon knockout interactions results in all nuclear models being accepted based on the obtained $p$-values.

Through a holographic model of QCD, we present a phenomenological approach to study the running of the strong coupling constant α_s in both non-perturbative and perturbative regimes. The renormalization of the metric tensor, driven by the Ricci Flow, and the breaking of conformal and chiral symmetries -thanks to introducing a double dilaton model and large-N_c corrections- allow us to relate the existence of an infrared fixed point in the coupling constant with a smooth matching to pQCD well above 2 GeV. This is done through a model with two fit parameters and one matching point. The proposed dilaton model yields linear Regge trajectories and decay constants for scalar, vector, and tensor meson families similar to their experimental counterparts. We finally study neutral and charged pion form factors to show an application of the running coupling constant obtained.

This paper presents Timepix2 Lite, a compact readout interface for the Timepix2 hybrid pixel detector, designed to provide high spatial and temporal resolution in a wide range of radiation detection applications. The Timepix2 Lite enables simultaneous energy and precise timing measurements on the nanosecond scale and supports advanced real-time data acquisition and visualization through the TrackLab software framework. Its compact and flexible architecture, enabled by the miniaturized design, allows seamless integration into diverse experimental setups, including nuclear spectroscopy, stratospheric balloon missions, and nanosecond-scale half-life measurements. As a case study, the half-life of the second excited state in 237Np was determined to be 67.5(7) ns for the 59.5 keV transition, consistent with previously reported results.

Dedicated R&D efforts are currently underway to couple a thin Cherenkov radiator to Silicon Photomultiplier (SiPM) arrays for precise charged particle Time-of-Flight (ToF) measurements. The prompt nature of Cherenkov radiation makes it an ideal candidate for achieving ultimate timing performance in a ToF detector. Using a thin radiator with a high refractive index, such as fused silica, enables the generation of a fast signal from charged particles that exceed the Cherenkov threshold. A crucial requirement for approaching the target time resolution is the optimization of both the radiator material and thickness, as well as the optical coupling to the SiPM arrays. In this work, we present the main factors that affect the time resolution and the expected performance achieved through a detailed Monte Carlo simulation and the comparison with beam test results.

In this work, we present a novel compact particle identification (PID) detector concept based on Silicon Photomultipliers (SiPMs) optimized to perform combined Ring-Imaging Cherenkov (RICH) and Time-of-Flight (TOF) measurements using a common photodetector layer. The system consists of a Cherenkov radiator layer separated from a photosensitive surface equipped with SiPMs by an expansion gap. A thin glass slab, acting as a second Cherenkov radiator, is coupled to the SiPMs to perform Cherenkov-based charged particle timing measurements. We assembled a small-scale prototype instrumented with various Hamamatsu SiPM array sensors with pixel pitches ranging from 2 to 3 mm and coupled with 1 mm thick fused silica window. The RICH radiator consisted of a 2 cm thick aerogel tile with a refractive index of 1.03 at 400 nm. The prototype was successfully tested in beam test campaigns at the CERN PS T10 beam line with pions and protons. We measured a single-hit angular resolution of about 4 mrad at the Cherenkov angle saturation value and a time resolution better than 50 ps RMS for charged particles with Z = 1. The present technology makes the proposed SiPM-based PID system particularly attractive for space applications due to the limited detector volumes available. In this work, we present beam test results obtained with the detector prototype and we discuss possible configurations optimized for the identification of ions in space applications.

In the present work, the strong decays of the discovered $P_c(4380)$, $P_c(4440)$, $P_c(4457)$ and their possible isospin cousins are systematically studied via the assignment that they are the meson-baryon molecular states. In detail, the strong decay constants, partial decay widths of their decay channels are calculated under the framework of QCD sum rules. The decay withes of the discovered $P_c(4380)$, $P_c(4440)$ and $P_c(4457)$ are in good agreement with the experiments. The predictions of the decays of these three related possible isospin cousins are presented which would shed light for their findings in experiment, in return, this may testify the assignments of the discovered $P_c$ states.

We identify an oscillatory solution that exists as a long-lived, bubble-like closed domain wall in the two-Higgs-doublet model (2HDM) under a $\mathbb{Z}_2$ symmetry constraint, and these structures emerge naturally during the late stages of domain wall decay. \\ \\ The longevity of these structures is attributed to a potential landscape characterized by two distinct vacuum regions, the oscillating region lies in one vacuum, while the constant outer region lies in the other. The lifetime of the structure depends on the parameter in the Lagrangian, we identify a specific parameter space where radiation is suppressed, the solution exhibits a maximum lifetime that goes up to infinity. \\ \\ The simpler two-complex-field system is first used to introduce the mathematical requirements of the structure before extending it to the more physical but complex 2HDM. Further Numerical verification via Floquet analysis shows these structures are stable under perturbation.

We present an exact analytical evaluation of the two-photon exchange (TPE) correction to the elastic lepton-proton differential scattering cross section at low-energies within the framework of heavy-baryon chiral perturbation theory. Our analysis focuses on the kinematic regime relevant to the ongoing MUSE experiment, and we therefore restrict the intermediate states to the dominant elastic channel. All loop integrals are evaluated analytically without approximations. Radiative and chiral recoil contributions of the proton are included, retaining kinematical and dynamical TPE corrections to the cross section through next-to-next-to-leading order [i.e., ${\mathcal O}(α/M^2)$] accuracy in the recoil expansion where $M$ is the proton mass. At this chiral order, pion-loop contributions demonstrate that structure-dependent TPE effects arise through the proton form factors. Our analytical results for the scattering cross section reveal non-vanishing residual proton structure effects of ${\mathcal O}(α/M^2)$, despite substantial cancellations between TPE box and crossed-box contributions. Such effects were entirely absent at this accuracy in our earlier analysis based on the soft-photon approximation. Although the next-to-leading-order contributions are numerically sizable, the next-to-next-to-leading-order TPE corrections are found to be small, thereby indicating that the chiral expansion exhibits reasonably good perturbative convergence.

We use numerical Monte Carlo simulations to determine the mass spectrum of the bosonic sector of the electroweak model in an external magnetic field of the electroweak-scale strength ($10^{20}\,{\rm T}$) at zero temperature. It is known that as the magnetic field gets stronger, the electroweak vacuum undergoes two consecutive crossover-type transitions, passing from (i) the conventional symmetry-broken homogeneous phase to (ii) an intermediate inhomogeneous vortex phase characterized by a (superconducting) condensate of electrically charged $W$ bosons and then to (iii) a homogeneous phase with a restored electroweak symmetry. We show that the spin component of the $W$ boson aligned with the direction of the magnetic field is the lightest excitation in all three phases. Its mass continuously decreases in the low-field broken phase and becomes very small in the intermediate phase. We argue that this nearly massless excitation corresponds to a Goldstone acoustic phonon mode associated with vibrations of the lattice of electroweak vortices. In the high-field symmetry-restored phase, where the vortices disappear, the lightest $W$ mass rises again. Neither Higgs nor $Z$ boson masses vanish across all studied phases and crossover transitions.

We prove that any quantum field theory, or more generally any probability distribution over tempered distributions in $\mathbb{R}^d$, admits a neural network description with a countable infinity of parameters. As an example, we realize the $2d$ Liouville theory as a neural network and numerically compute the three-point function of vertex operators, finding agreement with the DOZZ formula.

We review the main neutrino emission mechanisms operating in dense quark matter under strong magnetic fields, with particular emphasis on conditions expected in the interiors of compact stars. We discuss the direct Urca and neutrino synchrotron processes in unpaired quark matter, incorporating the effects of Landau-level quantization. For the direct Urca process, the quantization of the electron energy spectrum plays a critical role, whereas quark quantization can often be neglected at sufficiently high baryon densities. The resulting field-dependent neutrino emissivity is anisotropic and exhibits an oscillatory behavior as a function of magnetic-field strength. We explore the implications of these effects for magnetar cooling and for possible anisotropic neutrino emission that could contribute to pulsar kicks. In addition, we review the $ν\barν$ synchrotron emission process, which, although subdominant, provides valuable insights into the interplay between magnetic fields and weak interactions in dense quark matter. Overall, our analysis highlights the nontrivial influence of strong magnetic fields on neutrino production in magnetized quark cores, with potential consequences for the thermal and dynamical evolution of compact stars.

Top-quark pair production in heavy-ion collisions provides a unique opportunity to probe nuclear parton distribution functions and study the time evolution of strongly interacting matter, including the quark-gluon plasma. This work presents the observation and measurement of top-quark pair production in both proton--lead (p+Pb) and lead--lead (Pb+Pb) collisions using the ATLAS experiment at the Large Hadron Collider (LHC). In p+Pb collisions at a centre-of-mass energy of 8.16 TeV, top-quark pair production is observed in the lepton+jets and dilepton channels, with significances exceeding 5 standard deviations in each channel. The nuclear modification factor, $R_{p\mathrm{A}}$, is measured for the first time in this process, providing new insights into nuclear parton distribution functions. In Pb+Pb collisions at a centre-of-mass energy of 5.02 TeV, top-quark pair production is studied in the ($eμ$) final state, using datasets recorded in 2015 and 2018 with an integrated luminosity of 1.9 nb$^{-1}$. The measurement achieves a significance of 5.0 standard deviations and is compared to theoretical predictions based on various nuclear PDF sets. These measurements establish top-quark pairs as valuable tools for investigating heavy-ion collisions, offering novel insights into the dynamics of the quark-gluon plasma and nuclear matter.

Conserved currents of relativistic spin fluids derived from microscopic models are known to violate local thermodynamic relations. We present a systematic analysis of pseudo-gauge improvements in ideal spin hydrodynamics and identify a family of pseudo-gauges where standard thermodynamic relations are satisfied. We quantify pseudo-gauge ambiguities in the spin equation of state and derive universal thermodynamic relations that apply to conserved currents in any pseudo-gauge. As an application, we extract the thermodynamic variables and equations of state for free Dirac fermions and scalar fields.

When strongly supercooled cosmological first-order phase transitions (FOPTs) are sufficiently slow, super-horizon inhomogeneities can be generated. We compute these super-horizon curvature perturbations by employing a gauge-invariant, multi-fluid formalism. By resolving the gauge ambiguities inherent in conventional separate-universe simulations, we demonstrate that Primordial Black Holes are unlikely to be produced by these super-horizon inhomogeneities. We also derive a fitting formula for the resulting curvature perturbations and discuss potential observational constraints on FOPTs imposed by limits on primordial curvature perturbations and associated scalar-induced gravitational waves.

We search for a logarithmic 3-form representing the 6-point MHV gravity amplitude, requiring poles only on physical channels and residues matching factorization. Working in the Orlik-Solomon algebra on a De Concini-Procesi wonderful model, we restrict to the S3 x S3 invariant subspace and impose factorization boundary-by-boundary. The intersection of ten compatible 3|3 channels with two compatible 2-particle channels collapses to a unique one-dimensional candidate line. However, enforcing factorization on a crossing channel obstructs any single-valued global representative. We find that an orbit-mixed form, a linear combination over 20 permutation images indexed by bipartitions, satisfies the crossing constraint. This is consistent with a local-system picture: the global gravity form lives in a rank-20 bundle whose monodromy mixes bipartition chambers rather than flipping a single sign. All code and data are provided for reproducibility.

Higgs coupling deviations, at levels accessible to the high-luminosity LHC, can imply a phenomenological no-lose theorem for the next generation of collider facilities. Correlating Higgs coupling deviations from the SM expectation in the gauge boson sector with high-scale unitarity requirements, we estimate and compare the sensitivity that can be expected at a future hadron collider (operating at 100 TeV centre-of-mass energy) and a 10 TeV muon collider. Both muon and hadron colliders offer discovery potential for mass scales up to ${\cal{O}}(6~\text{TeV})$ where unitarity violation induced by (sub)percent Higgs coupling modifications is mended. We comment on how an intermediate precision FCC-ee programme can corroborate such deviations.

The A4 flavor symmetry has provided tremendous insight into the flavor structure of the lepton sector of the Standard Model, predicting a very good approximation to neutrino mixing angles, Tri-Bimaximal Mixing. A4 is spontaneously broken by a scalar called the flavon, and when this happens a number of degenerate vacua can form, resulting in so-called domain walls. These objects are not observed and hence need to be annihilated. This is usually done by explicitly breaking A4 by adding a bias term to the scalar potential. In this paper, we construct a new model invariant under A4 and Z4, which creates cosmologically viable domain walls, lifts the degeneracy of the vacuum giving a natural mechanism for domain walls to annihilate, as well as predicts realistic neutrino mixing angles; all utilizing cross couplings between flavons. The annihilation of the domain walls, with proper choice of wall tension and the consequent bias term, leads to a gravitational wave signal that is potentially detectable in near future gravitational wave experiments, and interestingly intersects with the observed Pulsar Timing Array signal.

We consider Loryons, particles beyond the Standard Model that receive a significant fraction of their masses from electroweak symmetry breaking, in the context of a two Higgs doublet model. Using scalar Loryons in the the $[1,1]$, $[1,3]$ (as well as the equivalent $[3,1]$) and the $[2,2]$ representations of the custodial $SU(2)_L \times SU(2)_R$ global symmetry as benchmarks, we study the constraints on the Loryon parameter space, focusing on unitarity, Higgs decay observables, and the absence of Loryon vacuum expectation values.

The central compact object (CCO) in the Vela Junior supernova remnant is a young neutron star whose relatively low X-ray flux and small distance suggest it has a mass high enough to activate fast neutrino cooling processes. Here we analyse all XMM-Newton MOS and pn and Chandra ACIS-S spectra of the Vela Junior CCO, with observations taking place over the 9 years from 2001 to 2010. We find that the best-fit flux and spectral model parameters do not vary significantly when treating each observation independently, and therefore we fit all the spectra simultaneously using various spectral models to characterize the predominantly thermal emission from the neutron star surface. Our results indicate the Vela Junior CCO has an atmosphere composed of hydrogen, a hot spot temperature (unredshifted) of 3.5x10^6 K, and a colder surface temperature of (6.6-8.8)x10^5 K. Possible absorption lines at ~0.6 keV and 0.9 keV provide evidence for the first-time of an average surface magnetic field B~3x10^10 G for this CCO, which is similar to the magnetic field of other CCOs. At the accurate new Vela Junior distance of 1.4 kpc, the observed luminosity that is dominated by the hot spot is ~5x10^32 erg s^-1. The luminosity from the rest of the colder surface is (1.3-4.0)x10^32 erg s^-1. The cool luminosity and temperature imply the Vela Junior CCO is indeed colder than many other young neutron stars and probably has a high mass that triggered fast neutrino cooling.

We propose a novel origin for the disparity between quark and lepton flavor mixing based on the refractive nature of neutrino masses. We postulate that the fundamental mixing in both the quark and lepton sectors is CKM-like, together with tiny vacuum neutrino masses, while the observed PMNS mixing matrix emerges dynamically from coherent forward scattering of neutrinos on an ultralight dark matter background. The resulting in-medium Hamiltonian rotates CKM mixing angles into large effective lepton mixings, naturally realizing quark--lepton complementarity without invoking new flavor symmetries. This framework links neutrino mass generation, flavor mixing, and dark matter, and predicts environment-dependent neutrino oscillation effects testable in current and future experiments.

Inspired by string theory and cosmological constant problem, it is plausible that the Universe's vacuum structure is characterized by a landscape of metastable vacua. The existence of dark matter and dark energy further suggests that the dark sector may inhabit its own "dark landscape". If the dark vacuum is metastable, bubbles of lower-energy phases can nucleate at an approximately constant rate. Because the Hubble expansion rate is monotonically non-increasing with cosmic time, such nucleation can eventually lead to percolation and completion of a dark-sector phase transition. In this work, we investigate the phenomenological consequences of this transition, focusing on the resulting stochastic gravitational-wave background and the potential formation of primordial black holes. We find that the gravitational wave spectrum peaks at $k_{\mathrm{peak}}=3.1 H_{\mathrm{PT}}$, with an amplitude $Ω_{\mathrm{GW}}^{\mathrm{peak}}\simeq1.5 Ω_γ(Δρ/ρ_{\mathrm{tot}})^2$. Furthermore, the formation of primordial black holes is suppressed due to $ΔN_{\mathrm{eff}}$ constraint.

We derive the QCD$\times$QED factorization theorem governing the leptonic decay $B^-\toμ^-\barν_μ(γ)$ at all orders in $α_s$ and $α$. Electromagnetic corrections to this decay probe multiple scales, which we disentangle through a sequence of effective field theories (EFTs). The resulting state-of-the-art prediction for the photon-vetoed rate includes the complete structure-dependent component and is accurate at the percent level, establishing the theoretical framework required for future high-precision measurements of this channel, which will allow for a clean determination of $|V_{ub}|$ and powerful tests of new physics. Our work presents the first complete study of QED effects to an exclusive $B$-meson decay at next-to-leading power (NLP) in the heavy-quark expansion. Important milestones are (i) the construction of the complete NLP operator basis in soft-collinear effective theory (SCET); (ii) the proposal of a "SCET-friendly" reduction scheme for the Dirac structures of four-fermion operators in dimensional regularization, which avoids power-enhanced evanescent operators; (iii) the consistent refactorization of endpoint-divergent convolution integrals and the first complete resummation of "rapidity logarithms" arising at the boundary between the contributions involving soft and hard-collinear quarks; (iv) the systematic discussion of the EFT below the scale of QCD confinement and the non-perturbative matching of SCET onto this low-energy theory; (v) the decoupling of pseudoscalar mesons in the context of heavy-hadron chiral perturbation theory, so that they can be integrated out for processes in which they do not appear as external particles. We perform a phenomenological analysis of direct and indirect contributions to the decay rate and radiation-energy spectrum, highlighting the importance of the chiral anomaly.

Using the modified Villain lattice Hamiltonian formulation of the 1+1d compact boson theory, we construct exactly solvable abelian chiral lattice gauge theories in two spacetime dimensions. As a concrete example, we derive an explicit quadratic lattice Hamiltonian for the "34-50" chiral gauge theory. We further show that $N$ copies of the modified Villain theory realize the $O(N,N;\mathbb{Z})$ T-duality transformations, which we then use to solve and analyze these lattice gauge theories.

We study the error rate of LLMs on tasks like arithmetic that require a deterministic output, and repetitive processing of tokens drawn from a small set of alternatives. We argue that incorrect predictions arise when small errors in the attention mechanism accumulate to cross a threshold, and use this insight to derive a quantitative two-parameter relationship between the accuracy and the complexity of the task. The two parameters vary with the prompt and the model; they can be interpreted in terms of an elementary noise rate, and the number of plausible erroneous tokens that can be predicted. Our analysis is inspired by an ``effective field theory'' perspective: the LLM's many raw parameters can be reorganized into just two parameters that govern the error rate. We perform extensive empirical tests, using Gemini 2.5 Flash, Gemini 2.5 Pro and DeepSeek R1, and find excellent agreement between the predicted and observed accuracy for a variety of tasks, although we also identify deviations in some cases. Our model provides an alternative to suggestions that errors made by LLMs on long repetitive tasks indicate the ``collapse of reasoning'', or an inability to express ``compositional'' functions. Finally, we show how to construct prompts to reduce the error rate.

A measurement of the fiducial cross section of the associated production of a Z boson and a high-$p_\mathrm{T}$ photon, where the Z decays to two neutrinos, and a search for anomalous triple gauge couplings are reported. The results are based on data collected by the CMS experiment at the LHC in proton-proton collisions at $\sqrt{s}$ = 13 TeV during 2016$-$2018, corresponding to an integrated luminosity of 138 fb$^{-1}$. The fiducial Z$γ$ cross section, where a photon with a $p_\mathrm{T}$ greater than 225 GeV is produced in association with a Z, and the Z decays to a $ν\barν$ pair (Z($ν\barν$)$γ$), is measured to be 23.3$^{+1.4}_{-1.3}$ fb, in agreement, within uncertainties, with the standard model prediction. The differential cross section as a function of the photon $p_\mathrm{T}$ has been measured and compared with standard model predictions computed at next-to-leading and at next-to-next-to-leading order in perturbative quantum chromodynamics. Constraints have been placed on the presence of anomalous couplings that affect the ZZ$γ$ and Z$γγ$ vertex using the $p_\mathrm{T}$ spectrum of the photons. The observed 95% confidence level intervals for $CP$-conserving $h_3^γ$ and $h_4^γ$ are determined to be ($-$3.4, 3.5) $\times$ 10$^{-4}$ and ($-$6.8, 6.8) $\times$ 10$^{-7}$, and for $h_3^\mathrm{Z}$ and $h_4^\mathrm{Z}$ they are ($-$2.2, 2.2) $\times$ 10$^{-4}$ and ($-$4.1, 4.2) $\times$ 10$^{-7}$, respectively. These are the strictest limits to date on $h_3^γ$, $h_3^\mathrm{Z}$ and $h_4^\mathrm{Z}$.

The Belle and Belle II experiments have collected an 1.6 ab$^{-1}$ sample of $e^+e^-$ collision data at center-of-mass energies near the $Υ(nS)$ resonances. In particular, the Belle II experiment collected a 19.2 fb$^{-1}$ sample of data at center-of-mass energies near the $Υ(10753)$ resonance. We study the following processes: $e^+e^-\to Υ(nS)η$, $e^+e^-\to γX_b(χ_{bJ}π^+π^-)$, and $e^{+}e^{-}\toχ_{bJ}(1P) γ$. These results provide additional information about the nature of the $Υ(10753)$ resonance and nearby structures. In addition, we measure the $B^{0}$ and $B^+$ meson mass difference, and $σ\left(e^+ e^-\to J/ψp\bar{p}\right)$ over a range of center-of-mass energies accessed via initial-state radiation.

Lepton flavor violating (LFV) interactions involving dark matter (DM) particles remain a largely unexplored area. In this study, we systematically investigate LFV DM interactions within the framework of effective field theories by analyzing astrophysical photons and positrons produced from DM annihilation. Employing the astrophysical photon and positron data collected by Fermi-LAT, INTEGRAL, XMM-Newton, and AMS-02, we place meaningful constraints on all leading-order effective operators involving a DM pair and a flavor violating charged lepton pair. Our analysis covers the three well-known DM candidates: a scalar, a fermion, and a vector particle. For the photon flux, we consider contributions from final-state radiation, radiative decay, and inverse Compton scattering, and examine their respective sensitivity regions across different DM masses and photon energies. We find that for DM masses below $\mathcal{O}(20\,\rm GeV)$, INTEGRAL provides the most stringent constraints on annihilation cross sections and effective operators in all three LFV channels, whereas AMS-02 offers the strongest constraints above $\mathcal{O}(20\,\rm GeV)$.

We investigate the frame dependence of distribution functions within the framework of generalized chiral kinetic theory. Based on the derived transformation rules governing the choice of frame, we analytically obtain the global equilibrium solution in the presence of vorticity and electromagnetic fields. Our results show that, under the assumption of a varying electromagnetic field, these equilibrium solutions can be uniquely determined.

We present the first dedicated analysis of cosmic-ray dark matter (CRDM) in paleo detectors. Owing to their large kinetic energies, CRDM particles generate nuclear-recoil tracks that extend to substantially larger lengths than those produced by dominant backgrounds from neutrinos and intrinsic radioactivity. Combined with the ultra-large effective geological exposure of $\mathcal{O}(10^{5})~\mathrm{t\,yr}$, paleo detectors provide a uniquely sensitive probe of sub-GeV DM. Considering both constant and vector-mediator interactions, we find that paleo detectors improve the sensitivity to the DM--proton scattering cross section by one to two orders of magnitude compared with the latest XENONnT limits.

We present the first systematic study of rigidly rotating protoquark stars based on isentropic equations of state (EOS) within the density-dependent quark mass (DDQM) framework. Using a quasi-static equilibrium approach, we follow the Kelvin--Helmholtz evolution from hot, lepton-rich matter to a cold, catalyzed quark star. Rotation substantially enhances the maximum stable mass (by up to $\sim 40\%$), equatorial radius, and key rotational observables, with the ratio of rotational kinetic to gravitational potential energy, $T_{\rm kin}/|W|$, reaching $0.18$--$0.19$ near the Keplerian limit, indicating a heightened susceptibility to gravitational-wave--emitting instabilities. Thermal evolution introduces a clear ordering: all stellar properties peak during the lepton-rich stages and decrease monotonically as the star cools. Compared to hadronic stars, rotating protoquark stars exhibit larger radii, higher moments of inertia, and stronger quadrupolar deformation, producing a distinct signature in the mass--radius--spin plane that can accommodate objects such as HESS~J1731--347 and PSR~J0740+6620. These results demonstrate that future multimessenger observations must account for both thermal history and rotation to robustly identify quark matter in compact stars.

We evaluate the QED coupling in the gradient-flow scheme in three and four space-time dimensions. Our general result applies to any theory with a U(1) gauge field coupled to arbitary other fields via arbitrary interactions. As an example, we consider QED with $n_\text{f}$ flavors in three and four space-time dimensions and evaluate the corresponding $β$ functions. In four dimensions, we find that the perturbative expansion of the $β$ function behaves much better than the corresponding expression in QCD. In three dimensions, we recover both the ultraviolet as well as the infrared fixed points of the QED coupling in the large-$n_\text{f}$ limit.

We present a unified overview of recent progress in the study of QCD radiation in heavy-quark jets, focusing on the dead-cone effect. Using precision data from LEP at $\sqrt{s}=91.2$~GeV, we demonstrate strong momentum-space suppression in charm and bottom quark jets, supported by Monte Carlo simulations with \textsc{Pythia}8, and provide a quantitative interpretation within the Modified Leading Logarithmic Approximation (MLLA) of perturbative QCD. We then extend the analysis to top-quark jets at $\sqrt{s}=1$~TeV, where finite lifetime effects and decay radiation introduce new conceptual challenges. A new method is presented to isolate the top-quark dead cone by separating production and decay radiation, and it is validated at both parton and hadron level using \textsc{Pythia}8. Together, these results establish a coherent framework for testing QCD radiation dynamics across all three heavy quarks.

We calculate the 1-loop $γZ$ box-graph correction to electron-quark scattering at low energy and low momentum transfer. Both electron and quark masses are kept non-zero. From our result, we extract coupling constants for the low-energy effective Lagrangian with parity-violating 4-fermion interaction terms. We study the zero-mass limits and show that a non-zero electron mass is sufficient to obtain finite, well-defined couplings which are insensitive to a hadronic mass cutoff. We finally discuss the impact of our results on the determination of the weak charge of the proton from polarized electron-proton scattering.

We investigate the dynamically generated $h_1$ state with spin-parity $J^P = 1^+$ and a mass around 1790~MeV, arising from the $K^* \bar{K}^*$ interaction within the chiral unitary approach. The partial decay widths into the $K_1(1270)\bar{K}$ and $b_1(1235)π$ channels are calculated via a triangular loop mechanism. In this mechanism, the $h_1$ state couples to $K^* \bar{K}^*$, and final-state interactions between $K^*$ and $\bar{K}^*$ proceed through pseudoscalar-meson exchange, leading to the final states $\bar{K}$ (or $π$) and $K_1(1270)$ [or $b_1(1235)$]. We also present the invariant mass distributions of a vector meson and a pseudoscalar meson originating from the decays of $K_1(1270)$ or $b_1(1235)$, along with the corresponding decay widths. Our results show that these decay widths are all of the order of a few MeV. We hope that future experiments can test the predictions presented here, thereby helping to identify this $h_1$ state.

We study the isentropic processes in a class of Anti de Sitter black holes coupled to non-linear electrodynamics. We demonstrate that such processes are classically forbidden but can proceed via quantum mechanical tunnelling. We compute the Euclidean action associated with the tunnelling process and analyze its dependence on the black hole charge, horizon radius, and the non-linear electrodynamics parameters characterizing each model. We find that the tunnelling probability is increasingly suppressed as the strength of the non-linearity is enhanced. We further find that smaller black holes exhibit a significantly higher tunnelling probability compared to larger ones, indicating a departure from classical behaviour. We conjecture that this behaviour may be universal across a broad class of black hole spacetimes. We discuss the implications of our results for entropy bounds and their potential relevance to the black hole information loss paradox.

We study the structure of bound states appearing in systems governed by the Coulomb and short-range interactions. We analyze the binding energies and wave functions of the bound states generated by the Coulomb plus short-range potential. We demonstrate that Coulomb-induced shifts of the binding energy are closely correlated with the spatial distribution of the wave function. Furthermore, we show that the asymptotic behavior of wave functions of weakly bound states is qualitatively altered by Coulomb repulsion, leading to a modification of the near-threshold mass scaling that is otherwise universal for short-range interactions.

We present a systematic quantum encoding framework for leptophilic extensions of the Standard Model, tailored to quantum simulation applications on near term and future quantum devices. Focusing on anomaly free $U(1)'_{\ell}$ gauge theories, we show that the leptonic charge structure admits a natural and scalable representation on qubit registers, where gauge symmetries and anomaly cancellation conditions are enforced directly at the level of quantum states. Within this framework, gauge invariant operators are mapped to unitary quantum circuits, ensuring the preservation of gauge symmetry under quantum evolution. As a proof of principle, we construct explicit circuits that encode scattering processes mediated by a leptophilic gauge boson $Z'_{\ell}$. Our results establish a reusable bridge between beyond the Standard Model gauge theories and quantum information science, providing a concrete pathway for simulating leptophilic gauge sectors within emerging quantum computing architectures.

We present a theoretical analysis of the process $Λ_c^+ \to p\bar{K}^0 π^0$ within the chiral unitary approach, with particular emphasis on the dynamically generated $N(1535)$ resonance. In addition to $N(1535)$, our model incorporates contributions from other intermediate resonances including $N(1650)$, $K^*(892)$, $K_0^*(1430)$, $N(1440)$, and $Σ(1750)$. The calculated invariant mass distributions and Dalitz plot are in good agreement with the recent Belle measurements. Our analysis highlights the crucial role of $N(1535)$ state in this decay channel and supports its interpretation as a dynamically generated state arising from coupled-channel meson-baryon interactions.

At LHC energies, the Drell--Yan ($Z/γ^{*}$) processes have a substantially large cross section. Their di-lepton ($\ell^+\ell^-$) final state contributes significantly to many resonant signal regions, making them one of the dominant backgrounds in numerous physics analyses. The study focuses on improving the discrimination and suppression of the $Z/γ^{*} \rightarrow \ell^{+}\ell^{-}$ background from the $H \rightarrow Zγ\rightarrow \ell^{+}\ell^{-}γ$ signal at $\sqrt{s}=13~\text{TeV}$ by leveraging Monte Carlo simulated data. The analysis introduces physics-motivated correlated observables derived from the two-dimensional $(P_{\mathrm{Higgs}}, θ_{Zγ})$ plane. These observables encode differences in angular and momentum information to enhance signal--background separation while maintaining high signal efficiency. We present a multivariate analysis (MVA) employing a Boosted Decision Tree (XGBoost) classifier. By incorporating additional physics-motivated correlated observables, the classifier achieves measurable improvements in performance. A significant increase in the area under the ROC curve (AUC) is observed in both the electron and muon channels, demonstrating the effectiveness of the expanded feature set. Further, optimised background rejection using $(P_{\mathrm{Higgs}}, θ_{Zγ})$ plane increases the signal-to-background ratio to 2.1\% and 3.4\% for the electron and muon channel respectively near the Higgs mass. This work demonstrates that combining kinematic correlations with interpretable multivariate techniques leads to improved sensitivity and robust background rejection. The approach is flexible and can be readily applied to a wide range of analyses, including rare Higgs decays, resonant searches, and studies beyond the Standard Model.

Within the framework of the extra-dimensional Randall-Sundrum set-up, we investigate the freeze-in production of Standard Model (SM) gauge-singlet scalar, fermionic, and massive vector dark matter (DM). Assuming that both the DM and SM fields reside on the IR brane and interact solely through the graviton and radion, we demonstrate that the observed DM relic abundance measured by Planck can be achieved across a wide range of reheating temperatures, all while naturally addressing the hierarchy problem, satisfying constraints from collider, early Universe cosmology including $Δ{N}_{\rm eff}$. We further show that the same set-up can accommodate TeV-scale leptogenesis capable of generating the observed baryon asymmetry of the Universe. Remarkably, we find that current graviton searches at the Large Hadron Collider (LHC) already impose strong constraints on the reheating temperature in this scenario.

The theoretical interpretation of jet observables in heavy-ion collisions is a complex task due to the intricate interplay of perturbative and non-perturbative effects. One way to reduce this complexity is to groom away soft, wide-angle radiation so that perturbative dynamics dominates. Even in this simplified scenario, there are competing explanations for the physical origin of the measured medium-induced modifications. In this paper, we present a minimal approach to compute groomed substructure observables. The core idea is to treat medium effects as an effective energy shift of the hard, vacuum-like substructure. This energy shift includes a gradual onset of colour decoherence effects and thus depends on the jet substructure itself. We first study a NLO-exact dijet configuration in vacuum and apply radiative energy-loss to the two subjets. We find that this minimal setup already captures the narrowing trend of groomed observables but it's not able to quantitatively describe the existing data. Next, we match the NLO matrix-element to a leading-logarithm accurate parton shower and perform a clustering algorithm to recover a two-prong system to which we again apply the energy-loss distribution. Despite its simplicity, the model results in a very good theory-to-data agreement (within $10\%$) for a broad range of observables including both ALICE and ATLAS kinematics. We also examine the discriminating power of groomed jet data in terms of colour decoherence effects and find that substructure-dependent energy loss yields an overall better agreement.

Scaling laws describing the dependence of neural network performance on the amount of training data, the spent compute, and the network size have emerged across a huge variety of machine learning task and datasets. In this work, we systematically investigate these scaling laws in the context of amplitude surrogates for particle physics. We show that the scaling coefficients are connected to the number of external particles of the process. Our results demonstrate that scaling laws are a useful tool to achieve desired precision targets.

The production of multiple jets or vector bosons in association with jets at the LHC provides a unique testing ground for Quantum Chromodynamics (QCD) in the high-energy regime. With the increasing precision of the ATLAS measurements, detailed studies have become possible on observables that probe different aspects of QCD, such as the topological configurations between vector bosons and jets, jet substructure features, and heavy-flavor jet contributions. These measurements also play a key role in improving the precision of the determination of the strong coupling constant. Recent ATLAS results in these areas are presented, offering new insights into QCD dynamics and the performance of state-of-the-art theoretical predictions.

The Higgs self-couplings remain only weakly constrained by current Large Hadron Collider (LHC) measurements, leaving ample room for physics beyond the Standard Model that could modify the structure of the Higgs potential. Multi-Higgs production processes provide a particularly sensitive probe of deviations in both the Higgs trilinear and quartic self-couplings. In this note, we summarize the current status of next-to-leading-order electroweak (EW) corrections to double-Higgs production computed within the Standard Model Effective Field Theory and Higgs Effective Field Theory frameworks, emphasizing how these calculations introduce sensitivity to the Higgs self-couplings beyond what is accessible at leading order. We discuss the key conceptual and technical differences between the two effective field theory approaches, including their treatment of higher-dimensional operators, renormalization procedures, and the structure of EW two-loop amplitudes. Despite these differences, both approaches yield broadly consistent constraints, illustrating the complementarity of double- and triple-Higgs measurements. With the high-luminosity LHC and future high-energy colliders on the horizon, these developments and further advances provide an essential foundation for extracting increasingly precise information on the dynamics of EW symmetry breaking.

The full exploitation of the physics potential of a multi-TeV muon collider will ultimately lie in the detector's ability to cope with unprecedented levels of machine-induced backgrounds. This contribution introduces the MUSIC (MUon System for Interesting Collisions) detector concept and presents its performance in the context of $\sqrt{s}$ = 10 TeV muon-antimuon collisions. The MUSIC detector is designed to mitigate machine-induced background effects while maintaining high efficiency and accuracy in the reconstruction of physics events, in particular in the Higgs boson sector and in the search for new physics. It features an all-silicon tracking system, a semi-homogeneous lead-fluorite crystal electromagnetic calorimeter, an iron-scintillator sampling hadronic calorimeter, and a superconducting magnet providing a 5 T magnetic field. Detailed detector simulations, accounting for the dominant machine-induced backgrounds, demonstrate promising performance in track, muon, photon, electron, and jet reconstruction, as well as jet flavor identification, highlighting the detector's strong potential for high-energy muon collider experiments.

The high-energy proton-proton collisions at the Large Hadron Collider provide the ideal conditions to study the rare processes predicted by the Standard Model (SM) such as the production of multiple electroweak bosons. These processes involve the self-interactions of the gauge bosons through triple and quartic gauge couplings (TGCs and QGCs), in addition to interactions with the Higgs boson. Therefore, precision measurements of multi-boson final states probe the electroweak symmetry breaking mechanism and allow to search for deviations from the SM. Four recently published ATLAS results are summarised in this document. The first two results are measurements of the production of di-boson final states in association with jets: the observation of $WW/WZ/ZZ$ production in semileptonic final states is described, and a measurement of the polarisation states of same-sign $W$ pairs is also discussed. The second set of measurements are related to the triple production of electroweak bosons: the first analysis reports the observation of $WWγ$ in the leptonic final state, while the second result presents the observation of triple vector boson production where at least one of the produced bosons is a $Z$ boson. All the measurements use datasets collected at a centre-of-mass energy of 13~TeV and corresponding to an integrated luminosity of 140~fb$^{-1}$.

We investigate a minimal extension of the Standard Model with a real singlet scalar and a singlet Dirac fermion acting as dark matter. Unlike a conventional singlet scalar setup, we assume that the singlet scalar does not acquire a vacuum expectation value at zero temperature. This decouples the scalar mixing angle from the Higgs-portal quartic coupling responsible for the strong first-order electroweak phase transition, allowing it to coexist with current collider and direct-detection constraints. The Higgs-singlet mixing is generated independently through a trilinear portal interaction. We check theoretical consistency conditions, various LHC limits on heavy scalar resonances, dark matter relic abundance, and direct detection bounds to delineate the viable parameter space. We perform a state-of-the-art analysis of the electroweak phase transition, which is shown to be achieved for a few benchmark points. We further compute the resulting stochastic gravitational wave spectra and find that several scenarios yield signals potentially observable at future space-based interferometers. Our results establish a unified and testable framework that connects collider phenomenology, first-order electroweak phase transition, and the resulting production of gravitational waves, along with the dark matter phenomenology, all within a simple renormalizable extension of the Standard Model.

For a quantum field of arbitrary mass and spin in the static patch of de Sitter spacetime, the Euclidean partition function receives contributions from edge modes localized on the horizon, expressible in terms of the Harish-Chandra character of the de Sitter group. Considering the flat limit and summing over all string fields, we obtain the partition function of edge modes in string theory near the Minkowski-Rindler horizon. Application of the Kronecker limit formula naturally yields a modular invariant one-loop partition function. The resulting expression generalizes the edge contribution of a massive vector boson in a spontaneously broken gauge theory to the infinite tower in string theory. It is naturally ultraviolet finite and amenable to a state-counting interpretation.

A reinterpretation of four-top quark ($t\bar{t}t\bar{t}$) production is presented using the full Run 2 dataset recorded by the CMS experiment, corresponding to an integrated luminosity of 138 fb$^{-1}$. The analysis targets BSM scenarios using the existing $t\bar{t}t\bar{t}$ production measurement, including constraints on effective field theory (EFT) operators, top-philic heavy resonances, and the top-Yukawa coupling. The results provide competitive limits on several new physics models and demonstrate the sensitivity of multi-top final states to a wide range of BSM effects.

Collisions among heavy ions, like Pb or Au, are a great tool to study the theory of strong interactions, that is Quantum Chromodynamics (QCD). In particular, these experiments are able to give insights on all the complex phases of matter that the theory of QCD allows. In this PhD Thesis we have investigated the initial stages of proton-ion collisions: in particular, we will focus on the first $\sim 0.4$ fm/c ($\sim 10^{-24}$ s) after the collision, which are dominated by very intense gluon fields, in a state called glasma. We investigated the effect of such fields on the dynamics of heavy quarks (charm and beauty) which are created and evolve in this medium. The effect of the initial gluon fields on heavy quarks is quite substantial, in particular we observe that the glasma provokes a $50\%$ dissociation rate on quark-antiquark pairs. Moreover, glasma fields have a large momentum anisotropy, and transmit a large part of such anisotropy to the heavy quarks which evolve in this medium. Finally, we have generalized our study to a non-boost invariant medium, and shown that fluctuations in rapidity do not lead to significant isotropization within glasma timescales.

We present the first lattice-QCD validation of multiple sum rules associated with quark-gluon decomposition of hadron mass by computing all components from first principles. We achieve this through nonperturbative renormalization of the QCD energy-momentum tensor, including its trace, in a gradient-flow scheme, followed by continuum extrapolations, two-loop matching to the $\overline{\mathrm{MS}}$ scheme, and zero-flow-time extrapolations. These ingredients enable a direct and simultaneous verification, in a common renormalization scheme and scale, of multiple energy-density-based and trace-based mass decomposition sum rules proposed in the literature. We demonstrate the framework for the $η_c$ and $J/ψ$ charmonia using three fine lattice spacings with a physical strange-quark and near-physical up- and down-quark masses. We present the first lattice-QCD results for the gravitational form factor $\bar{C}$. We find sizable gluonic contributions to charmonia masses at the hadronic scale, $\sim 15\%$ in the Lorcé and Metz-Pasquini-Rodini decompositions. The trace-anomaly contribution in the Ji sum rule is $\sim 6\%$, while the gluonic component of the trace anomaly in the Hatta-Rajan-Tanaka sum rule is $\sim 35\%$. The method is general and can be straightforwardly adopted for lattice-QCD calculations of mass and spin decompositions as well as gravitational form factors of other hadrons and nuclei.

Inflationary cosmology successfully accounts for the observed properties of primordial fluctuations using quantum field theory in an expanding background. However, the quantum nature of these fluctuations has not been experimentally established, since classical stochastic models could reproduce the observed two-point statistics by construction. Existing approaches to testing primordial quantumness focus primarily on Bell inequalities, which provide a sharp conceptual criterion but are difficult to implement with cosmological observables. In this work we adopt a falsification-based approach. We define a precise classical hypothesis for the origin of primordial perturbations (local stochastic fields admitting a positive probability distribution) and identify inequality constraints that must be satisfied within this class. We show how violations of these classicality inequalities can be probed using realistic cosmological observables, without invoking Bell tests or non-commuting measurement settings. We further identify symmetry-protected spectator sectors in which quantum coherence is parametrically preserved during inflation, allowing violations of observable magnitude to survive decoherence. Our results show that large-scale structure and future 21 cm surveys provide a viable and quantitative route to falsifying classical stochastic descriptions of primordial fluctuations.

Various theories beyond the Standard Model predict unusual signatures, including new, long-lived particles decaying at a significant distance from the collision point. These unique signatures are difficult to reconstruct and face unusual and challenging backgrounds. This contribution will focus on recent results from searches for unusual signatures motivated by dark sector models, utilising pp collision data collected by the ATLAS experiment at the Large Hadron Collider.

We demonstrate that every model for self-dual nonlinear electrodynamics in four dimensions has a $\mathsf{U}(1)$ duality-invariant extension to $4p>4$ dimensions.

We present a systematic framework for calculating the vacuum decay rate in D-dimensional electroweak theories, providing a unified treatment of quantum fluctuations for scalar, fermion, and gauge boson fields via a combined WKB expansion and dimensional regularization. This method ensures rapid convergence even at large angular momenta. Application to a $D=4$ standard model effective field theory gives $\log_{10}(γ\times \text{Gyr} \times \text{Gpc}^3)=183$. In the finite-temperature, dimensionally reduced standard model effective field theory with $D=3$, the values are $\log_{10}(Γ_T \times \text{Gyr} \times \text{Gpc}^3)=42.4$ at tree level and $116.1$ at two-loop order. The approach offers a general and efficient tool for analyzing vacuum stability and decay across dimensions.

In this paper, we review recent highlights in heavy-ion collisions and proton-proton collisions at top energies from STAR experiment at the Relativistic Heavy Ion Collider (RHIC) with key contributions from Chinese groups, including the Quark-Gluon Plasma (QGP) bulk properties, electromagnetic probes, heavy flavor and jets, antimatter hyper-nucleus, nuclear structure, global polarization, and nucleon spin structure. These data serve as important ingredients in the physics of Quantum Chromodynamics (QCD).

The $N=1$ supersymmetric Yang--Mills--Carroll--Field--Jackiw (SYM--CFJ) model is extended to include the quark sector of supersymmetric quantum chromodynamics (SQCD) in the presence of Lorentz--symmetry violation (LSV). The Lorentz--violating data are carried by a spurion chiral superfield whose components define a purely spacelike background vector $v_μ$ and fermionic bilinears, inducing a topological mass scale in the gauge sector. Working within a spurion effective field theory (EFT) and a small--LSV expansion, we classify the allowed parametric dependence of glueball observables on the induced mass scales. Using lattice Yang--Mills (YM) glueball masses only as a reference hadronic scale, we provide parametric sensitivity estimates and naturalness ranges for the topological mass $|v|$ relative to $Λ_{\rm YM}$.

In this work, we have conducted a comprehensive and systematic theoretical investigation of the full cascade decays of charged pseudoscalar mesons, specifically $D_s$, $D$, $B$, and $B_c$, into $τν_τ$, followed by the subsequent decay of the $τ$ via its dominant experimentally reconstructible channels: $τ\to πν_τ$, $τ\to ρν_τ$, $τ\to e \barν_e ν_τ$, and $τ\to μ\barν_μν_τ$. Our study is framed within the model-independent low-energy effective field theory approach, which incorporates the most general set of four-fermion operators, including those coupling to right-handed neutrinos. We provide precise Standard Model predictions for differential decay rate as a function of the final-state charged particle energy, develop an innovative and robust methodology for extracting the magnitudes of the new physics couplings using energy moments, and identify and characterize fixed points in the normalized energy distributions. The fixed points are invariant under new physics contributions described by the considered effective Hamiltonian.

Using instanton partition function for five dimensional $U(1)$ gauge theory with eight supercharges and a single adjoint massive hypermultiplet on the $Ω$ background, we give explicit expression for non-perturbative corrections to the topological string theory in the holomorphic limit. It was argued that in this case the theory is compactified on the twisted affine line bundle over $\mathbb{C}\times T^2$. We perform calculations in two ways. First we modify the integration contour by adding poles responsible for non-perturbative physics in accordance with a recent proposal. Then, we compute the genus zero Gopakumar-Vafa invariants for our case and evaluate the non-perturbative corrections to the partition function. We check that both calculations give the same result.

We determine the parametric conditions (pulse strength and duration) under which Schwinger's prediction for the pair-creation current - and therefore the probability - is significantly altered.

In this work, we use the master function approach to describe the CFT limit of the six-vertex model at the free fermion point. Using the ODE/IM correspondence, we obtain an explicit form of the master function. This allows us to compute the asymptotic expansion of the function $ω(λ, μ)$ describing the expectation values of the fermionic basis operators. As a result, we describe the entire Virasoro module of the corresponding CFT, including the integrals of motion as well.

The current study reports the design of advanced light-weight materials for high-energy accelerator beampipe applications. The objective is to optimize the combined requirements of high radiation length and stiffness properties of the designed materials. The present study targets conventional beampipe materials such as aluminum, titanium, and stainless steel as primary performance benchmarks. These conventional beampipes are used at synchrotron radiation sources, such as Indus-1 and Indus-2 in India, the Nuclotron-based Ion Collider Facility in Russia, and the ring synchrotron facility SIS 100/300 at the Facility for Antiproton and Ion Research in Germany. In this context, a series of ternary Mg--Al--Li alloys is systematically investigated to enhance the figure of merit. Two aluminum--rich alloys, A1 ($\mathrm{Al_{61.5}Li_{10.8}Mg_{27.7}}$) and A2 ($\mathrm{Al_{66}Li_{19.4}Mg_{14.6}}$), along with three magnesium-rich alloys, M1 ($\mathrm{Al_{23.9}Li_{29.3}Mg_{46.8}}$), M2 ($\mathrm{Al_{19}Li_{20.6}Mg_{60.4}}$), and M3 ($\mathrm{Al_{39.8}Li_{20.1}Mg_{40.1}}$) are explored. Thermodynamic stability, density, liquidus temperature, and phases are evaluated using Latin hypercube sampling within the Thermo-Calc TC-Python framework. Elastic properties are obtained from density functional theory calculations performed using the Vienna \textit{Ab Initio} Simulation Package. Our results show that, although the elastic moduli ($E$) of the investigated Mg-Al-Li alloys are comparable to those of conventional beampipe materials, their significantly higher radiation lengths ($X_0$) lead to an overall improvement in the figure of merit $X_0 E^{1/3}$.

Let $V$ be a Möbius vertex algebra and $G$ an abelian group of automorphisms of $V$. We construct $P(z)$-tensor product bifunctors for the category of $C_{n}$-cofinite grading-restricted generalized $g$-twisted $V$-modules (without $g$-actions) for $g\in G$ and the category of $C_{n}$-cofinite grading-restricted generalized $g$-twisted $V$-modules with $G$-actions for $g\in G$. In this paper, an automorphism $g$ of $V$ can be of infinite order and does not have to act semisimply on $V$, and the group $G$ can be an infinite abelian group containing nonsemisimple automorphisms of $V$.

The COSINE-100U (upgrade) will enhance the sensitivity of the COSINE-100 dark matter search by operating the detector array immersed in liquid scintillator (LS) at $-30^oC$. To validate the detector design for these conditions, we constructed a module using the COSINE-100U encapsulation and performed a dedicated long-term stability study. The module was first monitored at room temperature for ~110 days in air, followed by a one-week immersion in LAB-based LS to verify initial compatibility. Upon confirming stable optical performance, the temperature was lowered to $-33^oC$. During approximately 150 days of continuous operation at low temperature, we observed no degradation in performance. These results demonstrate the chemical and mechanical robustness of the encapsulation, confirming its suitability for the COSINE-100U physics run.

We analyze the CMS-reported t tbar excess within conventional Two-Higgs-Doublet Models of Types I, II, X, and Y, using the best-fit pseudoscalar parameters MA = 365 GeV, GammaA over MA = 2 percent, and tan beta = 1.28. Applying theoretical and experimental constraints, including stability, unitarity, perturbativity, flavor constraints, and collider bounds, we find that perturbativity limits the charged and heavy neutral Higgs masses to below about 723 GeV. Flavor constraints exclude Types II and Y, while the remaining parameter space in Types I and X is ruled out by recent t tbar Z measurements from ATLAS and CMS. We conclude that conventional Two-Higgs-Doublet Models cannot explain the observed t tbar excess, although toponium effects in the background modeling may modify this conclusion. This contribution is based on the proceedings of the 18th International Workshop on Top Quark Physics (TOP2025).

Motivated by recent BESIII measurements of the singly Cabibbo-suppressed processes $Λ^+_c \to ΛK^+ π^0$ and $Λ^+_c \to ΛK_S^0 π^+$, we investigate the process $Λ^+_c \to ΛK^0 π^+$ by taking into account the contribution from the low-lying excited baryon $Σ^*(1/2^-)$, dynamically generated via the $S$-wave pseudoscalar meson-octet baryon interaction, as well as from the intermediate resonances $K^*(892)$ and $N(1535)$. Our model successfully reproduces the BESIII $π^+K^0$ invariant mass distribution, and predicts a distinct cusp structure around 1.43~GeV in the $π^+Λ$ invariant mass distribution, which is associated with the predicted $Σ^*(1/2^-)$. Future high-precise measurements of this process at BESIII, Belle~II, and the proposed Super Tau-Charm Facility experiments will be crucial for testing the existence of $Σ^*(1/2^-)$ and advancing our understanding of the light baryon spectrum.

We perform a systematical investigation of the existence of the deuteron-like singly bottomed dibaryon resonance states with strangeness $S=-1,~-3,~-5$ in the chiral quark model. Two resonance states with strangeness $S=-1$ are obtained in the baryon-baryon scattering process. The first candidate is $ΣΣ_b$ in the $ΛΛ_b$ and $NΞ_b^*$ scattering process, with the resonance energy 6974.22 MeV - 6975.37 MeV and the decay width 14.450 MeV, respectively; the other one is $ΣΣ_b^*$ in the $NΞ_b$ and $NΞ'_b$ scattering process, with the resonance energy 6990.69 MeV - 7008.37 MeV and the decay width 43.790 MeV, respectively. The Root Mean Square (RMS) radius calculation shows that the former tends to be in a compact structure, while the latter tends to be in a molecular structure. Both of these resonance states are worthy of experimental exploration. Furthermore, it should be emphasized that the effect of channel-coupling is of great importance in exploring exotic hadron states, and investigating the scattering process may serve as an effective approach to identifying genuine resonances.

Quantum chromodynamics (QCD) with a general number of colors, $\Nc$, provides a powerful theoretical laboratory to explore the dynamics of non-Abelian gauge theories. Although $\Nc =3$ does not look a large number, the $1/\Nc$ expansion provides us with a very useful classification and book-keeping scheme for hadronic processes and sharpens conceptions otherwise obscured in real-world QCD with $\Nc = 3$. Important applications are dense QCD matter where the first principle methods for QCD are not available and many conceptual issues remain to be clarified. In this chapter we first review hadrons at large $\Nc$ from the viewpoint of quark-gluon dynamics, and then extend the discussions to hot/dense matter, focusing on confinement-deconfinement aspects. We emphasize how the large-$\Nc$ limit provides a unified organizing principle for hadronic and quark degrees of freedom in regimes where first-principle methods are limited. Two-color and isospin QCD, for which lattice simulations at finite density can be performed for a special reason, is reviewed.

We report a detailed investigation of three intermediate-to-old age open clusters, Berkeley 17, Berkeley 18, and Berkeley 39, utilizing precise astrometric and photometric data from Gaia DR3. Cluster membership was robustly determined through a probabilistic proper-motion analysis, yielding statistically significant samples of 600, 1042, and 907 stars, respectively. From the mean parallaxes of these members, we determine astrometric distances ranging from approximately 3.40 kpc for Berkeley 17 to 5.80 kpc for Berkeley 18. Isochrone fitting applied to the decontaminated color-magnitude diagrams constrains the cluster ages to 9.12 +/- 1.00 Gyr, 3.36 +/- 0.50 Gyr, and 5.10 +/- 0.50 Gyr, respectively. Interstellar reddening spans a wide range, from E(B-V) = 0.17 mag in Berkeley 39 to 0.58 mag in Berkeley 17. Structural parameters derived from King model fits to the radial density profiles, combined with mass function analyses, indicate that the clusters are dynamically relaxed systems with mass distributions broadly consistent with the canonical Salpeter slope. Our kinematic analysis reveals that Berkeley 17, Berkeley 18, and Berkeley 39 are part of the outer disk population.

The baryonic mass-size relation of galaxies links the total baryonic mass (stars plus gas) to the baryonic half-mass radius. In the first paper of this series, we showed that star-forming galaxies from the SPARC sample follow two distinct relations in the baryonic mass-size plane: one defined by high-surface-density (HSD), star-dominated, Sa-to-Sc galaxies, and one defined by low-surface-density (LSD), gas-dominated, Sd-to-dI galaxies. In this second paper, we study the structural relations between baryonic mass, half-mass radius, and mean surface density to constrain possible morphological transformations between star-forming and passive galaxies. We complemented the SPARC sample with $\sim$1200 passive galaxies that are nearly devoid of gas: ellipticals (Es), lenticulars (S0s), dwarf ellipticals (dEs) or dwarf spheroidals (dSphs), and the so-called `ultra-diffuse galaxies' (UDGs). Our results can be summarised as follows: (1) passive stellar components follow four distinct relations at high statistical significance, namely (i) ellipticals plus bulges, (ii) S0 disks, (iii) non-nucleated dwarfs (dEs, dSphs, UDGs), and (iv) nucleated dEs; (2) star-forming HSD disks (mostly Sa to Sc) overlap with S0 disks within 2$σ$ in the baryonic relations and within 1$σ$ in the stellar ones, so present-day spirals may simply evolve into lenticulars as they run out of gas; (3) star-forming LSD disks (mostly Sd to dI) are offset from non-nucleated passive dwarfs at more than 3$σ$ in the baryonic relations, but the two galaxy populations overlap within 1$σ$ in the stellar relations, suggesting that non-nucleated passive dwarfs may form from star-forming dwarfs only after gas removal; (4) UDGs extend the sequence of non-nucleated dEs/dSphs and may originate from the most diffuse star-forming LSD galaxies with no need for a substantial expansion of the stellar component.

We present the first high-resolution ($\sim$ 0.14") Atacama Large Millimeter/submillimeter Array (ALMA) Band 6 dust continuum and CO molecular line emission observations of the quadruple system HD 34700. In particular, HD 34700AaAb is a spectroscopic binary ($M_{\rm{bin}}=4\,M_\odot$) surrounded by two low-mass companions at large separations. Its circumbinary disk is highly substructured, featuring numerous spiral arms and a large cavity observed in infrared (IR) scattered light. We analyzed the CO line channel and intensity moment maps. By fitting a Keplerian model to the line channel emission, we identified the residual motions and conducted a line spectra analysis. We resolved an asymmetric continuum crescent on top of a dust ring at 0.39" (138 au), colocated with the IR ring. The CO molecule line emissions trace a smaller cavity in gas, whose edge aligns with the inner rim of the ring detected in H$α$ emission at 0.20" (65 au). The $^{12}$CO line emission and kinematics trace highly non-Keplerian motions ($\sim0.1Δ\upsilon_{\rm kep}$), and these CO spiral features align well with the spiral structures in scattered light. The $^{12}$CO line spectra analysis reveals a streamer above the southeastern disk plane, likely falling onto the disk. The $^{13}$CO and C$^{18}$O kinematics largely follow the disk's underlying Keplerian rotation, while $^{13}$CO exhibits tentative signs of anticyclonic vortex flows at the continuum crescent location. Our multimolecular line study suggests that the circumbinary disk of HD 34700A is highly perturbed in its upper layers, possibly warped and influenced by infalling material. While late-stage infall may account for the IR spirals and the formation of the vortex through Rossby wave instability, an embedded massive companion within the cavity may also contribute to these features.

Recent JWST observations indicate that massive quiescent galaxies (stellar mass $M_{*}\gtrsim 10^{10}~\mathrm{M_\odot}$) at high redshift ($z\gtrsim 6$) are more abundant than predicted by most existing galaxy formation simulations and semi-analytic models. Notably, the new COLIBRE simulations have succeeded in reconciling this tension, though the precise reason for their improved agreement with JWST data remains unclear. We demonstrate that the improved agreement is largely due to super-Eddington growth of supermassive black holes (BHs) at high redshift. We run a series of $(100~\mathrm{cMpc})^{3}$ simulations with the COLIBRE subgrid physics, varying the maximum allowed BH accretion rate in units of the Eddington rate. We show that only the fiducial COLIBRE model, which permits super-Eddington accretion, is consistent with the JWST constraints at $z \gtrsim 6$. Moreover, we find that in COLIBRE about $50$ per cent of BH mass growth at high redshift occurs in the super-Eddington regime, even though such events are extremely rare in time. Our work highlights the important role of super-Eddington accretion in simulations of galaxy formation for reproducing the observed early emergence of quenching of massive galaxies.

Aims. We investigate the radio loudness ($\mathcal{R}$) distribution in a large, homogeneous sample of radio galaxies. Methods. The sample is composed of galaxies from the ROGUE I/II catalogue belonging to the SDSS MGS and is divided into optically inactive radio galaxies (OPIRGs), optically active ones (OPARGs) and radio Seyferts. We use optical, mid-infrared and radio data to calculate the AGN bolometric luminosities, accretion rate ($λ$), black-hole mass ($M_{BH}$) and $\mathcal{R}$. Results. Contrary to some previous studies based on restricted samples, using our complete sample of objects with redshifts $z < 0.4$, we find no evidence of bimodality in $\mathcal{R}$. The highest $\mathcal{R}$ values are associated with extended radio structures. $\mathcal{R}$ is anti-correlated with $λ$, and spans about 2 dex at fixed $λ$. Radio Seyferts, OPARGs and OPIRGs form a sequence of increasing $M_{BH}$ with substantial overlap. Radio Seyferts show no $\mathcal{R}$-$M_{BH}$ correlation, whereas OPARGs and OPIRGs show a weak positive trend. From theoretical considerations, the observed two-dex spread in radio luminosity and $\mathcal{R}$ can be reproduced by a four-fold variation in the dimensionless magnetic flux $\varphi$ assuming realistic black-hole spins. Conclusions. The smooth distribution of radio loudness supports a common evolutionary path for all radio sources, with black-hole spin and magnetic field varying continuously. The radio loudness depends on black-hole mass and accretion rate, while moderate variations in $\varphi$ may account for the observed scatter in this relation.

I compare seven actively studied r-process nucleosynthesis scenarios against observed properties of r-process elements in the early Universe, and conclude that the most likely scenario to contribute to the site of elements below the third r-process peak is the magnetorotational r-process scenario, and that of the third peak is the common envelope jets supernova (CEJSN) r-process scenario. The collapsar and CEJSN r-process scenario might also contribute to the lighter r-process elements, and the binary neutron star (NS-NS) merger r-process scenario might contribute to the third r-process peak. The magnetar, the wind from the newly born NS, and the accretion-induced collapse of a white dwarf r-process scenarios fall short in explaining observations. They might exist, but cannot be major contributors to the r-process in the early Universe. To constrain r-process scenarios in the early Universe, I require that they explain the large scatter in the r-process abundances of very metal-poor stars, account for the correlation between light r-process nucleosynthesis and iron production, and the lack of correlation between the third peak r-process production and iron production, as inferred from very metal-poor stars. I discuss the diversity of the CEJSN r-process scenario and encourage extending its exploration.

The study of heavily obscured supermassive black hole (SMBH) growth in late-stage galaxy mergers is challenging: column densities $N_{\mathrm{H}}>10^{24},\mathrm{cm}^{-2}$ can block most nuclear emission, leaving significant gaps in the SMBH growth census. Millimeter-wave continuum emission offers a potential window into this obscured phase, as it can trace Active Galactic Nuclei (AGN) activity through mechanisms less affected by dust extinction. In this work, we test whether the observed correlation between millimeter ($\sim200,\mathrm{GHz}$) and hard X-ray (14 - 150,keV) luminosities can be used to plausibly identify hidden AGN in local (Ultra)Luminous Infrared Galaxies (U)LIRGs, including systems hosting confirmed dual AGN. We identify three sources -- one confirmed AGN and two strong candidates -- presenting significant evidence of AGN activity. The confirmed dual AGN lie within $\sim3σ$ of the mm--X-ray correlation, suggesting this relation can be used to identify hidden pairs. By combining the position of each source relative to this correlation with independent star formation rate constraints, we propose a method to disentangle AGN and star formation contributions for sources with measured column densities. While our analysis is based on a small, heterogeneous local sample and relies on empirical scaling relations, these results indicate that millimeter continuum emission may provide a useful complementary diagnostic for obscured SMBH growth. ALMA observations at high angular resolutions are particularly valuable for this approach, while future facilities such as the ngVLA will be essential to test its robustness in larger and more distant samples.

We present galaxy--galaxy lensing (GGL) measurements from the full six years of data from the Dark Energy Survey (DES Y6), covering $4031\,\mathrm{deg}^2$ and used in the DES Y6 $3\times2$pt cosmological analysis. We use the MagLim++ lens sample, containing $\sim 9$ million galaxies divided into six redshift bins, and the Metadetection source catalog, including $\sim 140$ million galaxies divided into four redshift bins. The mean tangential shear signal achieves a total signal-to-noise ratio (S/N) of $173$, corresponding to a $17\%$ improvement over DES Y3. After applying the scale cuts used in the cosmological analysis, with $R_{\min}=6\,\mathrm{Mpc}/h$ ($4\,\mathrm{Mpc}/h$) for the linear (nonlinear) galaxy-bias model, the S/N is reduced to $75$ (90). A comprehensive suite of validation tests demonstrates that the measurement is robust against observational and astrophysical systematics at the statistical precision required for the DES Y6 analysis. Although not used in the main cosmological analysis, we extract high--signal-to-noise geometric shear-ratio measurements from the galaxy--galaxy lensing signal on small angular scales. These measurements provide an internal consistency check on the photometric redshift distributions and shear calibration used in the $3\times2$pt analysis.

Extreme synchrotron blazars, characterized by a first peak in their broadband spectral energy distributions (SEDs) at frequencies exceeding $10^{17}$ Hz, often exhibit a second peak beyond 1~TeV. These sources serve as ideal laboratories for studying particle acceleration and radiation mechanisms in relativistic jets. In this work, we systematically analyze the $\sim$16-year Fermi-LAT observational data for 25 extreme high-synchrotron-peaked BL Lacs (EHBLs). The results indicate that the majority of these sources display stable or low flux levels in the GeV band, with only 6 sources showing significant variability at a confidence level exceeding 5$σ$. The time-averaged spectra over the 16-year period for most EHBLs are well described by a hard power-law model, with photon indices predominantly clustered between 1.7 and 1.8. Using Fermi-LAT data in conjunction with multiwavelength observations compiled from the literature, we construct broadband SEDs for these EHBLs and fit them with a one-zone synchrotron + synchrotron-self-Compton (SSC) model. We find that this simplified theoretical framework is sufficient for modeling the observed SEDs of most of these EHBLs, albeit requiring relatively higher electron energies compared to other $γ$-ray emitting HBLs, and at times under-representing the UV emission. Based on the SED fitting results, we investigate the physical properties of the emission regions in these EHBLs and compare them with those of other $γ$-ray emitting HBLs. Consistent with other GeV--TeV $γ$-ray-emitting BL Lacs, the jets in these EHBLs are marked by low radiation efficiency and low magnetization.

Bayesian model comparison implements Occam's razor through its sensitivity to the prior. However, prior-dependence makes it important to assess the influence of plausible alternative priors. Such prior sensitivity analyses for the Bayesian evidence are expensive, either requiring repeated, costly model re-fits or specialised sampling schemes. By exploiting the learned harmonic mean estimator (LHME) for evidence calculation we decouple sampling and evidence calculation, allowing resampled posterior draws to be used directly to calculate the evidence without further likelihood evaluations. This provides an alternative approach to prior sensitivity analysis for Bayesian model comparison that dramatically alleviates the computational cost and is agnostic to the method used to generate posterior samples. We validate our method on toy problems and a cosmological case study, reproducing estimates obtained by full Markov chain Monte Carlo (MCMC) sampling and nested sampling re-fits. For the cosmological example considered our approach achieves up to $6000\times$ lower computational cost.

In a recent publication [R. C. Tautz and I. Lerche, Res. Astron. Astrophys. 16, 162 (2016); arXiv:1606.08272], the authors present a derivation of the Green's function for the three-dimensional telegraph equation (also known as the heat wave equation, or relativistic heat conduction equation). We demonstrate that the closed-form expression derived in their Appendix A is incorrect. Specifically, the published solution lacks a Dirac delta term representing the ballistic wavefront and contains an algebraic error in the prefactor of the wake term. These omissions arise from the neglect of distributional derivatives when differentiating a Heaviside step function. We provide a rigorous derivation of the Green's function using the Fourier transform method, verify the correct limiting behavior as the damping vanishes, and pinpoint the exact mathematical step where the original derivation failed.

Surface chemistry on interstellar dust grains is recognized as a central component in astrochemical models, representing a plausible formation route for many of the observed complex molecular species. However, key parameters governing interstellar surface chemistry, such as diffusion energy barriers, remain poorly constrained. In particular, surface diffusion constitutes a fundamental step for the synthesis of complex organic molecules and plays a crucial role in understanding the desorption process. In this paper, the diffusion dynamics of carbon monoxide (CO) on amorphous solid water (ASW) surfaces, representative of interstellar ices, is modeled with quantum-chemical methods. Employing a representative ensemble of water clusters, each made by 22 molecules, diffusion energy barriers between the binding sites are computed using Density Functional Theory. Diffusion rate coefficients are then determined by applying the harmonic approximation of Transition State Theory. The results, in agreement with experimental studies, revealed a wide distribution of diffusion energies. This reflects the intrinsic topological heterogeneity of ASW surfaces, and highlights how surface mobility significantly influences CO's desorption dynamics and, as a consequence, surface-mediated reactivity in interstellar environments. We show that key parameters commonly employed in astrochemical models, like the ratio between binding and diffusion energy, should be carefully revised.

We investigate collapsar models with and without self-gravity under identical initial conditions to directly compare the effects of self-gravity on jet properties, such as opening angle, jet power, terminal Lorentz factor, and its variability. We compute a suite of time-dependent, three-dimensional GRMHD simulations of collapsars in evolving spacetime. We update the Kerr metric components due to the growth of the black hole mass and changes its angular momentum. The self-gravity is considered via perturbative terms. We present for the first time the process of jet formation in self-gravitating collapsars. We find that self-gravity leads to temporary jet quenching, which can explain some features in the gamma-ray burst prompt emission. We find no substantial difference in jet launching times between models with and without self-gravity. We observe that in the absence of self-gravity, the jet can extract more rotational energy from the black hole, while self-gravitating models produce narrower jet opening angles. We show that under certain conditions, self-gravity can interrupt the jet formation process, resulting in a failed burst. Our computations show that self-gravity significantly modifies the process of jet propagation, resulting in notably different jet properties. We show that the timescales, variability, and opening angle of jet depend on whether self-gravity is included or not. We argue that self-gravity can potentially explain certain prompt emission properties due to the jet quenching.

We investigate the connection between ionized gas kinematics and gas-phase metallicity gradients in 21 star-forming galaxies at $0.5 < z < 1.7$ from the MSA-3D survey, using spatially resolved JWST/NIRSpec slit-stepping observations. Galaxy kinematics are characterized by the ratio of rotational velocity to intrinsic velocity dispersion, $v/σ$, measured at $1.5\,R_e$, where $R_e$ is the effective radius. We find that dynamically hotter disks exhibit systematically flatter metallicity gradients, with a moderate anti-correlation between metallicity gradient and $v/σ$ (Pearson $r=-0.43$, $p=0.05$) and a linear fit yields a slope of $\sim 0.005$ dex per dex in $v/σ$, weaker than the dependence on stellar mass. A significantly stronger anti-correlation is observed with $R_e/σ$, interpreted as a proxy for the radial mixing timescale ($r=-0.59$, $p=0.005$), indicating that cumulative radial mixing more directly regulates chemical stratification. The metallicity gradients in our sample are uniformly shallow, indicating that efficient turbulent mixing in kinematically settled disks regulates the chemical structure of typical star-forming galaxies at $z\sim1$.

Aims. Massive stars impact their surrounding initiating star-formation along their photo-dissociation region. Once the HII region is formed it is unclear if and how the second generation of stars impacts its aspect and evolution. Methods. We performed high spectral resolution (R ~ 23400) Ha Fabry-Perot observations in five fields covering the Galactic HII region NGC 7538 and lead profiles multi-gaussian fitting to extract the parameters as peak intensity, width and velocity. We then analyse the kinematics of the ionised gas building kinematic diagrams and second order structure functions for every field. Results. The observations reveal a general blue-shifted ionised gas flow larger than 11 km s-1 in NGC 7538, consistent with previous studies. Profiles originating from features that are dark in Ha due to extinction or from outside the region show velocity dispersion larger than the one typically found for the Warm Interstellar Medium. The analysis of kinematic diagrams and second-order structure functions reveals non-thermal motions attributed to turbulence and large-scale velocity gradients. In the direction of the HII region itself the turbulence seems to be shock-dominated, with a characteristic scale length between ~ 0.72 and 1.46 pc. In this context, we propose that the kinematics of the central part of the region could be explained by the superposition of the outflow coming from IRS1 and a wind bow shock formed ahead IRS6.

The Gaia Alerts system is providing alerts on a variety of objects displaying a significant photometric change detected by the Gaia satellite from one passage to the next one over the same region of the sky. Among the over 22000 alerts published until the end of 2022, around 13 percent concern AGN or quasar candidates. We have embarked on a spectroscopic ground-based follow-up of some of those (including some selected by a different method specifically in galactic nuclei), to establish their true nature, and reveal the various phenomena leading to a change in magnitude of those AGN. The present paper deals with radio-quiet objects, while the radio-loud ones will be described in a companion paper. We confirm, on one hand, the AGN/quasar nature of 64 new candidates alerted by Gaia, and, on the other hand, obtained second-epoch spectra of over 200 known AGN also alerted for large photometric variations. The observed phenomena show a large variety: from Flares to Tidal Disruption Events (TDEs) and a large number of Changing Look Quasars (CLQs, 56 secure ones, plus 23 probable ones), not forgetting some rarer events like SNe, microlensing events or Extreme Coronal Line Emitters. This sample shows that variability is an excellent tool to detect new quasars, especially radio-quiet ones that otherwise would be undetected, and that a significant fraction of variable AGN/quasars, around 10 percent, presents the CLQ phenomenon. Some of the new CLQs are followed-up to monitor further changes and measure time scales.

We present a detailed temporal and spectral analysis of the blazar S5\,1044+71 using multi-wavelength data obtained from the \emph{Fermi}-LAT and Swift-XRT/UVOT telescopes. Applying the Bayesian block algorithm to the 3-day binned $γ$-ray lightcurve, we identify pronounced variability, including four major outbursts marked by significant flux enhancements. The highest flux recorded was $(1.1 \pm 0.2)\times 10^{-6}\,\text{ph}\,\text{cm}^{-2}\,\text{s}^{-1}$ on 57868.5 MJD. Each outburst comprises multiple components, and lightcurve profile analysis indicates predominantly symmetric temporal structures. The shortest variability timescale of 4.5 hours constrains the emission region to be located within 0.03 pc of the central engine, likely near the broad-line region (BLR). Additionally, two highest-energy photons were detected with energies of 46.4 GeV (on 57739.6 MJD) and 42.5 GeV (on 59161.9 MJD), observed outside the peak flaring activity. The fractional variability shows an overall increasing trend from UV/optical to $γ$-ray bands, with a noticeable dip in the X-ray range, consistent with the shape of the broadband spectral energy distribution (SED). The flux distributions during flares exhibit log-normal or double log-normal behavior, suggesting multiplicative variability processes and evolving emission zones. Cross-correlation analysis reveals a strong positive correlation between the $γ$-ray and X-ray bands, with X-rays lagging by 42.5 days. Broadband SED modeling across different flux states supports a one-zone leptonic scenario, with $γ$-ray emission produced via external Compton scattering of IR and BLR photons. High-flux states show harder electron spectra, elevated break energies, and reduced magnetic fields-features consistent with efficient particle acceleration and Compton dominance.

We introduce a complete model for dephasing due to line-of-sight acceleration (LOSA) in gravitational wave (GW) signals from stellar-mass binary black holes (BBHs) in three-body systems. Our prescription provides curvature- and projection-dependent phase features that are not recovered by local-expansion-based treatments. We perform parameter-space surveys and mock parameter inferences assuming the nominal sensitivity of the Einstein Telescope (ET) to identify the regime where the time-varying LOSA allows for separate constraints on the outer orbital parameters, in particular the tertiary mass and distance. We estimate that ET may detect a few to tens of such systems per year, provided that all binaries merge dynamically, and demonstrate that these constraints can be used to directly discriminate between a dynamical and AGN origin for BBHs. Finally, we reanalyse the GW190814 event and four O4a events finding no evidence for LOSA, with the previously claimed LOSA in GW190814 disappearing when a sufficiently long data segment is used.

A key question in extragalactic astronomy is how the star-formation rate density (SFRD) evolves over cosmic time. A powerful way of addressing this question is using radio-continuum observations, where the radio waves are unaffected by dust and are able to reach sufficient resolution to resolve individual galaxies. We present an investigation of the 1.4 GHz radio luminosity functions (RLFs) of star-forming galaxies (SFGs) and Active Galactic Nuclei (AGN) using deep radio continuum observations in the COSMOS and XMM-LSS fields, covering a combined area of $\sim 4\,\mathrm{deg}^2$. These data enable the most accurate measurement of the evolution in the SFRD from mid-frequency radio continuum observations. We model the total RLF as the sum of evolving SFG and AGN components, negating the need for individual source classification. We find that the SFGs have systematically higher space densities at fixed luminosity than found in previous radio studies, but consistent with more recent studies with MeerKAT. We attribute this to the excellent low-surface brightness sensitivity of MeerKAT. We then determine the evolution of the SFRD. Adopting the far-infrared - radio correlation results in a significantly higher the SFRD at $z > 1$, compared to combined UV and far-infrared measurements. However, using more recent relations for the correlation between star-formation rate and radio luminosity, based on full spectral energy distribution modelling, can resolve this apparent discrepancy. Thus radio observations provide a powerful method of determining the total SFRD, in the absence of dust-sensitive far-infrared data.

Context. The European Archive of the James Webb Space Telescope (eJWST) provides access to all data collected by the James Webb Space Telescope (JWST). JWST's capabilities span from studying early universe galaxy formation to probing exoplanet atmospheres. Specifically, for Active Galactic Nuclei (AGN), JWST offers unparalleled opportunities, enabling investigation into AGN phenomena with unprecedented detail through high-resolution imaging, spectroscopy, and photometric data. Aims. This study aims to compile and release a catalogue of all AGN observations conducted with JWST. Using eJWST, we systematically filter and organize these observations to facilitate access and retrieval of all of JWST's data products related to AGNs. Our goal is to provide the community with a valuable resource for their research. Methods. We compiled the AGN observations in eJWST using specific keywords set by the principal investigators in their proposals, manually reviewing the approved programs of JWST, as well as cross-matching all available observations with available AGN catalogues such as the Million Quasar catalogue, the SDSS MaNGA AGN catalogue, the CDFS catalogue, and others. Results. The resulting catalogue contains a total of 3,242 individual AGNs included in JWST observations. This is one of the first extensive collections of AGN observations from the JWST. It includes detailed information about the targets (name, coordinates, redshift), specifics of the JWST observations (instrument, aperture, filter, etc.), and provides links for data downloads.

Recently, Goldman (2024) obtained evidence for a large scale compressible, Burgers turbulence in the ism of a gravitationally lensed, star-forming galaxy at $z = 1.87$, with an outflowing wind. The turbulent timescale on the largest spatial scale has been found to be ~500 Myr . This together with the large spatial scale of~ 6.4 kpc suggest a large scale generating mechanism (such as tidal interaction or merger) that lasted for ~500 Myr. On the other hand, the outflowing wind is much younger and is probably the result of the intense star formation. Therefore, could it be that the star formation drives also turbulence on small scales? In the present work we utilize multi-point second order structure functions to find whether there exists also a small scale turbulence in this galaxy, and if so, try to identify its drivers. We obtained evidence for small scale turbulence whose largest spatial scale ~240 pc for the nebular gas velocity field and ~ 290 pc$ for the outflowing wind velocity field. These values suggest that stellar sub clumps or giant star clusters with an high concentration of young massive stars could be responsible for both the outflow and for the small scale turbulence.

We present a measurement of the angular Baryon Acoustic Oscillation (BAO) scale from the completed Dark Energy Survey (DES) dataset excluding the area of overlap with the Dark Energy Spectroscopic Instrument (DESI). We follow the same methodology and validation process as in the DES Y6 BAO analysis. We interpret the impact of this measurement in the context of the statistical preference for $w_0w_a$CDM over $Λ$CDM when combined with DES Y5 Type Ia supernovae (SN), Planck Cosmic Microwave Background (CMB) and DESI BAO. Based on our previous work, using the full Y6 DES BAO sample, in combination with SN, CMB and DESI DR1 BAO, added 0.3$σ$ in this preference (from 3.7$σ$ to 4.0$σ$), but this ignored possible correlations between datasets. Using our new DESI-independent DES BAO likelihood instead, we find a smaller increase in the statistical preference for $w_0w_a$CDM, from 3.7$σ$ to 3.8$σ$ when using DESI DR1 BAO, and from 4.0$σ$ to 4.1$σ$ when updating to the more recent DESI DR2 BAO. These significances reduce to 3.1$σ$ when using the new calibrated DES SN-Dovekie. Alongside this work, we publicly release BAOfit_wtheta, the BAO fitting code for the angular correlation function used in the DES Y6 BAO analysis.

Returning radiation is thought to play a key role in disk illumination of black hole X-ray binaries in the high-soft state, yet it has not been fully incorporated into XSPEC reflection models. We present a new table model for reflection spectroscopy that, for the first time, self-consistently accounts for the returning radiation. To isolate this effect, we adopt the standard disk-corona configuration but disable the corona, allowing the reflection spectrum to be produced solely by self-irradiation of the disk. Applying our model to the black hole X-ray binary 4U 1630$-$47, we report a rapidly spinning black hole ($a_* \sim 0.99$), a disk inclination of $i \sim 53^\circ$, a mass accretion rate of $\dot{M}_{\rm BH} \sim 15\% \, {\rm \dot{M}_{Edd}}$, and an electron density of $n_{\rm e} \sim 10^{21} \,\mathrm{cm^{-3}}$ to reproduce the observed reflection features. The model also yields a source distance of $D\sim 8.2 \, {\rm kpc}$, slightly below the commonly adopted value of $10 \, {\rm kpc}$ in the literature. Compared to the widely used relxillNS, our model naturally produces a harder high-energy reflection spectrum, fitting the data without invoking a Comptonized component.

We present the methodology for the weak lensing and galaxy clustering analyses of the Dark Energy Survey (DES) Year 6 data set. In this work, we design and validate the analysis pipeline for the cosmic shear, galaxy clustering plus galaxy$-$galaxy lensing ($2 \times 2$pt), and the joint analysis in the $3 \times 2$pt. Our framework accounts for key theoretical uncertainties, such as baryonic feedback and galaxy bias, incorporating both linear and non-linear models. We apply scale cuts in regimes where theoretical modeling becomes unreliable. The robustness of the pipeline is validated using mock data and simulations, confirming unbiased cosmological constraints and highlighting the importance of posterior projection effects in the validation process. As a result, we deliver robust and validated analysis pipelines for cosmic shear, $2 \times 2$pt, and $3 \times 2$pt in $Λ$CDM and $w$CDM scenarios, including a well-defined set of scales suitable for real data analysis, a robust prescription for theoretical systematics, and the theoretical covariance of the signal. This comprehensive methodology also lays the groundwork for future galaxy surveys such as the Vera C. Rubin Observatory Legacy Survey of Space and Time.

Gravitational lensing magnification alters the observed spatial distribution of galaxies and must be accounted for to prevent biases in cosmological probes of the large-scale structure. We investigate its effects on the Dark Energy Survey Year 6 galaxy clustering and galaxy-galaxy lensing analyses using the fiducial lens (position tracer) sample MagLim++. Magnification bias is parameterized by a coefficient that describes the response of the number of selected objects per unlensed area element to a change in the lensing convergence. We quantify this coefficient using the Balrog synthetic source injection catalog to account for the complexity of the selection function, and compare these results with simplified estimates. The resulting values of the magnification coefficients for each redshift bin are [3.16 $\pm$ 0.08, 2.76 $\pm$ 0.21, 4.09 $\pm$ 0.15, 4.42 $\pm$ 0.16, 4.90 $\pm$ 0.29, 4.83 $\pm$ 0.25]. Relative to Year 3, this analysis provides more precise and accurate magnification bias estimates through a larger Balrog area and reweighting to better match the data properties. The cosmological results are robust when tested against various magnification parameter prior choices and also when adding cross-clustering between lens redshift bins. Neglecting magnification, however, introduces significant systematic shifts: relative to the fiducial analysis with Gaussian priors centered on the Balrog-derived estimates, we observe shifts of 1.37$σ$ in $S_8$ and -0.84$σ$ in $Ω_m$ (with cosmic shear included: -0.61$σ$ in $S_8$ and -0.71$σ$ in $Ω_m$), in agreement with findings from simulated data, demonstrating that magnification must be modeled to avoid biases. Freeing the magnification bias in lens bin 2 leads to unphysical negative values, further justifying its exclusion from the fiducial Year 6 analysis.

We present a detailed photometric and spectroscopic analysis of the Type IIP supernova SN~2023zcu, which exploded in the galaxy NGC~2139 (redshift $z$ = 0.006). SN~2023zcu exhibits a well-sampled light curve covering the rise, plateau, and nebular phases. It has an optically thick phase of $100.6 \pm 0.6$ d with a magnitude drop of $\sim$1.7 mag in the {\em V} band during the transition between the plateau and the nebular phases. Weak emission features in the early-time spectra indicate a low-level interaction between circumstellar material (CSM) and the SN ejecta. The spectral evolution is well sampled and exhibits a prominent P-Cygni profile of H$α$, a defining characteristic of Type IIP SNe. Signatures of metal-line formation (e.g., \ion{Fe}{2}, \ion{Ca}{2} near-infrared triplet) are also evident in the spectra as the SN evolves. Spectral modeling with the radiative-transfer code \texttt{TARDIS} during the early photospheric phase (8.7--35.5 d since explosion) yields photospheric temperatures decreasing from $\sim$9,000 to $\sim$6,000 K and expansion velocities declining from $\sim$10,000 to $\sim$5,400 km s$^{-1}$. A tailored expanding photosphere method (EPM) fit based on the \texttt{TARDIS} models provides a distance estimate of $27.8 \pm 2.0$ Mpc. Nebular-phase spectra and bolometric light-curve modeling suggest a progenitor mass in the range 12--15 M$_\odot$. This thorough analysis helps to constrain progenitor properties and explosion parameters, thereby strengthening our understanding of Type IIP SNe.

G284.3$-$1.8 is a supernova remnant with a radio shell and thermal X-ray emission. Located near its center is the gamma-ray binary 1FGL J1018.6$-$5856, although the physical association between the two systems is not clear yet. Our X-ray spectroscopy with Suzaku reveals that G284.3$-$1.8 and 1FGL J1018.6$-$5856 have compatible absorption column densities of $N_\mathrm{H} = 6\textrm{--}7 \times 10^{21}~\mathrm{cm}^{-2}$, indicating that the two systems have similar distances. The actual distance is determined as $3~\mathrm{kpc}$ using $\mathrm{^{12}CO}$ ($J=1\textrm{--}0$) data obtained with NANTEN. The X-ray spectrum of G284.3$-$1.8 shows a strong K-shell emission line of Mg, confirming that the earlier claim that the SNR is one of the few Mg-rich SNRs. Comparing recent stellar models taking into account the "shell merger" processes, we find that the obtained Mg-to-Ne mass ratio of $M_\mathrm{Mg}/M_\mathrm{Ne} = 0.73^{+0.07}_{-0.03}$ and Si-to-Mg mass ratio of $M_\mathrm{Si}/M_\mathrm{Mg} = 0.44\pm0.03$ suggest a supernova explosion that would have left a neutron star. The characteristics of 1FGL J1018.6$-$5856, on the other hand, are better explained with a model in which its compact object is neutron star. The present results, therefore, would suggest a possible scenario where G284.3$-$1.8 and 1FGL J1018.6$-$5856 are both remnants of a common supernova explosion although further observational tests are necessary.

NGC 4395 is a nearby dwarf spiral galaxy hosting an active galactic nucleus (AGN) powered by an intermediate-mass black hole (IMBH, $M_{\rm BH} \sim 10^{4-5}\,M_\odot$). Recent optical continuum reverberation mapping studies have suggested potential lag variations between different epochs, offering important clues to the physical mechanisms governing variability in the vicinity of the central black hole. We present continuous intranight multi-band photometric monitoring of NGC 4395 based on three nights of observations with the Faulkes Telescope North and two nights with the Mephisto telescope. This represents the first systematic investigation of optical continuum lag stability in a robustly confirmed IMBH. By applying difference-imaging techniques to both the new observations and archival data, we significantly detect the optical inter-band lags of $\sim$5-15 minutes, which increase monotonically with wavelength. No obvious $u$-band lag excess is observed, implying a negligible fractional contribution from diffuse continuum (DC) emission to the optical continuum, in agreement with our spectral decomposition results. Remarkably, the inter-band lags remain stable over multi-year baselines. We attribute this long-term lag stability primarily to the minor DC contribution, a relatively steady disk-corona structure, and the unusually high X-ray-to-optical luminosity ratio characteristic of low-luminosity AGNs, which likely allows X-ray reprocessing to dominate over other potential variability mechanisms. Future facilities like Gemini/SCORPIO, with its simultaneous optical-to-near-infrared coverage, will be ideally suited to play an important role in advancing this field.

Changing-look active galactic nuclei (CLAGNs) represent a rare class of AGNs that undergo transitions from type 1 (characterized by the presence of broad emission lines in their spectra) to type 2 (absence of broad emission lines) or vice versa, over timescales ranging from months to years. Since normal type 1 and type 2 AGNs are known to show different polarization properties, detailed investigations of the CLAGN polarization can shed light on the underlying mechanisms responsible for the changing-look phenomenon. We present new (spectro)polarimetric observations of two changing-look AGNs located in the core of the inclined spiral galaxies NGC 1365 and NGC 2992. Both AGNs are radio emitters, thereby enabling a comparison of their polarization to the radio jet axis, which defines the accretion disk geometry. In the case of NGC 1365, the AGN shows polarization characteristics consistent with those observed in type 1 Seyferts, in particular polarization parallel to the radio jet. This intrinsic polarization is modified by the wavelength-dependent dichroic extinction that occurs in the galaxy bar and that rotates the polarization angle at the shortest wavelengths. NGC 2992, on the other hand, is so inclined that dichroic dust extinction in the disk completely dominates the polarization of the AGN, thus overwhelming any polarization due to scattering. Consequently, the polarization properties remain essentially constant between the different AGN states, and the faint broad lines observed in the polarized flux are most likely not scattered light. Differential dilution between the continuum and the narrow-line polarizations can explain the unusually high polarization measured in the emission lines.

Accretion onto black holes often leads to the launch of outflows that significantly influence their surrounding environments. The two primary forms of these outflows are X-ray disk winds-hot, ionized gases ejected from the accretion disk-and relativistic jets, which are collimated streams of particles often expelled along the rotational axis of the black hole. While previous studies have revealed a general association between spectral states and different types of outflows, the physical mechanisms governing wind and jet formation remain debated. Here, using coordinated NICER and MeerKAT observations of the recurrent black hole X-ray binary 4U 1630-472, we identify a clear anti-correlation between X-ray disk winds and jets: during three recent outbursts, only one type of outflow is detected at a time. Notably, this apparent exclusivity occurs even as the overall accretion luminosity remains within the range expected for a standard thin disk, characteristic of the canonical soft state. These results suggest a competition between outflow channels that may depend on how the accretion energy is partitioned between the disk and the corona. Our findings provide new observational constraints on jet and wind formation in X-ray binaries and offer a fresh perspective on the interplay between different modes of accretion-driven feedback.

The results of a study of the cosmic ray composition in individual events in the energy range (2-12.5) EeV using the muon correlation method is presented. The considered sample included showers with zenith angles less than 60 degrees recorded in the period 1974-2018. The existence of four separate groups of primary particles with different origins is confirmed. The obtained results have potential importance for understanding the composition of cosmic rays in the specified primary energy range.

We present a unified framework that combines early- and late-Universe observations to constrain three functional realizations of f(T,B) gravity: the linear, quadratic, and general power-law models. First, constraints on deviations from the standard weak interaction freeze-out temperature are derived using the most recent measurements of the primordial helium-4 mass fraction. Second, we perform a joint analysis incorporating five priors: Type Ia supernovae, baryon acoustic oscillations, cosmic chronometers, Big Bang Nucleosynthesis, and Cosmic Microwave Background in order to place bounds on the model parameters. The joint likelihood analysis significantly tightens the constraints compared to individual datasets. Third, we test the null, strong, and dominant energy conditions to evaluate the physical viability of the best-fit solutions across the cosmic redshift range. Our results show that all three f(T,B) models are consistent with current observations and exhibit stable behavior under the energy-condition criteria, supporting torsion-boundary modified gravity as a robust and viable alternative to General Relativity.

The formation mechanisms of intensely starbursting galaxies at high redshift remain unknown. One possible mechanism for triggering these starbursts is mergers and interactions, but detecting these at high redshift remains a challenge. Observations of high-redshift gravitationally lensed galaxies enable studies of the interstellar medium and environment of these extreme starbursts in detail. We used high angular resolution observations of dust continuum, CO(6-5), H2O(211-202), and H2O+(202-111) emission to constrain the ongoing processes in the z = 3.63 gravitationally lensed submillimeter galaxy H-ATLAS J083051.0+013224 (G09v1.97). We used PyAutoLens to create a de-magnified source plane CO(6-5) emission line cube and performed kinematic modeling using 3DBarolo. Additionally, we investigated the properties of the continuum and molecular line emission in the source plane. We find that the regions of CO(6-5) and H2O(211-202) emission match closely in the source plane but that the dust continuum emission is more compact. We find that our lens modeling results do not require more than one source, contrary to what has been found in previous studies. Instead, we find that G09v1.97 resembles a rotating disk with Vmax/sigma = 2.8 +/- 0.4 with evidence for residual emission indicative of non-circular motions such as outflows, tidal tails, or an additional background galaxy. We suggest that the origin of the non-circular motions may be associated with a bi-conical outflow, a tidal tail from an interaction, or indicate the possible presence of an additional galaxy. We calculate the dynamical mass, gas mass, star-formation rate, and depletion time for G09v1.97 and find a high star-formation rate and low gas depletion time. In combination, this suggests that G09v1.97 has recently undergone an interaction, triggering intense star formation, and is in the process of settling into a disk.

Double source plane strong lensing (DSPL) systems offer a robust, independent probe of cosmological parameters. The Chinese Space Station Telescope (CSST) is expected to discover hundreds of DSPLs, yet the survey modes and system configurations that best enable cosmological inference remain uncertain. To investigate the impact of varying signal-to-noise ratios (SNR) and Einstein radius ratios of DSPLs (denoted as $β^{-1}$ parameters) on cosmographic inference under different CSST survey modes (Wide Field (WF), Deep Field (DF), and Ultra-Deep Field (UDF)), we simulate and model mock lenses with Singular Isothermal Ellipsoid (SIE) mass profiles and Sérsic sources whose image properties are tailored to CSST specifications. Assuming a flat $w$CDM universe with fiducial values $Ω_{\rm m} = 0.30966$ and $w = -1$, and uniform priors of $Ω_{\rm m} \in [0, 1]$ and $w \in [-2, -1/3$), we find that the constraining power on cosmological parameters for a given DSPL system increases significantly with survey depth. For a representative DSPL system with two prominent arcs and a moderate $β^{-1}=1.17$, the constraints on ($w, Ω_{\rm m}$) improve from ($-1.28_{-1.00}^{+0.64}, 0.50_{-0.32}^{+0.28}$) in the WF to ($-1.59_{-0.32}^{+0.63}, 0.42_{-0.06}^{+0.15}$) in the UDF. Furthermore, we find that systems with smaller $β$ values yield tighter cosmographic constraints. We conclude that DSPL systems identified in UDF observations, particularly those with small $β$, are the most promising candidates for early-stage cosmological studies with CSST.

Determining how galaxies accumulate stellar mass is paramount to understanding the Universe. Two primary mechanisms drive this process: star-formation (SF) & mergers. Our understanding of star formation, and to some degree the processes that influence the baryon cycle (environment, gas supply, feedback, etc), are either relatively well constrained or will develop significantly over the coming decades via upcoming facilities (i.e. through their imprint on galaxy properties measured with deep multi-wavelength and spectroscopic data). However, the same can not be said for mergers. It is telling that we indirectly know hierarchical assembly through mergers is one of the most crucial processes that shape our Universe, but the robust observational measurement of mergers is almost non-existent outside of the local Universe - let alone how these mergers impact galaxy properties. This is not likely to significantly change in the coming decades as existing or approved facilities/surveys are inadequate in charactering mergers in the distant Universe. Motivated by this, we discuss an ambitious study to first explore mergers, and then the co-dependent astrophysical process that govern the accumulation of stellar mass over the last ~10billion years, and highlight the essential need for a 10m+ class multi-object spectroscopic facility.

Axion-like particles (ALPs) are hypothetical particles and compelling candidates for cold dark matter. Their existence could be probed through their conversions into photons in the presence of magnetic fields. In this work, we explore the effect of these photon-ALP conversions by searching for an attenuation in the observed gamma ray spectra of galactic sources that emit at energies of hundreds of TeV. We analyze data from the High-Altitude Water Cherenkov (HAWC) Observatory for the source 3HWC J1908+063. No evidence of photon-ALP conversions was found, and we set constraints on the ALP parameter space. Specifically, we derive exclusion limits for ALPs with masses in the range $10^{-8}~\mathrm{eV} \leq m_a \leq 10^{-6}~\mathrm{eV}$ and photon-ALP couplings in the range $10^{-12}~\mathrm{GeV}^{-1} \leq g_{aγ} \leq 10^{-10}~\mathrm{GeV}^{-1}$, based on HAWC observations.

We present a new framework for multi-line analysis that combines kernel principal component analysis (Kernel PCA), an unsupervised machine-learning method, and Kernel SHapley Additive exPlanations (Kernel SHAP), an explainable artificial intelligence (XAI) technique. To enable a comparison with PCA-based studies, which have been widely used in multi-line analyses, we apply our framework to integrated intensity maps of 13 molecular lines from Atacama Large Millimeter/submillimeter Array (ALMA) Band 3 archival data of the nearby galaxy NGC 1068. Previous PCA-based studies of NGC 1068 reported that physically meaningful structures are mainly captured up to the second component. In contrast, our framework can interpret physically meaningful features up to the fourth component. Furthermore, by comparing the results obtained from our framework with molecular column densities derived from local thermodynamical equilibrium (LTE) analysis, we suggest that the abundance of HCO+ is relatively enhanced in the molecular outflow region extending to a radius of about 400 pc from the galactic center, likely due to the effects of ultraviolet radiation and highly dense gas. These results show that our framework can provide data-driven insights into physical and chemical features that have not been clearly identified in previous studies. It also provides an efficient tool for interpreting the rapidly increasing amount of multi-line observational data.

As radio emission from normal galaxies is a dust-free tracer of star formation, tracing the star formation history of the Universe is a key goal of the SKA and ngVLA. In order to investigate how well radio luminosity traces star formation rate (SFR) in the early Universe, we have examined the radio properties of a JWST Paschen-alpha sample of galaxies at 1.0<=z<=1.8. In the GOODS-S field, we cross-matched a sample of 506 FRESCO Paschen-alpha emitters with the 1.23 GHz radio continuum data from the MeerKAT MIGHTEE survey finding 47 detections. After filtering for AGN (via X-ray detections, hot mid-infrared dust and extended radio emission), as well as blended sources, we obtained a sample of SFGs comprising: 11 cataloged radio detections, 18 non-cataloged detections (at ~3-5sigma) and 298 undetected sources. Stacking the 298 undetected sources we obtain a 3.3sigma detection in the radio. This sample, along with a local sample of Paschen-alpha emitters, lies along previous radio luminosity/SFR relations from local (z<0.2) to high redshift (z~1). Fitting the FRESCO data at 1.0<=z<=1.8 we find log(L_1.4GHz) = (1.31+/-0.17) x log(SFR_Pa-alpha) + (21.36+/-0.17) which is consistent with other literature relations. We can explain some of the observed scatter in the L_1.4GHz/SFR_Pa-alpha correlation by a toy model in which the synchrotron emission is a delayed/averaged tracer of the instantaneous Paschen-alpha SFR by ~10/75 Myr.

Using the Dark Energy Spectroscopic Instrument Milky Way Survey (DESI MWS), we examine the 6D space of the anticenter region of the stellar disk (150$^\circ$ $<$ Galactic longitude $<$ 220$^\circ$) using 61,883 main sequence turn-off stars. We focus on two well-known stellar overdensities in the anticenter, the Monoceros Ring (MRi) and Anticenter Stream (ACS). We find that the MRi overdensity has kinematics consistent with a tidally induced spiral arm, a type of dynamic spiral arm created by an interaction with a satellite galaxy, most likely the Sagittarius Dwarf Spheroidal galaxy (Sgr). We use the kinematics of the MRi to calculate the two most recent passage times of Sgr are 0.25 $\pm$ 0.09 Gyrs and 1.10 $\pm$ 0.23 Gyrs from the present day. We validate that the ACS is kinematically decoupled from the MRi because they are moving in opposite radial and vertical directions. We find that the kinematics associated with the ACS are not confined to our defined overdensity. The features we see in the ACS region are likely part of a broader distribution of stars with the same kinematic signature as detected in other places, like the vertical wave in the outer disk and phase spiral.

We present cosmology results combining galaxy clustering and weak gravitational lensing measured in the full six years (Y6) of observations by the Dark Energy Survey (DES) covering $\sim$5000 deg$^2$. We perform a large-scale structure analysis using three two-point correlation functions (3$\times$2pt): (i) cosmic shear from 140 million source galaxy shapes, (ii) galaxy clustering of 9 million lens galaxy positions, and (iii) galaxy-galaxy lensing from their cross-correlation. We model the data in flat $Λ$CDM and $w$CDM cosmologies. The combined analysis yields $S_8\equiv σ_8 (Ω_{\rm m}/0.3)^{0.5} = 0.789^{+0.012}_{-0.012}$ and matter density $Ω_{\rm m} = 0.333^{+0.023}_{-0.028}$ in $Λ$CDM (68\% CL), where $σ_8$ is the clustering amplitude. These constraints show a (full-space) parameter difference of 1.8$σ$ from a combination of cosmic microwave background (CMB) primary anisotropy datasets from Planck 2018, ACT-DR6, and SPT-3G DR1. Projected only into $S_8$ the difference is $2.6σ$. In $w$CDM the Y6 3$\times$2pt results yield $S_8 = 0.782^{+0.021}_{-0.020}$, $Ω_{\rm m} = 0.325^{+0.032}_{-0.035}$, and dark energy equation-of-state parameter $w = -1.12^{+0.26}_{-0.20}$. For the first time, we combine all DES dark-energy probes: 3$\times$2pt, SNe Ia, BAO and Clusters. In $Λ$CDM this combination yields a $2.8σ$ parameter difference from the CMB. When combining DES 3$\times$2pt with other low-redshift datasets (DESI DR2 BAO, DES SNe Ia, SPT clusters), we find a 2.3$σ$ parameter difference with CMB. A joint fit of Y6 3$\times$2pt, CMB, and those low-redshift datasets produces the tightest $Λ$CDM constraints to date: $S_8 = 0.806^{+0.006}_{-0.007}$, $Ω_{\rm m} = 0.302^{+0.003}_{-0.003}$, $h = 0.683^{+0.003}_{-0.002}$, and $\sum m_ν< 0.14$ eV (95\% CL). In $w$CDM, this combination yields $w = -0.981^{+0.021}_{-0.022}$.

We introduce IRMaGiC, an algorithm built based on RedMaGiC desgined to enhance the selection of Luminous Red Galaxies (LRGs) across the redshift range $1 \leq z \leq 2$. We show that this method extends the capabilities of the redMaGiC algorithm by applying it to simulated photometric data from the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) and the Nancy Grace Roman Space Telescope's High Latitude Wide Area Survey (HLWAS). By integrating infrared band coverage from Roman HLWAS with LSST's optical bands, IRMaGiC enables red-sequence calibration at higher redshifts. We demonstrate that IRMaGiC reduces scatter and bias in photometric redshift estimates for LRGs at higher redshift, providing more accurate redshift assessments compared to existing methods. Our findings suggest that incorporating infrared data can considerably improve the selection and redshift estimation of LRGs at higher redshift, offering substantial benefits for future cosmological surveys.

IGR J17091-3624 is a distinctive black hole X-ray binary exhibiting exotic variability, including complex "heartbeat" oscillations in its X-ray light curves, similar to those observed in GRS 1915+105, a system renowned for its structured, rapid X-ray variability but heavily obscured at optical wavelengths. In contrast, IGR J17091-3624 is less obscured, making it a more accessible target for optical investigations. Due to its weak radio emission, optical and infrared data are essential to probe the jet and outer disc behavior of IGR J17091-3624. This study presents the first long-term optical monitoring of IGR J17091-3624, using data from the Las Cumbres Observatory (LCO) over its 2011, 2016, and 2022 outbursts. We combine these observations with quasi-simultaneous X-ray data from Swift/XRT, RXTE, and NICER, employing light curve and variability analysis, spectral energy distributions, color-magnitude diagrams, and optical/X-ray correlations to investigate optical emission mechanisms. We find that the optical and X-ray fluxes are significantly correlated, following a power-law relation with the index 0.40\pm0.04, suggesting that the optical emission in IGR J17091-3624 is dominated by an X-ray-irradiated accretion disk. Based on optical spectral slope constraints, we estimate the extinction toward IGR J17091-3624 A_V = 4.3 to 6.6 mag. The global optical/X-ray correlation suggests a distance estimate of 8-17 kpc, in line with previous findings. High-cadence optical observations show tentative evidence of optical oscillations that may arise from reprocessed X-ray modulations, although confirming this will require higher time-resolution optical data.

I examine a recently proposed failed-supernova scenario for the fading of the yellow supergiant event M31-2014-DS1, and find that it requires unlikely fine-tuned parameters to work, if at all. In the failed-supernova scenario, most of the yellow supergiant collapsed to form a black hole. Due to the energy carried by neutrinos from the cooling, collapsing core, gravity decreases, leading to the ejection of a small fraction of the outer envelope, some of which remains bound. The fallback accreted gas possesses large angular-momentum fluctuations due to the pre-collapse envelope convection. The fallback material forms intermittent accretion disks around the black hole that launch jets (or disk wind), which unbind most of the bound material. The failed-supernova scenario for M31-2014-DS1 requires that only <1% of the bound material be accreted by the black hole, but the jets do not shut down the backflow for over 10 years. I find this fine-tuned requirement unlikely. I also find that, due to the rapid radiative cooling of the outflow interaction zone with the outer gas, the expected radiation is about an order of magnitude or more above the observed value. These, as well as earlier challenges raised against the failed-supernova scenario, make the alternative type II intermediate-luminosity optical transient scenario, in which fading is due to dust ejection in a violent binary interaction, more likely. The fading event M31-2014-DS1 does not support the failed-supernova scenario predicted by the neutrino-driven explosion mechanism of core-collapse supernovae.

Galaxy clustering is a sensitive probe of the expansion history and growth of structure of the universe, and key degeneracies can be broken by combining these data with measurements of cosmic shear and galaxy-galaxy lensing (a so-called 3$\times$2pt analysis). The largest and least biased statistical samples of galaxies for use in clustering analyses can be collected photometrically through large imaging surveys. However, selecting clean photometric subsamples for cosmology are crucial for avoiding contamination that can bias cosmological constraints. Here we present the MagLim++ galaxy sample, selected to optimize for cosmological constraining power and incorporating an array of novel quality cuts to identify and remove residual contamination. This sample comes from the full six years of observations from the Dark Energy Survey. We present measurements of the two-point angular clustering ($w(θ)$) of 9,186,205 galaxies distributed over 4031 sq. degrees and in six tomographic redshift bins centered at $\bar{z}\approx$ [0.31, 0.44, 0.62, 0.78, 0.90, 1.01]. These measurements are used as part of the 3$\times$2pt and other DES Y6 legacy cosmological analyses in companion works. We describe the battery of null tests and mitigation schemes implemented to address observational, astrophysical, and methodological systematics in the analysis. The resulting $w(θ)$ measurements have a S/N = 149 (90.2 for linear scales only), which we use to place galaxy-clustering-only constraints on the matter density of the Universe, $Ω_m=0.311^{+0.023}_{-0.035}$, and amplitude of galaxy clustering in each redshift bin, $b_iσ_8=[1.16^{+0.04}_{-0.06},\ 1.40^{+0.04}_{-0.06},\ 1.57^{+0.04}_{-0.06},\ 1.59^{+0.04}_{-0.05},\ 1.50^{+0.04}_{-0.05},\ 1.74^{+0.06}_{-0.08}]$.

Tidal disruption events (TDEs) represent a truly unique, and potentially very powerful, probe of the quiescent supermassive black hole (SMBH) population. Given current observational survey capabilities the vast majority of the TDEs discovered in the next decade will be observed only across optical-UV wavelengths. A set of questions of broad scientific interest relating to SMBH demographics and SMBH-galaxy correlations could in principal be answered by using TDE emission as an efficient means to constrain SMBH masses. In this paper we argue for using well-understood elements of TDE emission (the thermal X-ray continuum and late-time UV plateau) to derive empirical relationships between the more poorly understood early optical/UV flare and the black hole mass, before using these empirical relationships to measure TDE black hole masses simply and rapidly. We provide a publicly available code TDEFLARE which does this, showing (i) it produces results consistent with disk codes containing far more physics, (ii) it reproduces galactic scaling relationships at high ($>5σ$) significance, (iii) it produces reliable mass estimates for both partial and full disruptions, and (iv) it does not require late time data to derive mass constraints. We provide 89 TDE black hole mass constraints, derive the intrinsic black hole mass function implied by the current TDE population, and discuss the Malmquist-Hills bias, an important confounding factor in TDE science.

The dominant radiation mechanism that powers the prompt $γ$-ray emission in gamma-ray bursts (GRBs) remains poorly understood. High quality, time- and energy-resolved linear polarization measurements of prompt $γ$-ray photons can distinguish between synchrotron and inverse-Compton processes and provide crucial constraints on the outflow properties. This will be achieved by POLAR-2 that is proposed as a dedicated GRB polarimeter and successor to POLAR. The High-energy Polarimetry Detector (HPD) is one of the three instruments of POLAR-2 that features significantly improved sensitivity in the $(40-1000)$\,keV energy range and a detection area four times larger than that of POLAR. Here we demonstrate the capabilities of the HPD to constrain key physical model parameters by creating and fitting to synthetic sources using a time-resolved spectro-polarimetric theoretical model of prompt GRB emission. The time-resolved spectral and polarization fits are performed using a novel technique featuring maximum likelihood over an unbinned (in time and energy) list of detected events. The constrained model parameters directly relate to the underlying source physics that would reveal an accelerating, coasting or decelerating emission region. For a pulse fluence of $\mathcal{F}=10^{-5}\mathcal{F}_{-5}\,{\rm erg\,cm^{-2}}$ we can constrain the time-integrated polarization degree to an absolute accuracy ($1\,σ$) of about $2.5\mathcal{F}_{-5}^{\,-1/2}$ per cent, as long as source photons dominate over the background. In bright GRBs, such unprecedented accuracy at these energies will allow to distinguish between different models for the prompt GRB emission mechanism and constrain the magnetic field geometry, jet angular structure and outflow composition.

Astrometry from the Gaia satellite and from the long-term combination of Hipparcos and Gaia are now sensitive to sky-plane accelerations as low as $\approx$1 m/s/yr. This paper quantifies and explores an important caveat: apparent nonlinear motion due to a star's nonzero radial velocity can be indistinguishable from real astrometric acceleration. This nonlinear motion is parallel to the proper motion, so it can be both quantified and avoided by projecting apparent astrometric accelerations into components parallel and perpendicular to the proper motion. We illustrate this distinction for a sample of very nearby, fast-moving stars from the Hipparcos-Gaia Catalog of Accelerations (HGCA). We then generalize the effect of stellar radial velocity and projections of the astrometric acceleration to binary stars in which we observe the acceleration of both components. Finally, we demonstrate that the proper motion differences in the HGCA are statistically well-behaved even for the nearest and fastest-moving stars, at least in the component perpendicular to the proper motion. This distinction -- between astrometric acceleration parallel and perpendicular to proper motion -- could have important consequences for future missions reaching extreme astrometric sensitivities around nearby stars.

Galaxy clusters serve as a unique and valuable laboratory for probing cosmological models and understanding astrophysics at the high-mass limit of structure formation. Clusters that are dynamically relaxed are especially useful targets of study because of their morphological and dynamical simplicity. However, at redshifts z > 1, very few such clusters have been identified. We present results from new Chandra observations of the cluster SPT-CL J2215-3537 (hereafter SPT J2215), at z = 1.16, the second most distant relaxed, cool-core cluster identified to date. We place constraints on the cluster's total mass profile and investigate its thermodynamic profiles, scaling relations (gas mass, average temperature, and X-ray luminosity), and metal enrichment, resolving the cool core and providing essential context for the massive starburst seen in its central galaxy. We contextualize the thermodynamic and cosmological properties of the cluster within a sample of well-studied, lower-redshift relaxed systems. In this way, SPT J2215 serves as a powerful high-redshift benchmark for understanding the formation of cool cores and the evolution of massive clusters of galaxies.

We present a spectroscopic survey of field galaxies and lensed sources in the vicinity of the strong lensing galaxy cluster known as the Carousel lens at z=0.49. Using both Gemini/GMOS slitmask spectra and deep VLT/MUSE observations, we bring the total number of lensed sources up to 13, including three which were not previously known from imaging observations but are apparent in the MUSE data as emission-line sources. Of these sources, 10 have confident redshifts, and an additional 2 have tentative redshifts from likely Ly$α$ emission (including seven new redshifts determined here adding to those presented previously in Sheu et al. (2024)). The lensed sources span a redshift range from z=0.96 to 4.09 with most of them showing 3-5 images, including four sources displaying central or radial images. In total, we identify 43 images of these 13 sources. This lens system is remarkably symmetric and well-modeled by a simpler lens model than typical cluster lenses, and the large number of sources and their large range of redshifts make this cluster ideal for constraining cosmological parameters such as $w$ and $Ω_m$ as well as the cluster density profile. Additionally, we present a catalog of 57 unlensed field galaxies with confident redshifts, of which 49 are associated with the cluster. We measure a cluster velocity dispersion of about 1100 km s$^{-1}$ from which we estimate a halo mass $M_{200c} \approx 1.2 \times 10^{15} M_\odot$.

Studies of galactic bars have primarily focused on stellar bars, since they can be directly observed through ultraviolet to infrared wavebands. Cosmological as well as idealised simulations reveal that the dark matter (DM) haloes interact with baryonic matter, primarily the stellar bars, dynamically by means of the exchange of angular momentum. In these simulations, the spherical DM halo dynamically responds to interaction with the stellar bar by reshaping its orbital structure in the proximity of the stellar bar, forming a bar-like configuration, called as the Dark Matter (DM) bar. Using N-body simulations of Milky Way analogs we discuss the role of inner halo angular momentum, measured as halo spin parameter λ of the dark matter halo, on formation and evolutionary characteristics of the DM bars. Our systematic study involves haloes with initial spin configurations ranging from λ = 0 to 0.1. The result conveys that DM bar formation and its characteristics are extensively dependent on the initial spin parameter λ of the DM halo. We demonstrate that the strength of the dark matter bar gradually increases with an increase in halo spin in long-term evolution, with a significant impact of stellar bar buckling on dark matter bar strength. The evolutionary characteristics of the DM bar are strongly influenced by the initial spin of the host halo.

The stellar halo of the Milky Way contains the remnants of past accretion events, which could be detectable as substructures in the classical integrals of motion space, such as energy and angular momentum (E-Lz). However, our galaxy also contains a non-axisymmetric stellar bar, which traps stars in resonant orbits, leading to substructures in phase-space. Using a high-resolution magneto-hydrodynamic cosmological zoom-in simulation of a Milky Way analogue, we explore the connection between the bar and the accreted stellar halo. We find that the bar induces prominent substructures, or "ridges", in E-Lz, caused by the resonances. The most pronounced of these is caused by the corotation and the retrograde 1:1 resonances, with weaker ridges visible due to the prograde 1:1 and outer Lindblad resonance. The ridges are present across much of the stellar halo, with variations in radius due to the morphology of different orbital families. We explore the scattering of orbits at the resonances, finding that stars trapped at the 1:1 retrograde resonance become more circularised and have more negative angular momentum. Additionally, stars can move between the corotation and retrograde 1:1 families, thus alternating between prograde and retrograde motion. Due to these scatterings and the pre-existing metallicity gradients in the accreted population, the bar-induced substructures have distinct metallicities compared to stars in the surrounding phase-space. Our results suggest the need for caution when searching the Milky Way stellar halo for accreted substructures in both integral of motions and chemical spaces, since these can be induced by internal perturbations.

We investigate erythrohenosis -- the collision and merger of two red giants -- establishing an end-to-end model for this fundamental evolutionary channel in dense stellar environments. Combining three-dimensional SPH simulations of a binary with analytical modeling, we characterize the event from initial encounter to terminal explosion. We demonstrate that grazing encounters induce tidal capture and rapid orbital decay, accompanied by large-amplitude, nonlinear stellar oscillations. The subsequent inspiral spins up the common envelope into a stable, non-spherical equilibrium, powering a luminous precursor with quasi-periodic bursts. The terminal explosion, modeled with angular momentum conservation, produces an intrinsically flattened remnant that preserves a geometric memory, or morphomnesia, of its binary origin. The associated gravitational wave signal features a rapid, drag-dominated frequency evolution, identifiable by a unique time-varying apparent chirp mass. These results define a distinctive multi-stage observational fingerprint -- linking transient optical precursors, asymmetric nebulae, and anomalous gravitational wave chirps -- to guide identification in current and future multi-messenger surveys.

The Galactic Be star binary MWC 656 was long considered the only known Be star + black hole (BH) system, making it a critical benchmark for models of massive binary evolution and for the expected X-ray emission of Be+BH binaries. However, recent dynamical measurements cast doubt on the presence of a BH companion. We present new multi-epoch ultraviolet spectroscopy from the Hubble Space Telescope (HST), combined with high-resolution optical spectra, to reassess the nature of the companion. The far-ultraviolet spectra reveal high-ionisation features, including prominent N V and He II lines, which are absent in the spectra of normal Be stars and are indicative of a hot, luminous companion. Spectral modelling shows that these features cannot originate from the Be star or from an accretion disc around a compact object. Instead, we find that the data are best explained by a hot ($T_\mathrm{eff} \approx 85$ kK), compact, hydrogen-deficient star with strong wind signatures, consistent with an intermediate-mass stripped star. Our revised orbital solution and composite spectroscopic modelling yield a companion mass of $M_2 = 1.54^{+0.57}_{-0.46}\,\mathrm{M}_\odot$, definitively ruling out a BH and disfavouring a white dwarf. MWC 656 thus joins the growing class of Be + stripped star binaries. The system's unusual properties - including a high companion temperature and wind strength - extend the known parameter space of such binaries. The continued absence of confirmed OBe+BH binaries in the Galaxy highlights a growing tension with population synthesis models.

Observations from the James Webb Space Telescope (JWST) have uncovered supermassive black holes (SMBHs) with masses exceeding $10^6 \mathrm{M}_{\odot}$ at redshifts $z > 8$, posing significant challenges to existing models of early black hole formation and growth. Here we show, in a fully cosmological setting, that light seed black holes (LSBHs), remnants of Population III stars, can grow rapidly to $\sim10^4 \mathrm{M}_{\odot}$ in the early Universe. This growth is enabled by our novel black hole seeding prescription and the unprecedented resolution of our zoom-in cosmological simulations, which resolve the dense environments necessary for efficient accretion. Our results provide robust evidence that LSBHs can attain the masses required to serve as the dominant progenitors of the SMBH population observed at later cosmic epochs. These findings have far-reaching implications for the interpretation of JWST observations and future gravitational wave detections with LISA.

A small fraction of X-ray photons from gamma-ray bursts (GRBs), after escaping the relativistic jet, are scattered by electrons in the circumburst medium. Subsequent photon-photon absorption between the incoming MeV $γ$-rays and the back-scattered X-rays generate electron-positron pairs, enriching the surrounding medium with leptons. We investigate how these back-scattered photons modify the prompt GRB spectrum through $γ-γ$ absorption. In a dense and pair-loaded wind environment, the emerging spectra exhibit a broad attenuation structure, whose morphology is sensitive to the low-energy spectral index $α$. In particular, spectra with $α> -1$ develop a pronounced, saddle-shaped absorption between 1 and 100 MeV (rest frame). Such external MeV absorption could account for the spectral curvature seen in some bright GRBs, and may point to enhanced mass loss from their progenitor stars, consistent with early observations of core-collapse supernovae.

We present a detailed analysis of sloshing cold fronts in a sample of three nearby galaxy clusters (Abell 496, Abell 2029, and Abell 1644) observed with the Chandra X-ray Observatory. Cold fronts manifest as sharp edges in the X-ray surface brightness of the intracluster medium (ICM) in galaxy clusters. In the residual X-ray surface brightness maps, where the global ICM distribution has been subtracted, cold fronts generated by gas sloshing are observed at the boundaries of the spiral excesses. We perform a systematic and comprehensive study of the surface brightness edges along the spiral excesses. We find the deficit of the thermal pressure radially inward of the brightness edges, in contrast to stripping cold fronts that typically exhibit higher thermal pressure in brightness edges. Assuming that the sharp edges in the X-ray surface brightness distributions are sustained entirely by the gas bulk motions, we estimate the velocity gradients across the edges that are required to compensate for the deficit of the thermal pressure. We do not find statistically significant velocity gradients along the azimuthal direction. Our results suggest that alternative mechanisms such as magnetic fields and viscosity are necessary to maintain the sharpness of sloshing cold fronts.

Extended warm and cold gas nebulae, with complex morphologies and kinematics, have been observed in the centres of cool-core galaxy clusters. Their origin within the hot intracluster medium (ICM) is still puzzling, and among many mechanisms, positive feedback from the central active galactic nucleus (AGN) has been proposed. In this work, we performed a suite of very high-resolution hydrodynamic simulations of a Perseus-like cool-core galaxy cluster subject to self-regulated AGN jet feedback, which leads to realistic ICM properties. By explicitly following warm ionized, neutral, and molecular gas phases, we studied the complex interplay between AGN activity and the multi-phase ICM. While AGN feedback globally heats the ICM, we find that during the individual AGN jet bursts, hot material is also injected laterally to the jet axis, within the turbulent mixing layer. This material, as it expands, compresses the surrounding hot ICM, reducing the local cooling time, and leads to the formation of cold clumps on a characteristic timescale of $\sim 30$ Myr. By employing tracers, we explicitly track cooling within the affected regions, finding that very hot gas identified in high-compression, low-vorticity zones condenses in situ to form cold clumps. A statistical analysis reveals that the condensation of cold gas is highly promoted once the local turbulent Mach number, $σ_{hot}/c_{s,hot}$, in the hot gas component ($T \geq 10^7$ K) takes values around ~0.3. The presented process is a further important step in understanding the physical mechanisms that lead to the formation of cold gas in the cluster core. Our measured values of the characteristic turbulent Mach number, together with detailed multi-phase gas kinematics predictions, provide important theoretical tools to interpret future X-ray spectroscopy and deep radio data, ultimately to constrain the origin of cool-core cluster nebulae.

Strongly lensed star clusters have recently been detected up to redshift $z\approx 10$ in galaxy cluster fields using the James Webb Space Telescope (JWST). When pairs of mirror images of such star clusters appear across the lensing critical curve, it is usually assumed that both images will display identical spectral energy distributions (SEDs). However, this assumption may be invalidated in the presence of gravitational microlensing from stars or other compact objects in the lens, since microlensing will affect the SED contribution from bright stars within the star cluster independently in the two mirror images. Here, we explore under what circumstances mismatched mirror-image SEDs are likely to be observable, and argue that SED differences detectable in JWST observations of lensing-cluster fields will be limited to star clusters of mass $< 10^5\ M_\odot$ and ages $\lesssim 5$ Myr. The probability of severely mismatched mirror-image SEDs increases if the stellar initial mass function is very top-heavy and extends to stellar masses $\gg 100\ M_\odot$, as has been suggested to be the case for Population III stars. The prevalence of lensed star clusters with highly discrepant mirror-image SEDs could therefore serve as a probe of very massive stars and extreme stellar populations in the early Universe.

Conduction as a mechanism for explaining the disrupted cooling-flow in galaxy clusters has been mostly discounted, as the process is inefficient at transporting heat all the way from the cluster into the core. However, thermal conduction can be strongly enhanced when materials of significantly different temperature are brought into proximity, and thus into close thermal contact. Jets of active galactic nuclei may act as heat pumps by bringing low-entropy gas from the cluster core into thermal contact with the hot outer atmosphere of the cluster, significantly increasing the feedback efficiency of active galactic nuclei. We test this hypothesis by running a suite of 3D magnetohydrodynamic simulations of active galactic nuclei jets in a Perseus-like cluster, including anisotropic conduction. We find that the heat pump efficiency $η$ can reach up to 50\% of the maximum possible efficiency $η_{\rm max}$ if conduction operates near the Spitzer-Braginskii limit, while $η\approx f_{\rm sp}η_{\rm max}$ if conduction along the field lines is substantially suppressed below the Spitzer-Braginskii value by a factor $f_{\rm sp}$ by kinetic effects, as recently suggested. We further find that jet-induced thermal conduction is self-limiting: Magnetic draping during the uplift results in a magnetic field orientation close to perpendicular to the induced temperature gradients, significantly reducing conduction along the ideal conductive pathways. Thus, for conservative assumptions about thermal conduction suppression by $f_{\rm sp} \lesssim 0.1$, the heat pump effect leads to only marginal heat transfer and, correspondingly, to immaterial changes in the overall thermal evolution of cool core clusters beyond the isolated effects of conduction and jet-induced heating alone.

We demonstrate the capabilities of probabilistic diffusion models to reduce dramatically the computational cost of expensive hydrodynamical simulations to study the relationship between observable baryonic cosmological probes and dark matter at field level and well into the non-linear regime. We introduce a novel technique, Cosmo-FOLD (Cosmological Fields via Overlap Latent Diffusion) to rapidly generate accurate and arbitrarily large cosmological and astrophysical 3-dimensional fields, conditioned on a given input field. We are able to generate TNG300-2 dark matter density and gas temperature fields from a model trained only on ~1% of the volume (a process we refer to as `upscaling'), reproducing both large scale coherent dark matter filaments and power spectra to within 10% for wavenumbers k <= 5 h Mpc^-1. These results are obtained within a small fraction of the original simulation cost and produced on a single GPU. Beyond one and two points statistics, the bispectrum is also faithfully reproduced through the inclusion of positional encodings. Finally, we demonstrate Cosmo-FOLD's generalisation capabilities by upscaling a CAMELS volume of 25 (Mpc h^-1)^3 to a full TNG300-2 volume of 205 (Mpc h^-1)^3$ with no fine-tuning. Cosmo-FOLD opens the door to full field-level simulation-based inference on cosmological scale.

Fast X-ray transients (FXTs) are flashes of X-rays that last for a few hundreds of seconds to a few hours. An enigmatic population of these transients that did not have a clear origin has been known for several decades, mostly found serendipitously in soft X-ray imaging observations. Recent progress in this field by Einstein Probe has found that many FXTs are associated with gamma-ray bursts and the collapse of massive stars. Motivated by this, we searched the NuSTAR archive in the harder 3--79 keV band for $\sim1000$ s duration transients. From 204 Ms of exposure we present five candidate FXTs, four of which are spectrally hard, with power-law indices $-3<Γ<0$, standing them apart from FXTs discovered in the soft band. Three have potential associations with galaxies at $z=0.1-2$, implying 3--79 keV luminosities of $10^{43}$ to $10^{48}$ erg s$^{-1}$ and volumetric event rates of 125--2900 Gpc$^{-3}$ yr$^{-1}$. The properties of these NuSTAR FXTs most resemble low-luminosity gamma-ray bursts, and would be much more common than their higher-luminosity counterparts in this redshift range.

We present a reanalysis of the water maser system in NGC 4258 to reassess its geometric distance, commonly reported as approximately 7.6 Mpc with percent-level accuracy, a key anchor in extragalactic distance ladder calibrations and recent determinations of the Hubble constant. We introduce a method that relies exclusively on tracking individual maser components, rather than assuming a single averaged trajectory as in previous works, thereby avoiding arbitrary data averaging that can bias interpretations of the disk's geometry and dynamics. This approach requires spatially resolved individual maser components; consequently, the majority of observational epochs were excluded, as they lack sufficient spatial resolution to localize the maser position. We track individual maser components across multiple epochs and introduce an efficient marginalization method over nuisance parameters (angular radius and azimuth of each maser), reducing the number of free parameters from hundreds to 14. Our analysis reveals that the current observational cadence is insufficient to reliably track the individual masers, which our method relies on, given their intrinsic variability. Across a range of maser selections and model configurations, inferred distances span approximately between 6.7 to 8.1 Mpc, demonstrating significant sensitivity to how our method selects individual masers. Even visually and statistically robust fits can differ by several standard deviations, reflecting ambiguity in component identification across sparsely sampled epochs. We evaluate the impact of observational cadence on tracking fidelity and distance precision, and show that high-cadence monitoring is needed for our method to track individual masers and produce a robust anchor for cosmology.

The rapid emergence of supermassive black holes (SMBHs) in the early Universe poses a challenge to current models of black hole growth. One promising formation pathway is the direct collapse black hole (DCBH) scenario, in which gas in pristine, low-metallicity halos forms supermassive (or quasi-) stars leading to massive black holes seeds under specific environmental conditions. In this work, we investigate the potential host environments of DCBHs by coupling a semi-analytic model tracing BH formation and galaxy co-evolution with high-resolution N-body dark matter merger trees. This allows us to trace the population of DCBHs formed during the hierarchical assembly of a $\sim 10^{12} ~\rm M_\odot$ dark matter halo hosting a bright $10^9 ~\rm M_\odot$ quasar at redshift $z \approx 7$. We find that, when accounting for local fluctuations in the UV radiation field within this early cosmic structure, massive BH seeds can form via direct collapse as early as $z \approx 22$. Even under more stringent conditions for heavy seed formation, tens of DCBHs are predicted to emerge within the simulated overdensity down to $z \sim 14$, at which point metal enrichment of the intergalactic medium inhibits further episodes of direct collapse. A significant fraction of the massive black hole population formed at $z > 14$ is expected to survive in satellite galaxies that do not merge with the central halo down to $z \approx 7$. We show that the existence of such a population of ungrown heavy BH seeds can be probed through deep JWST observations targeting regions surrounding bright high-redshift quasars, and we discuss tailored observational strategies to detect and identify these elusive systems.

The discovery by JWST of a substantial population of compact "Little Red Dots" (LRDs) presents a major puzzle: their observed spectra defy standard astrophysical interpretations. Here, we show that LRD spectra are naturally reproduced by emission from an accreting Direct Collapse Black Hole (DCBH). Using radiation-hydrodynamic simulations, we follow the growth of the DCBH seed via a dense, compressionally heated, collisionally ionized accretion flow. The model self-consistently reproduces the screen responsible for the observed Balmer absorption, while allowing UV/optical emission to partially escape, along with reprocessed infrared radiation. Crucially, this structure is not a blackbody and requires no stellar contribution: the UV continuum originates entirely from reprocessed DCBH radiation, attenuated only by a small amount of dust with an extinction curve consistent with high-redshift galaxies. This single framework simultaneously explains the key observational puzzles of LRDs: (a) weak X-ray emission, (b) metal and high-ionization lines alongside absent star-formation features, (c) overmassive black holes, (d) compact morphology, (e) abundance and redshift evolution -- linking them directly to pristine atomic-cooling halos, (f) long-lived ($>100$ Myr), slowly variable phases driven by radiation pressure. Our findings indicate that JWST is witnessing the widespread formation of heavy black hole seeds in the early Universe.

eROSITA substantiated earlier indications that Loop-I is the northern part of an extended bipolar Galactic-bubble structure, but the southern bubble was not established in nonthermal emission and the shock strength was not robustly measured in either bubble. After using eROSITA data to map the bubble edges, we analyzed edge-adjacent radio and $γ$-ray data to remove foregrounds, test if the southern bubble can be detected in nonthermal emission, and measure the corresponding high-latitude spectra of both bubbles. Data were stacked parallel to the eROSITA bubble edges traced by an edge detector, in the same method used previously to pick up weak signals in the smaller, nested Fermi bubbles; the detected brightness jumps were then used to measure the spectrum. We detect ($>5σ$) both bubble tips in both radio and $γ$-rays, and find a radio spectrum corresponding to high, Mach $3$-$5$ shocks. The southern bubble is fainter, by $\sim$an order of magnitude in radio, its edge propagating into a medium roughly half as dense. The results indicate that these eROSITA bubbles are older, evolved counterparts of the Fermi bubbles, arising from an earlier collimated high-energy outburst from the Galactic center.

One of the largest source of uncertainties in the predictions of stellar models comes from the internal transport mechanisms. In close massive binaries, previous theoretical studies suggest that tides systematically boost chemical mixing. However, observations do not reveal any clear period-nitrogen enrichment trend, challenging these predictions. In addition, comprehensive examinations of the interplay between tidal interactions, angular momentum and chemicals transport have so far been scarce. We investigate the interplay between tidal interactions and rotational mixing, and the impact of the angular moment transport (AMT) assumptions. We examine whether tidal interactions enhance or suppress chemical mixing by computing grids of genec binary models with various AMT treatments. In order to independently assess the role of tidal interactions, we systematically compute model variations of single stars with identical initial conditions. Our investigations reveal that tides can either enhance or suppress mixing relative to single-star models, and that the outcome is highly sensitive to the AMT assumptions. We identify a key contrast between the two types of computed models: in close systems subject to tides, magnetic models predict that the mixing efficiency is mostly determined by the orbital configuration, whereas in hydrodynamic models it also depends on the assumed initial velocity. As a result, hydro models may display non-monotonic period-enrichment trends, or even period-enrichment correlations. These results highlight the importance of the AMT assumptions in modeling binaries with tidal interactions. The sensitivity of the predictions of hydro models to initial conditions extends the size of the period-enrichment parameter space they cover, allowing them to accommodate for peculiar observed systems, i.e., with mild enrichment at short periods, or high enrichment at longer periods.

A crucial physical quantity in determining the 21-cm signal during cosmic dawn is the inhomogeneous background of Ly$α$ photons originating from the first galaxies. As these photons travel through the intergalactic medium, their scattering cross-section is often approximated as a delta function at resonance due to computational cost. That is, photons with emitted wavelengths between Ly$α$ and Ly$β$ are assumed to travel in straight lines until they redshift into the Ly$α$ resonance. However, due to the damping wing in the Ly$α$ cross-section, this approximation fails as the frequency of the photon approaches the resonant frequency, resulting in multiple scatterings events that could be separated by non-negligible distances. Some previous works studied this effect of Ly$α$ multiple scattering by running computationally heavy radiative-transfer simulations. However, robustly interpreting the cosmic 21cm signal requires exploring a large parameter space of astrophysical uncertainties, motivating more computationally-efficient approaches. Here we incorporate Ly$α$ multiple scatterings in the public, semi-numerical simulation 21cmFAST. We employ Monte Carlo simulations to study the trajectories of Ly$α$ photons on different scales. We find that the distance distributions of Ly$α$ photons with respect to the absorption point can be modeled as analytical functions that are governed by a single parameter. Upon implementing the distance distributions in 21cmFAST, we find that the multiple scattering effect is important (about 50% difference in the 21-cm power spectrum) only at high redshifts before the spin temperature is fully coupled to the kinetic temperature. Furthermore, we find that Ly$α$ multiple scattering does not enhance Ly$α$ heating, and that the combined effect is negligible, especially under realistic X-ray heating scenarios.

Cosmic voids, the large underdense regions of our Universe, have emerged over the past decade as powerful cosmological laboratories: their simple dynamics, sensitivity to local gravitational effects and cosmic expansion, and ability to span large volumes, make them uniquely suited to test fundamental physics. Fueled by advances in theory, simulations, and observations, void science has matured into a precision tool for constraining the parameters of the standard cosmological model and its possible extensions. In this review, we provide a comprehensive description of the statistical tools developed to characterize voids, the theoretical models that link them to cosmological parameters, and the methodologies used to extract information from survey data. We highlight the growing synergy between void-based observables and other cosmological probes, and showcase the increasingly stringent constraints derived from voids measured from current survey data and expected from future missions. With the advent of the next generation of galaxy surveys, voids are poised to play a central role in the future of cosmology, turning what was once regarded as emptiness into one of the most promising frontiers of fundamental science.

GRB~221009A has been the brightest gamma-ray burst (GRB) observed to date, and its afterglow has been characterised with unprecedented detail at TeV energies by LHAASO. Quite puzzlingly, it is also the most energetic GRB known. Among the riddles posed by this mysterious source, however, the sheer energetics are hardly the most intriguing: an unprecedented, narrow, luminous emission line at around 10 MeV has been uncovered by a detailed spectral analysis of \textit{Fermi}/GBM data immediately following the brightest peak in the GRB prompt emission and the peak of the TeV afterglow. As noted in the discovery article, the temporal evolution of the line properties can be explained as being due to high-latitude emission from a geometrically thin, relativistically expanding shell where annihilation of a large number of electron-positron pairs took place. We show that this interpretation yields stringent constraints on the properties of such shell, that point to a process that happens at radii typical of external shocks. We then demonstrate that the shell could have been the blastwave associated with the GRB precursor, with the line arising after pair loading of such blastwave as it was illuminated by the bright and hard radiation of the GRB main event. The scenario, which also explains the abrupt initial rise of the LHAASO afterglow, requires the progenitor of the GRB to have been surrounded by a circum-stellar medium (CSM) extending out to a few $10^{15}\,\mathrm{cm}$, with a density $n_\mathrm{ext}\sim 10^{8}-10^{9}\,\mathrm{cm^{-3}}$ reminiscent of those found from studies of Type IIn supernovae. This provides a precious clue to the nature of the progenitor of this peculiar GRB, which could also be present in other bursts that feature a long quiescence followed by a bright emission episode with a hard spectrum.

An increasing number of discoveries of isolated and quenched dwarf galaxies are challenging the idea that the present-day local environment of low-mass systems is the main determinant of their quenching. We present new Hubble Space Telescope (HST) data of one such system, the dwarf galaxy Canes Venatici C (CVn C). CVn C is a low-mass (3.4(+4.2-2.6)*10^6 M_sun) galaxy with a Tip of the Red Giant Branch distance of 8.43(+0.47-0.32) Mpc determined from the resolved stars in the HST imaging, which we also use to derive CVn C's structural parameters. CVn C's distance places CVn C in the Local Volume and in an isolated environment with the most tidally influential L* galaxy > 5Rvir away. Additional constraints from the HST color-magnitude diagram, archival Far-Ultraviolet (FUV), and neutral hydrogen (HI) data show that CVn C is quenched, with no evidence of star formation in the last 100 Myr and no detectable gas (MHI < 1.5*10^6 M_sun). Circumstantial evidence suggests that CVn C may have quenched via past interactions with the L* galaxy NGC 4631 (L_K = 10^10.4 L_sun), and was possibly sent on an extreme backsplash orbit by the tidal dissolution of a subhalo group. However, other quenching mechanisms-such as stripping via the cosmic web-cannot be ruled out. CVn C adds to the growing number of quenched dwarf galaxies in under-dense environments, a population that will be critical to defining the mass and environment regimes in which different quenching mechanisms operate.

The Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) will produce unprecedented volumes of heterogeneous astronomical data (images, catalogs, and alerts) that challenge traditional analysis pipelines. The LSST Dark Energy Science Collaboration (DESC) aims to derive robust constraints on dark energy and dark matter from these data, requiring methods that are statistically powerful, scalable, and operationally reliable. Artificial intelligence and machine learning (AI/ML) are already embedded across DESC science workflows, from photometric redshifts and transient classification to weak lensing inference and cosmological simulations. Yet their utility for precision cosmology hinges on trustworthy uncertainty quantification, robustness to covariate shift and model misspecification, and reproducible integration within scientific pipelines. This white paper surveys the current landscape of AI/ML across DESC's primary cosmological probes and cross-cutting analyses, revealing that the same core methodologies and fundamental challenges recur across disparate science cases. Since progress on these cross-cutting challenges would benefit multiple probes simultaneously, we identify key methodological research priorities, including Bayesian inference at scale, physics-informed methods, validation frameworks, and active learning for discovery. With an eye on emerging techniques, we also explore the potential of the latest foundation model methodologies and LLM-driven agentic AI systems to reshape DESC workflows, provided their deployment is coupled with rigorous evaluation and governance. Finally, we discuss critical software, computing, data infrastructure, and human capital requirements for the successful deployment of these new methodologies, and consider associated risks and opportunities for broader coordination with external actors.

The persistent Hubble tension and the lack of a fundamental explanation for dark energy motivate the exploration of alternative mechanisms capable of reproducing late-time cosmic acceleration. In this work, we revisit gravitationally induced particle creation as a phenomenological non-equilibrium process that can effectively mimic a dynamical dark-energy component. Within the thermodynamic framework of open systems, we model the production of an unspecified particle species with constant intrinsic equation-of-state parameter and consider four phenomenological parametrisations of the particle-creation rate. The modified continuity and Friedmann equations lead to an effective negative pressure and a redshift-dependent effective equation of state, which we constrain using Cosmic Chronometers, Pantheon+ supernovae, DESI DR2 BAO, a compressed CMB likelihood, and SH0ES data. Using the full dataset combination, we find that particle-creation models provide fits comparable to $Λ$CDM, yielding $H_0 \simeq 69.3\,\mathrm{km\,s^{-1}\,Mpc^{-1}}$ and present-day effective dark-energy equation-of-state values close to $w^{\rm eff}_{\rm DE}(0)\simeq -1$, with all models predicting an accelerating Universe ($q_0\simeq -0.55$). When the Hubble tension is assessed using early- and late-time dataset splits, particle-creation scenarios reduce its statistical significance to the $\simeq 2.4σ$--$3σ$ level, compared to the $4.3σ$ discrepancy obtained in $Λ$CDM. Although deviations from $Λ$CDM remain mild and Bayesian model comparison indicates no statistical preference between models, gravitationally induced particle creation emerges as a viable late-time extension of the standard cosmological model and provides a consistent phenomenological framework for exploring departures from $Λ$CDM.

The evolution of radio jet morphology and its energetics is significantly influenced by the environment in which the host galaxy resides. As giant radio galaxies (GRGs) often extend to the scale of entire galaxy clusters ($\sim$Mpc) and beyond, they are a suitable class of objects for studying jet--intracluster medium interactions. This paper presents a multiwavelength study of a GRG, J1007+3540, using the LOFAR Two-metre Sky Survey second data release (LoTSS DR2) at 144 MHz and the upgraded Giant Metrewave Radio Telescope (uGMRT) at 400 MHz. The source has a projected linear extension of 1.45 Mpc and is hosted by MaxBCG J151.77665+35.67813, within the WHL 100706.4+354041 cluster. At both frequencies, the source exhibits clear signatures of recurrent jet activity, a one-sided, extended, tail-like diffuse structure with a morphological break in the tail. The estimated radiative ages of the inner lobes and outer north lobe are $\sim$140 Myr and $\sim$240 Myr, respectively. In addition to the radio analysis, we performed optical--to--infrared spectral energy distribution modelling. The host galaxy is an evolved elliptical system with a stellar mass of $\log_{10}(M_\star/M_\odot) = 11.0$ and an old stellar population age of $\sim$12 Gyr. The high infrared-derived star formation rate ($\sim106~M_\odot$~yr$^{-1}$) of the source implies significant dust-obscured star formation, potentially linked to merger-driven gas inflows. J1007+3540 presents a rare combination of a restarted jet, a detached tail-like structure, and unusual spectral flattening beyond the tail break, which is very rare to report together in a GRG. This rare and remarkable system offers a unique laboratory for probing the interplay between active galactic nucleus activity, star formation, and environmental effects in cluster-surrounded GRGs.

We construct posterior distributions of the equation of state (EoS) for matter beyond the inner crust of neutron stars by incorporating finite nuclei (FN) constraints within relativistic mean field models. These constraints are implemented in three complementary ways: (i) through theoretical bounds on the EoS, (ii) implicitly via nuclear matter parameters, and (iii) explicitly by enforcing consistency with experimental binding energies and charge radii of selected nuclei. The resulting low-density nucleonic EoSs are subsequently matched to a model-agnostic speed-of-sound parametrization, constrained by astrophysical observations, including NICER mass-radius measurements, tidal deformability limits from GW170817, and lower bounds on the maximum neutron-star mass inferred from radio pulsar observations. We find that the admissible range of the transition density is strongly sensitive to the choice of the low-density EoS. In particular, the inclusion of explicit FN constraints significantly reduces the allowed parameter space of the nucleonic EoS at low densities, narrowing the transition-density range by nearly a factor of two. Consequently, neutron-star properties inferred from EoSs with explicit FN constraints differ substantially, with especially pronounced effects for low-mass neutron stars and their correlations with nuclear matter parameters. A quantitative comparison, using metrics based on Mahalanobis distance, shows consistency of the explicit constraints with PSRs J0740+6620, J0030+0451, and J0437-4715, but suggest a possible tension with PSR J0614-3329. These findings underscore the critical importance of a consistent treatment of finite-nuclei properties for reliably inferring the behavior of high-density matter and the presence of possible phase transitions from astrophysical observations.

We present multi-wavelength observations of the first recorded low state of the intermediate polar BG CMi. Optical monitoring of the source by members of the American Association of Variable Star Observers reveals a decrease of ~0.5 mag that lasted ~50 d in early 2025. During the low state the optical timing properties imply that BG CMi underwent a change in the accretion mode, as power at the spin frequency $ω$ dramatically dropped. An XMM-Newton observation revealed a substantial decrease in intrinsic absorption and a slight increase in intrinsic X-ray luminosity, compared to archival Suzaku data. Timing analysis of the X-ray light curves shows that power shifted from the orbital frequency $Ω$ (prominent in Suzaku data) to $2Ω$ in the low state XMM-Newton data, along with the strengthening of certain orbital sidebands. We suggest that BG CMi transitioned to disk-overflow accretion, where the white dwarf accreted matter via both a disk and a stream, the latter becoming more dominant during the low state due to a decrease in the mass and size of the disk.

Compact object mergers are multi-messenger sources and progenitors of some gamma-ray bursts (GRBs), primarily understood by gamma-ray observations, while poorly constrained in the prompt low-energy phase. A long-lasting X-ray emission was discussed as afterglows following several short-duration ($\lesssim$2 s) bursts, yet this prompt X-ray component was not directly observed or confirmed. Here we report the discovery of a minutes-long ($\sim$560 s) flash of soft X-rays immediately following the short ($\sim$0.4 s) GRB 250704B. The long-soft bump points to a distinct phase of prompt emission in X-rays detected by Einstein Probe in an event that otherwise appear as an ordinary short GRB, showing that long-lasting X-ray emission is likely a common feature of merger-driven bursts and a promising electromagnetic counterpart to gravitational-wave sources.

This is a collection of papers presented by the JEM-EUSO Collaboration at the 39th International Cosmic Ray Conference (ICRC 2025) (Geneva, Switzerland, July 14--24, 2025).

Fast Radio Bursts (FRBs) are millisecond-duration radio transients of extragalactic origin. Classifying repeating FRBs is essential for understanding their emission mechanisms, but remains challenging due to their short durations, high variability, and increasing data volume. Traditional methods often rely on subjective criteria and struggle with high-dimensional data. In this study, we apply an unsupervised machine learning framework that combines Uniform Manifold Approximation and Projection (UMAP) and Hierarchical Density-Based Spatial Clustering of Applications with Noise (HDBSCAN) to eight observed parameters from FRB 20220912A. Our analysis reveals three distinct clusters of bursts with varying spectral and fluence properties. Comparisons with clustering studies on other repeaters show that some of our clusters share similar features with sources such as FRB 20201124A and FRB 121102, suggesting possible common emission mechanisms. We also provide qualitative interpretations for each cluster, highlighting the spectral diversity within a single source. Notably, one cluster shows broadband emission and high fluence, which are typically seen in non-repeating FRBs. This raises the possibility that some non-repeaters may be misclassified repeaters due to observational limitations. Our results demonstrate the utility of machine learning in uncovering intrinsic diversity in FRB emission and provide a foundation for future classification studies.

The luminous accreting pulsar SMC X-1 is an appropriate target to explore the accretion dynamics. SMC X-1 shows unique quasi-periodic flux variability of 40-65$\,$days known as superorbital modulation. To constrain the accretion structure of SMC X-1 based on timing and spectral study, we have analysed X-ray data of SMC X-1 observed by Suzaku and NuSTAR at various epochs between 2011 and 2022. The spectral analysis shows that the hydrogen column density ($N_\mathrm{H}$) increases from $1.1 \times 10^{22}\,\mathrm{cm^{-2}}$ to $1.24 \times 10^{23}\,\mathrm{cm^{-2}}$ as the flux decreases with the superorbital modulation. The neutral iron K$α$ line at 6.4$\,$keV has a broad width of 0.3$\,$keV, and its equivalent width increases as toward superorbital low states. The line broadening is consistent with Keplerian motion at the inner disc rather than the stellar wind velocity of the donor star. These findings support that the superorbital modulation is a consequence of X-ray attenuation by the warped accretion disc. To test this interpretation, we have conducted photon transport simulations of a system consisting of a neutron star, a warped disc, and optically-thin disc atmosphere. Occultation of the central source by the disc successfully reproduces the observed variations in the equivalent width of neutral iron K$α$ line, pulse profiles, and flux in hard X-rays. Notably, a disc precession angle of approximately $30^\circ$ can account for the observational features. For the radiation pattern of the photon source, the preferred beam width corresponds to a standard deviation of $30^\circ$.

Pulsar timing arrays (PTAs) have recently reported compelling evidence for the presence of a gravitational-wave background signal. Mapping the gravitational-wave background is key to understanding how it is formed, since anisotropy is a tracer for, for example, a supermassive black hole binary origin. In this work we refine the frequentist regularised gravitational-wave mapping analysis developed in our previous work (as part of the MeerKAT PTA 4.5-year data release). We derive a point-spread function describing the angular resolution of a PTA. We investigate how the point spread function changes for different PTA constellations and determine the best possible angular resolution achievable within our framework. Using simulated data, we demonstrate that previous methods do not capture the actual resolution - especially in regions of the sky with a high density of pulsars. We propose an improved scheme that accounts for a variable local resolution and test it using realistic simulations of the latest MeerKAT dataset. We demonstrate that we are able to identify a continuous gravitational wave signal in a region with good pulsar sky coverage with approximately a factor of two increase in significance compared to our previous method.

PSR J2021+4026 is a remarkable $γ$-ray pulsar exhibiting repeated transitions between high $γ$-ray flux (HGF) and low $γ$-ray flux (LGF) states. With 17-yr Fermi-LAT monitoring, we reveal persistent secular evolution and enhanced spin-down rate variability within individual emission states -- beneath the quasi-periodic state transitions. After removing discrete jumps, the jump-corrected flux $δF_γ$ shows a three-phase evolution: rise ($+2.02^{+0.17}_{-0.15}\%~\mathrm{yr}^{-1}$), decline ($-3.72^{+0.34}_{-0.47}\%~\mathrm{yr}^{-1}$), and rapid rise ($+14.9^{+6.4}_{-4.4}\%~\mathrm{yr}^{-1}$), with all rates quoted relative to the long-term mean flux $\langle F_γ\rangle=7.8\times 10^{-10}\,\mathrm{erg}\,\mathrm{cm}^{-2}\,\mathrm{s}^{-1}$. Moreover, the flux of the LGF state is gradually approaching the stable HGF level at a rate of $+0.72 \pm 0.11\%~\mathrm{yr}^{-1}$. These results demonstrate that secular flux evolution in PSR J2021+4026 operates largely independently of discrete state transitions, yet jointly with them drives the system toward a stable high-flux equilibrium.

We investigated the properties of boxy/peanut-shaped (B/PS) bulges in a sample of 71 galaxies from the Edge-on Galaxies in the Pan-STARRS Survey (EGIPS) and 20 simulated galaxies from Illustris TNG50 using multicomponent photometric decomposition. For each real and simulated galaxy, we obtained a suitable photometric model in which the B/PS bulge was represented by a dedicated 2D photometric function. For real galaxies, we found that more flattened X-structures are generally residing in larger B/PS bulges. When tested against the galaxy masses, we verified that both larger bulges and more flattened X-structures are typically found in more massive galaxies. Since large bars are also known to reside in more massive galaxies, we conclude that the flatness of X-structures in larger B/PS bulges has a physical origin, rather than being solely a result of projection effects due to differences in observed bar viewing angles. When comparing the properties of B/PS bulges between EGIPS galaxies and TNG50 galaxies, with bars rotated for different viewing angles, we found that B/PS bulges in TNG50 are considerably smaller and less luminous in terms of total intensity. This is consistent with previous studies of bar properties in TNG50, indicating the B/PS bulges in TNG50 differ from those in real galaxies, as do their parent bars.

Galaxies in the nearby Universe, particularly dwarf systems, exhibit inner mass profiles of dark matter haloes that systematically depart from canonical cold dark matter expectations, signalling an interplay between baryonic feedback and the collisionless halo. We update an analytical cusp-core transition model by incorporating the effect of supernova-driven mass loss. Adapting this model to SPARC galaxies, we measure the energy conversion efficiency epsilon, defined as the fraction of supernova feedback energy that is used to change the central dark-matter potential. We find epsilon ~ 0.01 for nearby SPARC galaxies. Building on these measurements, we compare the dynamical energy required for a cusp-core transformation with the feedback energy available over burst cycles and identify a cusp-core transition forbidden region on the halo-stellar mass plane where transformation cannot occur. Galaxies with halo masses from 10^8 to 10^11 M_sun lie outside the forbidden region, whereas ultra-faint dwarf galaxies < 10^8 M_sun, galaxy groups and clusters > 10^11 M_sun fall within it, consistent with their high central densities and the inefficiency of core formation at very low and very high masses. This approach also explains the observed diversity of inner density profiles in low-mass systems, showing that both the star formation rate and the energy conversion efficiency govern them, with the latter emerging as a key parameter setting the strength of the cusp-core transition. Beyond the cusp-core problem, our observationally inferred energy conversion efficiency provides a model independent benchmark that strongly constrains galaxy formation models.

In this study, we revisit the well-known cosmic microwave background (CMB) anomaly referred to as Hemispherical Power Asymmetry (HPA), using CMB temperature maps from the Planck mission public release 4 (PR4) and the WMAP nine-year data release. Employing the Local Variance Estimator (LVE) method, we systematically reexamine the properties of HPA to investigate possible frequency dependence as well as scale dependence in its amplitude and direction. We model the HPA as a scale-dependent dipole modulation following a power-law form, rather than assuming a scale-invariant case. Our analysis incorporates seven cleaned frequency-specific CMB temperature maps from both the Planck and WMAP missions to test the robustness of the observed asymmetry across instruments and frequency channels. We find that the dipolar modulation characteristic of HPA is present in all cases examined, with consistent estimates of the preferred direction and scale-dependent variation in dipole amplitudes. These results support the conclusion that the observed asymmetry is unlikely to arise from instrumental artifacts or data-processing effects, and instead points toward a persistent large-scale feature of the CMB sky with a possible cosmological origin.

Doppler lensing, a relativistic effect resulting from the peculiar velocities of galaxies along the line of sight, provides insight into the large-scale structure of the Universe. Relativistic simulations are essential for modeling Doppler lensing because they incorporate gravity and motion in spacetime. We compare two relativistic $N$-body simulation frameworks, $\texttt{GEVOLUTION}$ and $\texttt{SCREENING}$, to calculate Doppler lensing convergence in cosmic voids of different sizes and halos of different masses. Our analysis reveals scale-dependent performance: $\texttt{SCREENING}$ shows larger differences in small voids (radius range: 15--25 Mpc/h) with a mean absolute relative difference of 38.5\%, due to linearized dynamics failing in nonlinear regimes. Medium voids (25--35 Mpc/h) show better agreement (9.5\% mean difference). For large voids (35--45 Mpc/h), $\texttt{SCREENING}$ exhibits intermediate differences (16.9\% mean difference) with central instabilities. Moreover, our Doppler convergence analysis with massive halos ($10^{11.5}$--$10^{14} {~h^{-1}\mathrm{M}_\odot}$) demonstrates excellent consistency (1.6--3.6\% mean difference). These findings provide clear guidance for simulation choice: $\texttt{GEVOLUTION}$ is recommended for precision studies critical to $Λ$CDM or modified gravity tests, while $\texttt{SCREENING}$ offers a computationally efficient alternative for relativistic treatments with large catalogs of voids and halos, assisting future astrophysical surveys.

Type Ia supernovae are a cornerstone of modern cosmology, providing first evidence for cosmic acceleration and new tests of dark energy. Son et al. 2025 (S25) claim a strong redshift evolution in standardized supernova luminosities driven by supernova progenitor age, with dramatic cosmological implications: rapidly evolving dark energy, decelerating expansion, and a $9σ$ tension with $Λ$CDM. We show that the underpinning evidence required for this conclusion -- the supernova progenitor-age dependence, the redshift-dependent age difference, and their combined impact -- is either negligible or relies on effects already corrected for in modern supernova analyses. First, the S25 analysis omits the standard host-galaxy stellar mass correction that captures known environmental dependencies that also correlate with stellar age. Applying this correction to the S25 sample, we find no dependence of standardized supernova brightness on host age. Independent data also show no significant difference at low-redshift in standardized brightness between star-forming galaxies and several Gyr older quiescent galaxies of the same stellar mass. Second, the S25 scenario predicts strong redshift evolution of the host-mass effect. Data from the Dark Energy Survey supernova survey measure evolution of $-0.028 \pm 0.034~\mathrm{mag}\,z^{-1}$, consistent with zero and altering the dark-energy equation-of-state measurement ($w$) by $<$0.01 if included. Third, we demonstrate that the claimed $\sim5$~Gyr progenitor age difference between nearby and distant supernovae is overstated by factors of three to five largely due to a conflation of host galaxy age with supernova progenitor age. We conclude that type~Ia supernova cosmology remains robust for current measurements of dark energy.

Dwarf galaxies have a wide variety of structures, such as dark matter (DM) distribution, stellar-to-halo mass ratio, and stellar density. Recent high-resolution simulations have shown a variety of stellar-to-halo mass ratios for dwarf galaxies with a DM halo mass of $\sim 10^9 M_{\odot}$ at $z=0$. In this study, we performed cosmological $N$-body/smoothed-particle hydrodynamic zoom-in simulations of dwarf galaxies with the highest gas and DM particle mass resolutions of 2.37 $M_{\odot}$ and 12.8 $M_{\odot}$, respectively. The stellar-to-DM halo mass ratio of one of our simulated dwarf galaxies was $\sim 10^{-4}$, typical for satellites of the Milky Way. The stellar mass ($10^5 M_{\odot}$) and half-mass radius (68 pc) were also similar to those of the satellites of the Milky Way. The power-law slope of the DM halo was $α= -1.1$. On the other hand, the other simulated galaxy exhibited a stellar-to-halo mass ratio of $\sim 10^{-3}$ and a steeper power-law slope ($α=-1.9$) than the other; the presence of baryonic matter deepened the cusp. The mass of $>10^6 M_{\odot}$ and a half-mass radius of $\sim 36$ pc of this galaxy were similar to those of ultra-compact dwarf galaxies rather than the satellites of the Milky Way. This DM halo grew in mass earlier than the former one, and the central DM density was higher than that of the other even in the DM-only simulations.

We present upgraded Giant Metrewave Radio Telescope (uGMRT) observations of 0731+438, an \mbox{FR II} radio galaxy at a redshift of 2.429 with two lobes separated by 82 kpc. A blueshifted, faint and broad \mbox{H{\sc i}} 21 cm absorption line with velocity full width at half maximum (FWHM) $\sim 600\,\rm km\,s^{-1}$ is detected against the southern radio lobe that is 47 kpc from radio core, indicating a neutral hydrogen outflow associated with jet-gas interaction beyond the host galaxy. The outflow has a mass outflow rate of $\sim\,0.4T_{\rm s}Ω\rm\, M_\odot\,{\rm yr}^{-1}$, which could increase to $\sim\,4.0T_{\rm s}Ω\rm\,M_\odot\,{\rm yr}^{-1}$, corresponding to an energy outflow rate of $2.4T_{\rm s}Ω\times10^{40}$ -- $1.5T_{\rm s}Ω\times10^{41}\,\rm erg\,s^{-1}$, where $T_{\rm s}$ is the spin temperature and $Ω$ is the solid angle of the outflow. Previous optical observations identified an extended emission line region aligned with the radio axis, ionized by the central Active Galactic Nucleus (AGN). Within this region, a warm and ionized outflow with a mass outflow rate of $\sim\,50\rm\, M_\odot\,{\rm yr}^{-1}$ and an energy outflow rate of $\sim1.7\times10^{43}\,\rm erg\,s^{-1}$ was detected. We propose that both the extended emission line region and the optical outflow are results of synergistic effect between jet and AGN radiation. The AGN likely exerts negative feedback on the host galaxy, as evidenced by the gas expulsion by the jet and the high velocity dispersion of ionized gas observed optically. So far, detections of jet-driven neutral hydrogen outflows remain rare. The high redshift, large outflow radii, substantial mass outflow rate and energy outflow rate of the neutral hydrogen outflow in 0731+438 expand the known parameter space of such outflows.

We report the third observation of the Seyfert-1.9 active galactic nucleus (AGN) MCG-05-23-16 with the Imaging X-ray Polarimetry Explorer (\textit{IXPE}), together with optical spectro-polarimetry obtained at the Very Large Telescope (VLT), and combined with archival near-ultraviolet, optical and near-infrared polarimetric data. No X-ray polarization was detected in the 2-8 keV band, with a 99\% confidence upper limit of $\leq$2.9\%, further reduced to $\leq$2.5\% when combined with the two past IXPE observations of the same target. Monte Carlo simulations suggest that equatorial coronal models are disfavored if the AGN is indeed a type-1.9/2 AGN, but coronae coplanar to the accretion disk remain consistent if the source is less inclined than previously assumed. \textit{VLT}/FORS2 data reveal a typical type-2 spectrum in total flux, a broad H$α$ line in polarized flux, and strongly wavelength dependent polarization degree and angle, rotating by nearly 70$^\circ$ across the optical band. Comparison with historical measurements confirms long-term stability of the polarization spectrum and a $\sim$90$^\circ$ rotation in the near-ultraviolet. Interpreting the multi-wavelength polarization relative to the AGN ionization axis indicates that the main obscurer is not a compact circumnuclear torus, but a distant kpc-scale dust lane crossing the galaxy. This result implies that MCG-05-23-16 is in fact a type-1 AGN seen through foreground dust. The low X-ray column density becomes consistent with the absence of polarization, provided that the nuclear inclination is low.

We present imaging and spatially resolved spectral analyses of eight Chandra data taken for the Circinus galaxy in $\approx$ 22 years to reveal neutral iron-K$α$ emission on a circumnuclear scale ($\sim$ 10--100 pc) and search for time variability in the emission. By simulating and taking account of point-source emission from the active galactic nucleus (AGN), we detect iron-line emission $\sim$ 20--60 pc away from the nucleus, particularly in the eastern and western regions. In the two regions, possible time variability in the line flux was also detected. Our spectral analysis then finds that the observed equivalent widths can reach $\sim$ 2 keV and the slopes of underlying continua are rather inverted with $Γ< 0$. These are consistent with a scenario in which the iron emission originates from clouds illuminated by AGN X-rays; our result could provide the first extragalactic example of AGN X-ray echoes. In this scenario, we estimated the physical sizes of the illuminated clouds based on the timescale of variability to be less than 6 pc. Furthermore, we compared the iron emission distribution with the cold molecular distribution inferred by Atacama Large Millimeter/submillimeter Array (ALMA) observation of CO($J$=3--2), revealing that in the region of bright iron-line emission, the molecular emission seems to be weak. This might suggest that the AGN X-ray emission affects the chemical composition in the form of AGN feedback.

Using new Institut de Radioastronomie Millimétrique (IRAM) 30m telescope $\rm N_2H^+$, $\rm C^{18}O$ $J$=1-0 and Atacama Pathfinder Experiment (APEX) telescope $\rm ^{13}CO$ and $\rm C^{18}O$ $J$=2-1 maps together with archival far-infrared continuum data, and $\rm ^{12}CO$, and $\rm ^{13}CO$ $J$=1-0 data, we present a comprehensive analysis of the massive filament CFG024.00$+$0.48 (G24) across clump-to-cloud scales. Our results show that G24 is an $\sim$80 pc giant filament with a total mass of $\sim$$10^5$ M$_{\odot}$. In the different tracers the filament width is measured to be about $\sim$2 times the beam size of the observations, as expected for power-law density distributions, giving beam-deconvolved widths in the range from 0.8 to 2.8 pc. We determine a line-of-sight thickness of $\sim$2.2 pc demonstrating that G24 is not an edge-on, flatten structure. The virial parameter obtained from line mass ($α_{\rm line,vir}=M_{\rm line,vir}/M_{\rm line}$) from the $\rm C^{18}O$ (1-0) data is 0.85, and that obtained from $Herschel$-based H$_2$ column density is 0.52, suggesting G24 is globally close to virial equilibrium. The distribution of the 40 dust clumps appears to have a ''two-tier'' fragmentation pattern. For the clump groups, the separation, with a mean/median of 3.68/3.46 pc, is very close to expected length associated with the maximum fragmentation growth rate of $λ_{\rm max}=3.55 \pm0.32$ pc estimated for the dust. However, the longitudinal centroid velocity profiles of $\rm C^{18}O$ and $\rm N_2H^+$ show oscillation patterns with wavelengths of 9.8$\pm$0.1 pc and 9.9$\pm$0.1 pc, respectively. This is $\sim$2 times larger than the corresponding values of $λ_{\rm max}$ of 4.96$\pm$0.63 pc and 4.65$\pm$1.34 pc, respectively. This suggests that the velocity structure is not dominated by flows directly associated with the fragmentation seen in the dust emission.

We analyse a sample of 225 star-forming regions from the DESIRED-E project, each with simultaneous determinations of the electron temperature from ionized nitrogen and oxygen, $T_{\rm e}$([NII]) and $T_{\rm e}$([OIII]), respectively. We derive new empirical relations connecting the gas-phase metallicity to the global electron temperature, $T_{\rm e}$(H$^+$), as determined via radio observations. We establish two calibrations: one assuming a homogeneous temperature distribution ($t^2 = 0$, the ``direct method''), and another accounting for internal temperature fluctuations ($t^2 > 0$). Applying these calibrations to 460 radio observations of Galactic HII~regions spanning Galactocentric distances from $\sim0.1$ to 16 kpc, we determine the radial O/H gradient in the Milky Way under both assumptions. We further compare these nebular gradients to independent metallicity estimates from young O- and B-type stars and Cepheid variables. We find that the $t^2 > 0$ calibration yields a gradient in excellent agreement with stellar-based determinations, whereas the $t^2 = 0$ method underestimates metallicities by up to $\sim$0.3 dex. This discrepancy cannot be reconciled by invoking oxygen depletion onto dust grains or nucleosynthetic processing via the CNO cycle in massive stars. We also find that one widely used relation in the literature, assuming $t^2 = 0$, produces an excessively steep gradient -- likely due to the use of outdated atomic data and pre-CCD observations. Finally, we explore potential azimuthal variations in the Galactic metallicity distribution driven by the presence of the spiral arms, finding no evidence for variations larger than $\sim$0.1 dex with respect to the general radial gradient.

Radio halos are megaparsec-scale diffuse, non-thermal radio sources located at the centers of galaxy clusters, tracing relativistic particles and magnetic fields in the intra-cluster medium. Their origin is generally attributed to cluster mergers that generate turbulence and re-accelerate aged electrons. We study the diffuse radio emission, spectral properties, and the connection between thermal and non-thermal components in the massive galaxy clusters Abell 773 and Abell 1351 ($M_{500} \sim 7 \times 10^{14}\,M_{\odot}$), both of which are dynamically disturbed. We combine LOFAR LoTSS-DR2 observations at 144 MHz with uGMRT observations at 650 MHz, supplemented by archival XMM-Newton X-ray imaging. We confirm that both clusters host radio halos extending up to a largest linear size of $\sim 2$ Mpc. We measure an integrated spectral index $α_{144}^{650} \sim -1.0$ for both clusters. The radio halo in Abell 773 resembles a classical halo and follows a sublinear radio--X-ray surface brightness relation. In contrast, Abell 1351 shows a more complex and asymmetric morphology, influenced by embedded radio sources including the brightest cluster galaxy, a tailed radio galaxy, and a ridge-like feature. These contaminating sources lead to deviations from the sublinear trend in the point-to-point radio--X-ray analysis of Abell 1351.

Our understanding of the $γ$-ray sky has greatly advanced, yet studying the unresolved $γ$-ray background (UGRB) can unveil the nature of the faintest $γ$-ray source populations in the Universe. Statistical cross-correlations between the UGRB and tracers of large-scale cosmic structure allow us to infer which sources contribute the most to this emission. In this work, we examine the angular correlation between the UGRB and the matter distribution traced by galaxies, using twelve years of Fermi Large Area Telescope (LAT) observations along with three years of Dark Energy Survey (DES) data. We detect a correlation with a signal-to-noise ratio of 7.96, primarily driven by large angular scales. We then perform a multi-tracer analysis that combines this measurement with the cross-correlation between $γ$ rays and DES weak lensing. The two single-tracer results are mutually consistent, and their combination yields a total significance of 8.6, firmly establishing the extragalactic origin of the UGRB. Intriguingly, the properties inferred for the sources contributing to the UGRB show departures from those of the resolved γ-ray population, suggesting that the faint end of the $γ$-ray sky is not a simple extrapolation of currently resolved sources.

Between the groundbreaking detections of stellar-mass black holes by LIGO/Virgo/KAGRA and JWST's revelation of a surprisingly abundant population of supermassive black holes, one crucial missing link remains: the elusive intermediate-mass black holes (IMBHs). IMBHs represent a key phase in the hierarchical growth of black holes, yet they have persistently evaded detection. Traditional methods, effective for both actively accreting and quiescent black holes, have largely failed to uncover this hidden population. Here, we argue that novel observational strategies--particularly time-domain variability studies of active galactic nuclei (AGN) and tidal disruption events--provide a promising path forward. Finding IMBHs will resolve critical gaps in our understanding of black hole formation and the various mechanisms driving their subsequent growth. The upcoming Vera C. Rubin Observatory, with its unprecedented capacity to monitor the dynamic sky, stands to revolutionize our ability to detect these long-sought IMBHs, shedding new light on the assembly history of black holes across cosmic time.

We analyze the polarization of optically thin, stationary, axisymmetric black hole jets at scales of order the light cylinder radius. Our work generalizes the face-on results of Gelles et al. (2025) to arbitrary viewing inclination. Due to a combination of geometry and relativistic aberration, the polarization of the jet is not left-right symmetric, and the degree of asymmetry can shed light on both the viewing angle and the plasma bulk Lorentz factor. We show that there is always a radius in the jet at which the polarization transitions from azimuthal to radial; this radius is different along the spine and limb of the jet. We propose metrics that can be used to constrain the black hole spin, inclination angle, and plasma Lorentz factor from these polarimetric signatures, and we discuss the impact of limb-brightening on these measurements. We anticipate that these polarimetric signatures can be studied with current or forthcoming data in M87, NGC 315, NGC 4261, Centaurus A, Cygnus A, and other systems. Observations of the polarization of the base of the counter-jet in higher inclination sources would provide a particularly promising probe of black hole spin.

We quantify the utility of HCN and HNC to characterize gas conditions in the nearby starburst galaxy NGC 253. We use measurements from the Atacama Large Millimeter/Submillimeter Array (ALMA) Large Program ALCHEMI: the ALMA Comprehensive High-resolution Molecular Inventory. Using different subsets of the eight total HCN and HNC transitions measured by ALCHEMI, we test the number and combinations of transitions necessary for constraining the temperature, H$_2$ volume and column densities, cosmic-ray ionization rate, and beam-filling factor in three representative regions within NGC 253. We use these combinations of HCN and HNC transitions to constrain chemical and radiative transfer models and infer the gas conditions using a Bayesian nested sampling algorithm combined with neural network models for increased efficiency. By comparing the shapes of the resulting posterior distributions, as well as the medians and uncertainties for each gas parameter, from each test case to what we obtain with the full set of eight transitions (the control), we quantify how well each test reproduces the control. We find that multiple transitions each of both molecules are required to obtain a median parameter value within a factor of 2 of the control with an uncertainty less than 2-3 times that of the control. We also find that transition combinations that feature a range of upper-state energies are most effective. We show that single transitions, such as HCN J = 1-0 or 3-2, are among the worst-performing combinations and result in parameter values up to an order of magnitude different than the control.

The matter distribution in the Universe exhibits a rich variety of structures forming the cosmic web. These structures arise from the anisotropic gravitational collapse of primordial density fluctuations and define the pathways along which galaxies flow from voids to high-density clusters. Local density variations within these structures play a fundamental role in driving the environmental evolution of galaxies. To characterise filament boundaries, we analysed galaxy overdensity profiles around filaments in two redshift ranges: $0.05 < z < 0.1$ and $0.1 < z < 0.3$. Perpendicular and parallel profiles were derived by averaging galaxy overdensity as a function of distance. Characteristic scales and central overdensities were then analysed by fitting analytical models, specifically exponential and power-law families. We also introduced normalised density profiles to account for survey incompleteness. The perpendicular overdensity profiles show a nearly constant value in the central regions $D_{fila} < 1$ Mpc, decreasing at distances up to $\approx 10$ Mpc. The mean physical widths (scale radii) at $0.05 < z < 0.1$ and $0.1 < z < 0.3$ are $2.39 \pm 0.69$ and $5.56 \pm 2.29$ Mpc, respectively. This scale difference between redshift ranges is also evident in the normalised profiles. Conversely, profiles along filaments remain constant at distances larger than $\approx 20$ Mpc from the nearest intersection. Our results show that the influence zone of cosmic filaments extends up to $\sim 10$ Mpc from their spines. Furthermore, a mild evolution in structural parameters is observed over the past $\sim 4$ Gyr, suggesting that filaments undergo measurable changes even at relatively low redshifts.

The microlensing event OGLE-2011-BLG-0950 exhibits the well-known ``Planet/Binary'' degeneracy, in which distinct lens configurations produce similar light curves but imply substantially different mass ratios between the lens components. A previous study suggested that high-resolution imaging could break this degeneracy through differences in the lens-source relative proper motion. In this work, we identify a new planetary model for this event that arises from a newly identified degeneracy, simultaneously reproducing the observed light curve and remaining consistent with the relative proper motion measured from high-resolution imaging. By combining constraints from the light-curve modeling and high-resolution observations, we infer a lens system consisting of a $\sim 1~M_{\odot}$ host star orbited by a $\sim 1.5~M_{\rm Jup}$ planet, with a projected separation of about 2 or 8 au, subject to the ``Close/Wide'' degeneracy. Our reanalysis of the color-magnitude diagram further indicates that the source star has unresolved companions that contribute non-negligible blended light, highlighting the importance of carefully accounting for source and lens companions in future Roman microlensing analyses. Finally, we show that adopting a single mass--luminosity relation significantly underestimates the uncertainties in the inferred lens properties for host masses $\gtrsim 1~M_{\odot}$.

Covariance matrices are essential cosmological probes of fundamental physics, providing information on numerous fundamental physical parameters and varying with any change in the underlying cosmology. However, this cosmology dependence, while providing excellent information, also makes them computationally intensive to compute, as a new covariance matrix must explicitly be calculated for every variation in cosmology before comparisons to observational data can be made. In this paper, we develop an efficient model for estimating the parameter dependence of the covariance matrix of the matter power spectrum by Taylor expanding around a known value of the parameter space. This method allows us to use a relatively small number of input cosmologies, specifically one fiducial cosmology and two further cosmologies for each parameter. We explicitly calculate the covariance matrices for these cosmologies and then develop a new model that allows us to interpolate from these the form of the covariance matrix with a cosmology that is located elsewhere in that given parameter space without explicit perturbation theory calculations. This method speeds up covariance matrix calculations in new cosmologies by orders of magnitude compared to explicit perturbation theory calculations at each point in a given parameter space. Using different approximations, we develop three versions of our interpolated covariance matrix and validate the model by recreating all of our input cosmologies using all three forms, both with and without super-sample covariance corrections in each case, and show that the models provide robust recreations of the original results, with the different approximations being valid in certain regimes.

We present a comprehensive temporal and spectral study of the flat-spectrum radio quasar PKS~0402$-$362 using \textit{Fermi}-LAT/Swift-XRT/UVOT observations spanning from MJD 54686-60321. The $γ$-ray light curve exhibits multiple phases of enhanced activity, with the fractional variability parameter ($F_{\mathrm{var}}$) showing larger amplitudes at longer timescales, consistent with variability trends observed in other FSRQs. Statistical analysis of the flux and spectral index distributions using the Anderson--Darling test and histogram fitting reveals that both distributions deviate from a single log-normal form and are better represented by a double log-normal profile, indicating two distinct flux states. A search for quasi-periodic oscillations in the $γ$-ray emission using the Lomb--Scargle periodogram identified a significant periodic signal at $\sim$413~days with a confidence level exceeding $3σ$. However the proximity of the timescale to one year and limited number of observed cycles prevents a definitive interpretation. Broadband spectral energy distributions for six flux states were modeled using a one-zone leptonic framework incorporating synchrotron, synchrotron self-Compton (SSC), and external Compton (EC) components. The SEDs are well reproduced with physically reasonable parameters: high-flux states exhibit harder electron spectra and lower magnetic field strengths ($B \sim 0.2--0.6\,\mathrm{G}$), while low-flux states show softer spectra and stronger magnetic fields ($B \sim 1.3\,\mathrm{G}$). The fitted break energy decreases during high-flux states, suggesting enhanced radiative cooling and a transition toward a particle- or kinetic-energy-dominated jet. These trends are consistent with the ``harder-when-brighter'' behavior commonly observed in blazars.

Carbon monoxide is thought to be the second most abundant molecule in the Universe. This makes observation of both its parent isotopologue ($^{12}$C$^{16}$O) and its stable isotopologues, $^{13}$C$^{16}$O, $^{12}$C$^{18}$O, $^{12}$C$^{17}$O, $^{13}$C$^{18}$O and $^{13}$C$^{17}$O, important in variety of objects. Here the MARVEL (Measured Active Rotational-Vibrational Energy Levels) algorithm is used to determine precise rotational vibrational energy levels for the five minor isotopologues of carbon monoxide in their electronic ground state. A review of 27 literature sources yields 3716, 1454, 89, 728 and 57 validated transitions for $^{13}$C$^{16}$O, $^{12}$C$^{18}$O, $^{12}$C$^{17}$O, $^{13}$C$^{18}$O and $^{13}$C$^{17}$O, respectively, giving 863, 499, 33, 345 and 45 empirically determined, rotation vibration energy levels, respectively.

We investigate the impact of flickering variability in jet power on the luminosity and morphology of radio galaxies. We use a Lagrangian particle method together with relativistic hydrodynamics simulations using the PLUTO code to track the evolution of electron spectra through particle acceleration at shocks and cooling processes. We introduce an adapted version of this method which improves tracking of adiabatic cooling in regimes where low density jet material mixes with high density from the ambient medium in the lobes. We find that rapid increases in jet power can lead to large increases in hotspot luminosity due to the interaction of a travelling shock structure with the pre-existing shock structure at the jet head. We show that in some cases it may be possible to identify a bright region of emission corresponding to a shock travelling along the jet axis. We find that the time-averaged radiative efficiency of variable jets is similar to their steady counterparts, but find significant departures from this on an instantaneous basis. We suggest that, together with environmental effects and differences in the average powers of jets, variable jet powers may have a significant impact on how we understand the diversity of radio jets seen in observations and have significant implications for interpretations of jet powers, energy budgets and luminosity-linear size diagrams.

When an object is scattered near a supermassive black hole (SMBH), tidal oscillations excited within it reduce its orbital energy, leading to capture by the SMBH. This process, called tidal capture, can also occur when the object approaches even closer to the SMBH, resulting in a partial tidal disruption event (pTDE). Previous studies on pTDEs of main-sequence stars have shown that as the disruption intensifies, dynamical effects dominate over tidal oscillations, causing the remnant material to acquire a kick velocity instead of being captured by the SMBH. In this work, we performed hydrodynamic numerical simulations of pTDEs involving giant stars. We found that for weaker disruptions, the dynamical behavior of the remnant material resembles that of main-sequence stars. However, as the disruptions deepen, the remnant material transitions from gaining energy to losing energy, leading to capture by the SMBH. This behavior markedly differs from that of main-sequence stars, demonstrating that the presence of a compact core significantly influences the dynamical processes in pTDEs. Our simulations reveal that the energy change of the remnant material strongly correlates with asymmetric mass -- lossspecifically, the difference in mass outflow between the Lagrange points L1 and L2. This suggests that the energy change stems from asymmetric mass loss, consistent with conclusions from previous studies on main-sequence stars. However, quantitative analysis contradicts earlier models, indicating that the dynamical model of pTDEs requires further refinement. Finally, we discuss the characteristics of repeating pTDEs produced by this process and their potential observability, as well as the implications for the long-term orbital evolution of high eccentricity extreme mass ratio inspiral systems.

Ground-based observations around 1.4 $μ$m are normally limited by strong absorption of telluric water-vapor. However, Dome A, Antarctica has exceptionally dry conditions that offer a unique opportunity for observations in this band. We designed a new filter covering 1.34--1.48 $μ$m, namely $W'$, and installed it on the Antarctic Infrared Binocular Telescope (AIRBT) at Dome A in 2025. AIRBT comprises two identical 15 cm optical tube assemblies and two InGaAs cameras equipped with $J$ and $W'$ filters, respectively. With this Early Data Release (EDR), we aim to evaluate the performance of the $W'$ band at Dome A to observe objects with water-vapor features. This EDR covers $\thicksim 20 \ \mathrm{deg^2}$ in the Galactic plane using $\thicksim 20,000$ images in three nights. For 2 s exposures, the 5 $σ$ limiting magnitude histogram peaks at $J \thicksim 11.5$ mag (Vega) and $W' \thicksim 9.9$ mag, respectively. The $J-W'$ vs $J-H$ color-color diagram distinguishes ultracool candidates with water-vapor-absorption features from reddened early type stars. Furthermore, later-type stars tend to exhibit stronger water-vapor absorption. Some sources show larger $ΔW'$ than $ΔJ$ across the three nights, which we attribute to variations of their water-vapor-absorption depth. We conclude that it will be efficient to search for ultracool stars and estimate their spectral subtypes using $W'$ band imaging at Dome A, where the atmospheric transmission is high and stable.

This study extends the 3D magnetohydrodynamic (MHD) simulation of a jet emerging from a binary neutron star (BNS) merger presented in Pavan et al. (2023), in which an incipient jet was manually injected into the realistic environment imported from a previous general-relativistic MHD simulation of a merging BNS system. The jet evolution is followed up to almost 10 seconds without loss of resolution. Our results reveal that the jet faces challenges in penetrating the dense surroundings, leading to a barely successful outflow that exhibits structural asymmetries and low Lorentz factors. By the end of the extended simulation, 98% of the jet energy is converted to kinetic form and its angular structure is stabilized. The physical quantities inferred thus provide reliable inputs for afterglow emission calculations. This work demonstrates a method for simulating jets in 3D up to nearly ballistic regimes that is general and ready to be applied to any jet in a BNS merger context.

Context. In the Local Group, dwarf spheroidal galaxies (dSphs) and ultra-faint dwarf galaxies (UFDs) exhibit large velocity dispersions. These values are generally attributed to the presence of substantial amounts of dark matter (DM), in line with the predictions of the standard model of galaxy formation. However, alternative, more conservative explanations exist, such as non-virialized dynamical states induced by tidal interactions, the presence of stellar streams, and artificial inflation of the velocity dispersion caused by binary-star orbital motion. Aims. We study the dynamical evolution of UFDs using purely stellar ("dry") dynamics, without invoking DM. We dynamically evolve our systems up to a Hubble time and compare our results with observational studies and previous theoretical work. Methods. We employ direct high precision NBODY simulations performed with the NBODY6++GPU code. We explore the role of binaries in inflating the velocity dispersion of low-mass host galaxies. We also present both the stellar and dynamical evolution of the stellar population, which is necessary to properly interpret our results. Results. We find that, in all our models, the UFD remains globally quasi-stationary for approximately 3000 Myr. Subsequently, the system undergoes mass segregation and experiences a phase resembling core collapse. Red giants and white dwarfs (WD) are found to play significant, but distinct, roles. Red giants provide the dominant contribution to the luminosity, whereas WDs constitute the largest fraction of the non-luminous component, accounting for approximately 13% of the total stellar population. Finally, if not taken into account properly, velocity dispersion measurements can be strongly biased by the presence of a significant binary population, which can lead to substantial overestimates of velocity dispersion in UFDs

The AT2017gfo kilonova transient remains a unique multi-messenger event thanks to its proximity (z=0.00987) and the possibility to investigate time-resolved spectra, providing evidence of r-process nucleosynthesis. The kilonova signal was extensively studied in the spectral and time domains, giving key insights into the chemical composition and physical properties of the ejecta. Here, we report the discovery of a novel correlation between two fundamental observables: the peak energy of the EF_E spectrum, Ep, and the isotropic-equivalent luminosity, Liso. In particular, we show that up to about 2.5 days after the merger, the AT2017gfo spectrum evolves according to: log10[Ep/eV] = -0.13 (+0.02/-0.02) + 0.62 (+0.02/-0.02) * log10[Liso/(1e41 erg/s)] (68% C.L.) while in subsequent epochs, Ep remains almost constant with Liso, flattening around 1 eV. Exploiting simulations from a state-of-the-art radiative transfer code, we demonstrate that our kilonova model inherently predicts this peculiar correlation, suggesting a new diagnostic tool for comparing observables against simulations. Future kilonova observations will provide additional insight into the physics behind the Ep-Liso correlation.

We present a study of the transition of Neutral Hydrogen (HI) gas from the end of the Epoch of Reionization (EoR) to late-time large-scale structure. We examine the signature of the transition as traced through the redshifted 21-cm line with SKA-Low at $3 < z < 7$. To do so, we use the semi-numerical simulation \textsc{21cmFAST} to model the HI during the EoR and add a HI-halo based post-processing model of the late-time HI. This approach gives a robust estimate of the amplitude of the HI temperature field and predicts the observable power spectrum during the transition period. We find that our simulation pipeline reproduces the expected power spectrum trends from existing observations and theory, in addition to replicating current observational constraints on $Ω_{\text{HI}}$. Our simulations predict a drop in power of four orders of magnitude between $4 < z < 7$. Assuming an inhomogeneous recombination model, we find a flattening of the power due to lingering neutral islands masking the late-time HI signal for $5 < z< 6.5$. Using SKA-Low deep survey parameters, we find HI power spectrum detectability at scales $k \leq 1$ $h$ Mpc$^{-1}$ for redshifts $3< z < 7$, even when using the horizon limit to mitigate foregrounds. Our results suggest a sufficient SNR of the HI power spectrum tracing the underlying halos $z < 5$, which can be used for late-time cosmology. Our results suggest that the resulting $Ω_{\rm HI}$ constraints can trace different reionization scenarios such as a decreased escape fraction. This study implies that deep SKA-Low observations for $3< z< 7$ will be an important probe to constrain reionization parameters as well as cosmological models.

Pulsar timing arrays (PTAs) can detect the low-frequency stochastic gravitational-wave background (GWB) generated by an ensemble of supermassive black hole binaries (BHBs). Accurate determination of BHB merger timescales is essential for interpreting GWBs and constraining key astrophysical quantities such as black hole (BH) occupation fractions and galaxy coalescence rates. High-accuracy $N$-body codes such as \texttt{Griffin} can resolve sub-pc BHB dynamics but are too costly to explore a wide range of initial conditions, motivating the need for surrogate models that emulate their long-term evolution at much lower computational cost. We investigate neural ordinary differential equations (NODEs) as surrogates for the secular orbital evolution of BHBs. Our primary contribution is a parameterised NODE (PNODE) trained on an ensemble of $N$-body simulations of galaxy mergers spanning a two-dimensional parameter space defined by the initial orbital eccentricity and particle resolution $(e_i, N)$, with the learned vector field explicitly conditioned on these parameters. A single PNODE thereby learns a simulation-parameter-conditioned dynamical model for the coupled evolution of the BH pair's orbital state across the ensemble, yielding smooth trajectories from which stable hardening and eccentricity growth rates can be extracted. The PNODE accurately reproduces the secular evolution of the specific orbital energy and angular momentum, and the corresponding Keplerian orbital elements, for held-out trajectories, with modest generalisation to a partially unseen high-resolution case. Combining PNODE predictions with semi-analytical prescriptions for stellar hardening and gravitational-wave emission yields BHB merger timescales consistent with those obtained from direct $N$-body inputs within current theoretical uncertainties.

We report the discovery of an exceptionally eccentric binary system, BE Lyncis (BE Lyn), which hosts the nearest known black hole (BH) to Earth. Through the analysis of $\textit{TESS}$ photometry combined with an extensive set of times of maximum light spanning 39 years, we identify BE Lyn as a high-amplitude $δ$ Scuti star in a binary with an orbital period of $\approx15.9$ years and an extraordinary orbital eccentricity of $e=0.9989^{+0.0008}_{-0.0021}$ ($>0.9968$ at 95% confidence) -- the highest reliably measured for any binary system. Dynamical constraints impose an upper limit on the orbital inclination of $i \lesssim 4.0^{\circ}$, corresponding to a companion mass of $M_2 \gtrsim 17.5~M_{\odot}$, which unequivocally favors a black hole. This system provides a unique laboratory for studying asteroseismology in strong gravitational fields, the formation of black holes via asymmetric supernovae, and the evolution of extreme binary systems. Our work demonstrates, for the first time, the successful application of the light-travel time effect in a pulsating variable to unveil a dormant black hole, establishing a novel method for BH detection in non-interacting binaries.

We present a joint analysis of galaxy clustering and galaxy--galaxy lensing measurements from BOSS galaxies using a simulation-based emulation method combined with a halo occupation distribution model. Our emulators are constructed with the Aemulus $ν$ simulations, a suite of $wν$CDM $N$-body simulations with massive neutrinos as independent particle species. We combine small-scale analysis of clustering from $0.1h^{-1}$Mpc to $60.2~h^{-1}$Mpc and lensing from $1.7h^{-1}$Mpc to $60.2~h^{-1}$Mpc to perform cosmological constraints. We split the BOSS galaxies into three redshift bins to measure their clustering and employ galaxies from Dark Energy Camera Legacy Survey and Hyper Suprime-Cam as source galaxies to measure lensing separately. We find that the addition of lensing significantly improves the constraining power on $S_{8}=σ_8(Ω_m/0.3)^{0.5}$, with a weak improvement for $fσ_{8}$. Our results of $fσ_{8}$ indicate tensions of around $1\sim4σ$ below the results of CMB observations of Planck. For $S_{8}$, our results are also lower than Planck, and the tension can be mitigated when considering possible systematics in lensing measurement. As a byproduct, our analysis prefers a non-zero neutrino mass but without strong significance, with the constraining power dominated by the clustering. Given the accuracy and precision of our model and the observational data, it is anticipated that larger and higher-quality spectroscopic datasets will improve the constraints on this fundamental property in the near future.

We report the discovery of multiple compact molecular features exhibiting extremely broad velocity widths toward the W44 molecular cloud. ALMA CO $J$=3--2 data reveal eight ``Petit--Bullets'' surrounding the previously known ``Bullet.'' Each Petit--Bullet shows a distinct V-shaped structure in position--velocity space, reminiscent of the Y-shaped morphology of the Bullet, suggesting a common origin. These features are interpreted as the result of high-velocity plunges of compact gravitational objects into dense molecular gas. The spatial and kinematic properties of the Petit--Bullets suggest that the plunging material was not a single object but rather a small cluster of compact bodies. A virial mass of $1.0\!\times\! 10^{5}\, M_\odot$ inferred from their velocity dispersion is comparable to that of typical globular clusters. Momentum analysis further implies that the main Bullet likely formed by an isolated black hole. These findings provide new evidence for dynamical interactions between halo clusters and disk molecular gas.

Statistics on tidal disruption events (TDEs) may be contaminated by repeating TDEs (rTDEs), which have been extensively discovered recently. However, the origin of rTDEs remains unclear. In addition, no statistical research on rTDEs with time intervals $>5$ years has been made yet. In this work, we searched for rTDEs with time intervals of 5--19 years using CRTS data in a sample of 16 ZTF BTS TDEs at $z<0.05$. We found 2 rTDE candidates, AT 2019azh and AT 2024pvu, with time intervals of 13.2 and 17.1 years, respectively. The peak luminosities of CRTS flares are close to those of ZTF flares. For the CRTS flare of AT 2024pvu, using GALEX UV observations near the peak, we measured a blackbody temperature of $\sim19500$ K, consistent with TDEs and higher than SNe. Moreover, we estimated the expected number of SNe in the sample to be $\lesssim0.08$, and hence the probability that both CRTS flares are SNe is only 0.3\%. Therefore, the possibility that both CRTS flares are SNe can be ruled out, and it is likely that both are TDEs. Using the two rTDEs, we inferred that the TDE rate is 2--3 orders of magnitude higher than the average over 5--19 years prior to TDE detection. Considering another two rTDEs with intervals of $\sim$2 years in the sample and possible rTDEs missed by CRTS, rTDEs with intervals of $<20$ years may account for 25\%--60\% of the TDE sample. We prefer to explain rTDEs as repeating partial TDEs, but the possibility of independent TDEs cannot be ruled out and requires future observational tests.

We present the first systematic spectroscopic observations of extragalactic transients from the Dark Energy Spectroscopic Instrument (DESI), as part of the DESI Transients Survey program. With 5,000 fibers and an ${\sim} 8$ deg$^2$ field of view, we exploit DESI as a machine for the discovery and classification of transients. We present transient classifications from archival DESI data in Data Releases 1 and 2, relying on a combination of a secondary target program and serendipitous observations. We also present observations from the first 6 months of the DESI spare fiber program dedicated to transients. The program is run in coordination with a dedicated DECam time-domain survey, serving as a pathfinder for what we will be able to achieve in conjunction with the Rubin Observatory Legacy Survey of Space and Time (LSST). We classify over 250 transients, of which the majority were previously unclassified. The sample comprises thermonuclear and core-collapse supernovae and tidal disruption events (TDEs), including a TDE observed before its discovery in imaging. We demonstrate DESI's ability to classify a population of faint transients down to $r\sim 22.5$ mag during main survey operations, with negligible impacts on DESI's main observations.

Recent work has demonstrated that widely used strong-line oxygen abundance indicators, such as O3N2, $\rm R23$, and $\widehat{\rm R}$, suffer from large uncertainties when applied to high-redshift galaxies. We show that this loss of precision primarily arises because, at fixed \Oabund, galaxies span a wide dynamic range in ionization parameter and nitrogen enrichment. Here we develop updated indicators that explicitly incorporate both effects via the proxies O32 and N2O2. We define ${\rm R}_{\rm u}\equiv \rm R23+α_1 O32+α_2 N2O2$, $\widehat{\rm R}_{\rm u}\equiv \rm \widehat{R}+β_1 O32+β_2 N2O2$, and ${\rm O}_{\rm u}\equiv \rm O3N2+γ_1 O32+γ_2 N2O2$, and calibrate \Oabund~as low-order polynomials in each composite indicator. Applied to a JWST sample with $T_{\rm e}$-method abundances, the updated indicators substantially tighten the correlations with \Oabund, boosting adjusted coefficients of determination from $\mathbb{R}^2\lesssim 0$ (classical indicators) to $\mathbb{R}^2\gtrsim 0.5$ for the full sample and to $\sim 0.7$ at $z>2$. The residuals reveal a redshift evolution in the mapping between \Oabund, strong lines, ionization, and nitrogen enrichment, with a pivotal turning point near the cosmic noon ($z\sim 2$). Our calibrations provide a practical, physically grounded path to precise metallicity measurements in the JWST era and a firmer basis for quantifying early chemical enrichment and feedback.