# Research Papers — 2026-04 · bundle page 29 of 233 > Full-text markdown of papers #1401–1450 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-04/page/29 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/28.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/30.md) ### Green Shielding: A User-Centric Approach Towards Trustworthy AI Authors: Aaron J. Li, Nicolas Sanchez, Hao Huang, Ruijiang Dong, Jaskaran Bains, Katrin Jaradeh, Zhen Xiang, Bo Li, Feng Liu, Aaron Kornblith, Bin Yu Published: 2026-04-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.24700v1 Large language models (LLMs) are increasingly deployed, yet their outputs can be highly sensitive to routine, non-adversarial variation in how users phrase queries, a gap not well addressed by existing red-teaming efforts. We propose Green Shielding, a user-centric agenda for building evidence-backed deployment guidance by characterizing how benign input variation shifts model behavior. We operationalize this agenda through the CUE criteria: benchmarks with authentic Context, reference standards and metrics that capture true Utility, and perturbations that reflect realistic variations in the Elicitation of model behavior. Guided by the PCS framework and developed with practicing physicians, we instantiate Green Shielding in medical diagnosis through HealthCareMagic-Diagnosis (HCM-Dx), a benchmark of patient-authored queries, together with structured reference diagnosis sets and clinically grounded metrics for evaluating differential diagnosis lists. We also study perturbation regimes that capture routine input variation and show that prompt-level factors shift model behavior along clinically meaningful dimensions. Across multiple frontier LLMs, these shifts trace out Pareto-like tradeoffs. In particular, neutralization, which removes common user-level factors while preserving clinical content, increases plausibility and yields more concise, clinician-like differentials, but reduces coverage of highly likely and safety-critical conditions. Together, these results show that interaction choices can systematically shift task-relevant properties of model outputs and support user-facing guidance for safer deployment in high-stakes domains. Although instantiated here in medical diagnosis, the agenda extends naturally to other decision-support settings and agentic AI systems. --- ### The Chameleon's Limit: Investigating Persona Collapse and Homogenization in Large Language Models Authors: Yunze Xiao, Vivienne J. Zhang, Chenghao Yang, Ningshan Ma, Weihao Xuan, Jen-tse Huang Published: 2026-04-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.24698v1 Applications based on large language models (LLMs), such as multi-agent simulations, require population diversity among agents. We identify a pervasive failure mode we term \emph{Persona Collapse}: agents each assigned a distinct profile nonetheless converge into a narrow behavioral mode, producing a homogeneous simulated population. To quantify persona collapse, we propose a framework that measures how much of the persona space a population occupies (Coverage), how evenly agents spread across it (Uniformity), and how rich the resulting behavioral patterns are (Complexity). Evaluating ten LLMs on personality simulation (BFI-44), moral reasoning, and self-introduction, we observe persona collapse along two axes: (1) Dimensions: a model can appear diverse on one axis yet structurally degenerate on another, and (2) Domains: the same model may collapse the most in personality yet be the most diverse in moral reasoning. Furthermore, item-level diagnostics reveal that behavioral variation tracks coarse demographic stereotypes rather than the fine-grained individual differences specified in each persona. Counter-intuitively, \textbf{the models achieving the highest per-persona fidelity consistently produce the most stereotyped populations}. We release our toolkit and data to support population-level evaluation of LLMs. --- ### Can Current Agents Close the Discovery-to-Application Gap? A Case Study in Minecraft Authors: Zhou Ziheng, Huacong Tang, Jinyuan Zhang, Haowei Lin, Bangcheng Yang, Qian Long, Fang Sun, Yizhou Sun, Yitao Liang, Ying Nian Wu, Demetri Terzopoulos, Xiaofeng Gao Published: 2026-04-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.24697v2 Discovering causal regularities and applying them to build functional systems--the discovery-to-application loop--is a hallmark of general intelligence, yet evaluating this capacity has been hindered by the vast complexity gap between scientific discovery and real-world engineering. We introduce SciCrafter, a Minecraft-based benchmark that operationalizes this loop through parameterized redstone circuit tasks. Agents must ignite lamps in specified patterns (e.g., simultaneously or in timed sequences); scaling target parameters substantially increases construction complexity and required knowledge, forcing genuine discovery rather than reliance on memorized solutions. Evaluating frontier models including GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5 under a general-purpose code agent scaffold, we find that all plateau at approximately 26% success rate. To diagnose these failures, we decompose the loop into four capacities--knowledge gap identification, experimental discovery, knowledge consolidation, and knowledge application--and design targeted interventions whose marginal contributions serve as proxies for corresponding gaps. Our analysis reveals that although the general knowledge application capability still remains as the biggest gap across all models, for frontier models the knowledge gap identification starts to become a major hurdle--indicating the bottleneck is shifting from solving problems right to raising the right problems for current AI. We release SciCrafter as a diagnostic probe for future research on AI systems that navigate the full discovery-to-application loop. --- ### NeuroClaw Technical Report Authors: Cheng Wang, Zhibin He, Zhihao Peng, Shengyuan Liu, Yufan Hu, Carl Yang, Lifang He, Lichao Sun, Xiang Li, Yixuan Yuan Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24696v3 Agentic artificial intelligence systems promise to accelerate scientific workflows, but neuroimaging poses unique challenges: heterogeneous modalities (sMRI, fMRI, dMRI, EEG), long multi-stage pipelines, and persistent reproducibility risks. To address this gap, we present NeuroClaw, a domain-specialized multi-agent research assistant for executable and reproducible neuroimaging research. NeuroClaw operates directly on raw neuroimaging data across formats and modalities, grounding decisions in dataset semantics and BIDS metadata so users need not prepare curated inputs or bespoke model code. The platform combines harness engineering with end-to-end environment management, including pinned Python environments, Docker support, automated installers for common neuroimaging tools, and GPU configuration. In practice, this layer emphasizes checkpointing, post-execution verification, structured audit traces, and controlled runtime setup, making toolchains more transparent while improving reproducibility and auditability. A three-tier skill/agent hierarchy separates user-facing interaction, high-level orchestration, and low-level tool skills to decompose complex workflows into safe, reusable units. Alongside the NeuroClaw framework, we introduce NeuroBench, a system-level benchmark for executability, artifact validity, and reproducibility readiness. Across multiple multimodal LLMs, NeuroClaw-enabled runs yield consistent and substantial score improvements compared with direct agent invocation. Project homepage: https://cuhk-aim-group.github.io/NeuroClaw/index.html --- ### Contextual Linear Activation Steering of Language Models Authors: Brandon Hsu, Daniel Beaglehole, Adityanarayanan Radhakrishnan, Mikhail Belkin Published: 2026-04-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.24693v1 Linear activation steering is a powerful approach for eliciting the capabilities of large language models and specializing their behavior using limited labeled data. While effective, existing methods often apply a fixed steering strength to all tokens, resulting in inconsistent steering quality across diverse input prompts. In this work, we introduce Contextual Linear Activation Steering (CLAS), a method that dynamically adapts linear activation steering to context-dependent steering strengths. Across eleven steering benchmarks and four model families, it consistently outperforms standard linear activation steering and matches or exceeds the performance of ReFT and LoRA in settings with limited labeled data. We therefore propose CLAS as a scalable, interpretable, and accurate method for specializing and steering large language models. --- ### Diffusion-Guided Feature Selection via Nishimori Temperature: Noise-Based Spectral Embedding Authors: Vasiliy S. Usatyuk, Denis A. Sapozhnikov, Sergey I. Egorov Published: 2026-04-27 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.24692v1 We propose Noise-Based Spectral Embedding (NBSE), a physics-informed framework for selecting informative features from high-dimensional data without greedy search. NBSE constructs a sparse similarity graph on the samples and identifies the Nishimori temperature $β_N$ the critical inverse temperature at which the Bethe Hessian becomes singular. The corresponding smallest eigenvector captures the dominant mode of an intrinsically degree-corrected diffusion process, naturally reweighting nodes to prevent hub dominance. By transposing the data matrix and applying NBSE in feature space, we obtain a one-dimensional spectral embedding that reveals groups of redundant or semantically related dimensions; balanced binning then selects one representative per group. We prove that coloured Gaussian perturbations shift $β_N$ by at most $O(\barσ^2)$, guaranteeing robustness to measurement noise. Experiments on ImageNet embeddings from MobileNetV2 and EfficientNet-B4 show that NBSE preserves classification accuracy even under aggressive compression: on EfficientNet-B4 the accuracy drop is below $1\%$ when retaining only $30\%$ of features, outperforming ANOVA $F$-test and random selection by up to $6.8\%$. --- ### Can LLMs Act as Historians? Evaluating Historical Research Capabilities of LLMs via the Chinese Imperial Examination Authors: Lirong Gao, Zeqing Wang, Yuyan Cai, Jiayi Deng, Yanmei Gu, Yiming Zhang, Jia Zhou, Yanfei Zhang, Junbo Zhao Published: 2026-04-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.24690v1 While Large Language Models (LLMs) have increasingly assisted in historical tasks such as text processing, their capacity for professional-level historical reasoning remains underexplored. Existing benchmarks primarily assess basic knowledge breadth or lexical understanding, failing to capture the higher-order skills, such as evidentiary reasoning,that are central to historical research. To fill this gap, we introduce ProHist-Bench, a novel benchmark anchored in the Chinese Imperial Examination (Keju) system, a comprehensive microcosm of East Asian political, social, and intellectual history spanning over 1,300 years. Developed through deep interdisciplinary collaboration, ProHist-Bench features 400 challenging, expert-curated questions across eight dynasties, accompanied by 10,891 fine-grained evaluation rubrics. Through a rigorous evaluation of 18 LLMs, we reveal a significant proficiency gap: even state-of-the-art LLMs struggle with complex historical research questions. We hope ProHist-Bench will facilitate the development of domain-specific reasoning LLMs, advance computational historical research, and further uncover the untapped potential of LLMs. We release ProHist-Bench at https://github.com/inclusionAI/ABench/tree/main/ProHist-Bench. --- ### Governing What You Cannot Observe: Adaptive Runtime Governance for Autonomous AI Agents Authors: German Marin, Jatin Chaudhary Published: 2026-04-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.24686v1 Autonomous AI agents can remain fully authorized and still become unsafe as behavior drifts, adversaries adapt, and decision patterns shift without any code change. We propose the \textbf{Informational Viability Principle}: governing an agent reduces to estimating a bound on unobserved risk $\hat{B}(x) = U(x) + SB(x) + RG(x)$ and allowing an action only when its capacity $S(x)$ exceeds $\hat{B}(x)$ by a safety margin. The \textbf{Agent Viability Framework}, grounded in Aubin's viability theory, establishes three properties -- monitoring (P1), anticipation (P2), and monotonic restriction (P3) -- as individually necessary and collectively sufficient for documented failure modes. \textbf{RiskGate} instantiates the framework with dedicated statistical estimators (KL divergence, segment-vs-rest $z$-tests, sequential pattern matching), a fail-secure monotonic pipeline, and a closed-loop Autopilot formalised as an instance of Aubin's regulation map with kill-switch-as-last-resort; a scalar Viability Index $VI(t) \in (-1,+1)$ with first-order $t^*$ prediction transforms governance from reactive to predictive. Contributions are the theoretical framework, the reference implementation, and analytical coverage against published agent-failure taxonomies; quantitative empirical evaluation is scoped as follow-up work. --- ### Aycromo: An Open-Source Platform for Automatic Chromosome Detection in Metaphase Images Based on Deep Learning Authors: Jorge L. A. Lima, Filipe R. Cordeiro Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24685v1 Chromosome analysis is a fundamental step in the diagnosis of genetic diseases, but the manual karyotyping workflow is time-consuming and heavily dependent on expert specialists, often requiring several days per patient. Although Deep Learning models have achieved high performance in chromosome detection, most proposed solutions remain restricted to research prototypes or lack graphical interfaces suitable for clinical use. In this work, we present Aycromo, an open-source desktop platform for AI-assisted cytogenetic analysis. Built on Electron and ONNX Runtime, the tool allows cytogeneticists to load pre-trained models, compare architectures through an integrated benchmarking module, and manually correct detections via an interactive annotation interface, all without command-line interaction. Preliminary experiments on metaphase images from the CRCN-NE dataset demonstrate that YOLOv11 achieves 99.40% mAP@50, while the platform reduces per-slide analysis to seconds --- ### Benchmarking Pathology Foundation Models for Breast Cancer Survival Prediction Authors: Fredrik K. Gustafsson, Constance Boissin, Johan Vallon-Christersson, David A. Clifton, Mattias Rantalainen Published: 2026-04-27 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.24679v1 Pathology foundation models (PFMs) have recently emerged as powerful pretrained encoders for computational pathology, enabling transfer learning across a wide range of downstream tasks. However, systematic comparisons of these models for clinically meaningful prediction problems remain limited, especially in the context of survival prediction under external validation. In this study, we benchmark widely used and recently proposed PFMs for breast cancer survival prediction from whole-slide histopathology images. Using a standardized pipeline based on patch-level feature extraction and a unified survival modeling framework, we evaluate model representations across three independent clinical cohorts comprising more than 5,400 patients with long-term follow-up. Models are trained on one cohort and evaluated on two independent external cohorts, enabling a rigorous assessment of cross-dataset generalization. Overall, H-optimus-1 achieves the strongest survival prediction performance. More broadly, we observe consistent generational improvements across model families, with second-generation PFMs outperforming their first-generation counterparts. However, absolute performance differences between many recent PFMs remain modest, suggesting diminishing returns from further scaling of pretraining data or model size alone. Notably, the compact distilled model H0-mini slightly outperforms its larger teacher model H-optimus-0, despite using fewer than 8% of the parameters and enabling significantly faster feature extraction. Together, these results provide the first large-scale, externally validated benchmark of PFMs for breast cancer survival prediction, and offer practical guidance for efficient deployment of PFMs in clinical workflows. --- ### Leveraging LLMs for Multi-File DSL Code Generation: An Industrial Case Study Authors: Sivajeet Chand, Kevin Nguyen, Peter Kuntz, Alexander Pretschner Published: 2026-04-27 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2604.24678v1 Large language models (LLMs) perform strongly on general-purpose code generation, yet their applicability to enterprise domain-specific languages (DSLs) remains underexplored, especially for repository-scale change generation spanning multiple files and folder structures from a single natural-language (NL) instruction. We report an industrial case study at BMW that adapts code-oriented LLMs to generate and modify project-root DSL artifacts for an Xtext-based DSL that drives downstream Java/TypeScript code generation. We develop an end-to-end pipeline for dataset construction, multi-file task representation, model adaptation, and evaluation. We encode DSL folder hierarchies as structured, path-preserving JSON, allowing single-response generation at repository scale and learning cross-file dependencies. We evaluate two instruction-tuned code LLMs (Qwen2.5-Coder and DeepSeek-Coder, 7B) under three configurations: baseline prompting, one-shot in-context learning, and parameter-efficient fine-tuning (QLoRA). Beyond standard similarity metrics, we introduce task-specific measures that assess edit correctness and repository structural fidelity. Fine-tuning yields the most significant gains across models and metrics, achieving high exact-match accuracy, substantial edit similarity, and structural fidelity of 1.00 on our held-out set for multi-file outputs. At the same time, one-shot in-context learning provides smaller but consistent improvements over baseline prompting. We further validate practical utility via an expert developer survey and an execution-based check using the existing code generator. --- ### A Functorial Formulation of Neighborhood Aggregating Deep Learning Authors: Sun Woo Park, Yun Young Choi, U Jin Choi, Youngho Woo Published: 2026-04-27 Categories: cs.LG, math.AT arXiv: https://arxiv.org/abs/2604.24672v1 We provide a mathematical interpretation of convolutional (or message passing) neural networks by using presheaves and copresheaves of the set of continuous functions over a topological space. Based on this interpretation, we formulate a theoretical heuristic which elaborates a number of empirical limitations of these neural networks by using obstructions on such sets of continuous functions over a topological space to be sheaves or copresheaves. --- ### The Price of Agreement: Measuring LLM Sycophancy in Agentic Financial Applications Authors: Zhenyu Zhao, Aparna Balagopalan, Adi Agrawal, Dilshoda Yergasheva, Waseem Alshikh, Daniel M. Bikel Published: 2026-04-27 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.24668v3 Given the increased use of LLMs in financial systems today, it becomes important to evaluate the safety and robustness of such systems. One failure mode that LLMs frequently display in general domain settings is that of sycophancy. That is, models prioritize agreement with expressed user beliefs over correctness, leading to decreased accuracy and trust. In this work, we focus on evaluating sycophancy that LLMs display in agentic financial tasks. Our findings are three-fold: first, we find the models show only low to modest drops in performance in the face of user rebuttals or contradictions to the reference answer, which distinguishes sycophancy that models display in financial agentic settings from findings in prior work. Second, we introduce a suite of tasks to test for sycophancy by user preference information that contradicts the reference answer and find that most models fail in the presence of such inputs. Lastly, we benchmark different modes of recovery such as input filtering with a pretrained LLM. --- ### Benchmarking Source-Sensitive Reasoning in Turkish: Humans and LLMs under Evidential Trust Manipulation Authors: Sercan Karakaş, Yusuf Şimşek Published: 2026-04-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.24665v1 This paper investigates whether source trustworthiness shapes Turkish evidential morphology and whether large language models (LLMs) track this sensitivity. We study the past-domain contrast between -DI and -mIs in controlled cloze contexts where the information source is overtly external, while only its perceived reliability is manipulated (High-Trust vs. Low-Trust). In a human production experiment, native speakers of Turkish show a robust trust effect: High-Trust contexts yield relatively more -DI, whereas Low-Trust contexts yield relatively more -mIs, with the pattern remaining stable across sensitivity analyses. We then evaluate 10 LLMs in three prompting paradigms (open gap-fill, explicit past-tense gap-fill, and forced-choice A/B selection). LLM behavior is highly model- and prompt-dependent: some models show weak or local trust-consistent shifts, but effects are generally unstable, often reversed, and frequently overshadowed by output-compliance problems and strong base-rate suffix preferences. The results provide new evidence for a trust-/commitment-based account of Turkish evidentiality and reveal a clear human-LLM gap in source-sensitive evidential reasoning. --- ### Dual Control of Linear Systems from Bilinear Observations with Belief Space Model Predictive Control Authors: Daniel Cao, Beixi Du, Andrew Lowitt, Sunmook Choi, Sarah Dean, Yahya Sattar Published: 2026-04-27 Categories: math.OC, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.24663v1 We study finite-horizon quadratic control of linear systems with bilinear observations, in which the control input affects not only the state dynamics but also the partial observations of the state. In this setting, the separation principle can fail because control inputs influence the future quality of state estimates. State estimation requires an input-dependent Kalman filter whose gain and error covariance evolve as functions of the control inputs. To address this challenge, we propose a belief-space model predictive control ($\texttt{B-MPC}$) method that plans directly over both the estimated state and its error covariance. In particular, $\texttt{B-MPC}$ plans with a deterministic surrogate of the belief evolution defined by the input-dependent Kalman filter. Through numerical experiments in two synthetic settings, we show that $\texttt{B-MPC}$ can outperform both the separation-principle controller and its MPC variant in favorable regimes, and that these gains are accompanied by lower estimation covariance and more uncertainty-aware action choices. --- ### Information bottleneck for learning the phase space of dynamics from high-dimensional experimental data Authors: K. Michael Martini, Eslam Abdelaleem, Paarth Gulati, Ilya Nemenman Published: 2026-04-27 Categories: physics.data-an, cs.AI, cs.IT arXiv: https://arxiv.org/abs/2604.24662v2 Identifying the dynamical state variables of a system from high-dimensional observations is a central problem across physical sciences. The challenge is that the state variables are not directly observable and must be inferred from raw high-dimensional data without supervision. Here we introduce DySIB (Dynamical Symmetric Information Bottleneck) as a method to learn low-dimensional representations of time-series data by maximizing predictive mutual information between past and future observation windows while penalizing representation complexity. This objective operates entirely in latent space and avoids reconstruction of the observations. We apply DySIB to an experimental video dataset of a physical pendulum, where the underlying state space is known. The method, with hyperparameters of the learning architecture set self-consistently by the data, recovers a two-dimensional representation that matches the dimensionality, topology, and geometry of the pendulum phase space, with the learned coordinates aligning smoothly with the canonical angle and angular velocity. These results demonstrate, on a well-characterized experimental system, that predictive information in latent space can be used to recover interpretable dynamical coordinates directly from high-dimensional data. --- ### Nonparametric Instrumental Variable Analysis Without Structural Equations: Debiased Inference on Functionals of Inverse Problems with No Solutions Authors: Zikai Shen, Nathan Kallus, Dimitri Meunier, Houssam Zenati, Arthur Gretton, Aurélien Bibaut Published: 2026-04-27 Categories: stat.ML, econ.EM, math.ST, stat.ME arXiv: https://arxiv.org/abs/2604.24660v2 We consider debiased inference on finite-dimensional functionals of infinite-dimensional least-squares solutions to inverse problems as a way to avoid having to assume exact solutions exist. Such assumptions are substantive and not innocuous, and their failure may imperil inference when we impose them on the statistical model. Our approach instead allows us to conduct inference on a quantity that is defined regardless of solutions existing and coincides with the usual estimands when they do. For the case of instrumental variables, this means we can motivate the analysis with structural models but these do not need to hold exactly for the semiparametric inferential procedure to remain valid. --- ### The Last Human-Written Paper: Agent-Native Research Artifacts Authors: Jiachen Liu, Jiaxin Pei, Jintao Huang, Chenglei Si, Ao Qu, Xiangru Tang, Runyu Lu, Lichang Chen, Xiaoyan Bai, Haizhong Zheng, Carl Chen, Zhiyang Chen et al. Published: 2026-04-27 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.24658v3 Scientific publication compresses a branching, iterative research process into a linear narrative, discarding the majority of what was discovered along the way. This compilation imposes two structural costs: a Storytelling Tax, where failed experiments, rejected hypotheses, and the branching exploration process are discarded to fit a linear narrative; and an Engineering Tax, where the gap between reviewer-sufficient prose and agent-sufficient specification leaves critical implementation details unwritten. Tolerable for human readers, these costs become critical when AI agents must understand, reproduce, and extend published work. We introduce the Agent-Native Research Artifact (ARA), a protocol that replaces the narrative paper with a machine-executable research package structured around four layers: scientific logic, executable code with full specifications, an exploration graph that preserves the failures compilation discards, and evidence grounding every claim in raw outputs. Three mechanisms support the ecosystem: a Live Research Manager that captures decisions and dead ends during ordinary development; an ARA Compiler that translates legacy PDFs and repos into ARAs; and an ARA-native review system that automates objective checks so human reviewers can focus on significance, novelty, and taste. On PaperBench and RE-Bench, ARA raises question-answering accuracy from 72.4% to 93.7% and reproduction success from 57.4% to 64.4%. On RE-Bench's five open-ended extension tasks, preserved failure traces in ARA accelerate progress, but can also constrain a capable agent from stepping outside the prior-run box depending on the agent's capabilities. Our code is open-sourced at https://github.com/Orchestra-Research/Agent-Native-Research-Artifact. --- ### AgentWard: A Lifecycle Security Architecture for Autonomous AI Agents Authors: Yixiang Zhang, Xinhao Deng, Jiaqing Wu, Yue Xiao, Ke Xu, Qi Li Published: 2026-04-27 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2604.24657v1 Autonomous AI agents extend large language models into full runtime systems that load skills, ingest external content, maintain memory, plan multi-step actions, and invoke privileged tools. In such systems, security failures rarely remain confined to a single interface; instead, they can propagate across initialization, input processing, memory, decision-making, and execution, often becoming apparent only when harmful effects materialize in the environment. This paper presents AgentWard, a lifecycle-oriented, defense-in-depth architecture that systematically organizes protection across these five stages. AgentWard integrates stage-specific, heterogeneous controls with cross-layer coordination, enabling threats to be intercepted along their propagation paths while safeguarding critical assets. We detail the design rationale and architecture of five coordinated protection layers, and implement a plugin-native prototype on OpenClaw to demonstrate practical feasibility. This perspective provides a concrete blueprint for structuring runtime security controls, managing trust propagation, and enforcing execution containment in autonomous AI agents. Our code is available at https://github.com/FIND-Lab/AgentWard . --- ### DepthKV: Layer-Dependent KV Cache Pruning for Long-Context LLM Inference Authors: Zahra Dehghanighobadi, Asja Fischer Published: 2026-04-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.24647v1 Long-context reasoning is a critical capability of large language models (LLMs), enabling applications such as long-document understanding, summarization, and code generation. However, efficient autoregressive inference relies on the key-value (KV) cache, whose memory footprint grows linearly with sequence length, leading to a major memory bottleneck. To mitigate this overhead, KV cache pruning methods discard cached tokens with low attention scores during inference. Most existing methods apply a uniform pruning ratio across layers, implicitly assuming that all layers contribute equally to overall model performance. We show that this assumption is suboptimal, as layers differ significantly in their sensitivity to pruning. We propose DepthKV, a layer-dependent pruning framework that allocates a fixed global KV budget across layers based on their sensitivity, rather than using a uniform allocation. Across multiple models and tasks, DepthKV consistently outperforms uniform pruning at the same global pruning ratio, demonstrating more effective utilization of the KV cache budget through layer-dependent allocation. --- ### K-MetBench: A Multi-Dimensional Benchmark for Fine-Grained Evaluation of Expert Reasoning, Locality, and Multimodality in Meteorology Authors: Soyeon Kim, Cheongwoong Kang, Myeongjin Lee, Eun-Chul Chang, Jaedeok Lee, Jaesik Choi Published: 2026-04-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.24645v1 The development of practical (multimodal) large language model assistants for Korean weather forecasters is hindered by the absence of a multidimensional, expert-level evaluation framework grounded in authoritative sources. To address this, we introduce K-MetBench, a diagnostic benchmark grounded in national qualification exams. It exposes critical gaps across four dimensions: expert visual reasoning of charts, logical validity via expert-verified rationales, Korean-specific geo-cultural comprehension, and fine-grained domain analysis. Our evaluation of 55 models reveals a profound modality gap in interpreting specialized diagrams and a reasoning gap where models hallucinate logic despite correct predictions. Crucially, Korean models outperform significantly larger global models in local contexts, demonstrating that parameter scaling alone cannot resolve cultural dependencies. K-MetBench serves as a roadmap for developing reliable, culturally aware expert AI agents. The dataset is available at https://huggingface.co/datasets/soyeonbot/K-MetBench . --- ### Probing CLIP's Comprehension of 360-Degree Textual and Visual Semantics Authors: Hai Wang, Xiaochen Yang, Mingzhi Dong, Jing-Hao Xue Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24642v1 The dream of instantly creating rich 360-degree panoramic worlds from text is rapidly becoming a reality, yet a crucial gap exists in our ability to reliably evaluate their semantic alignment. Contrastive Language-Image Pre-training (CLIP) models, standard AI evaluators, predominantly trained on perspective image-text pairs, face an open question regarding their understanding of the unique characteristics of 360-degree panoramic image-text pairs. This paper addresses this gap by first introducing two concepts: \emph{360-degree textual semantics}, semantic information conveyed by explicit format identifiers, and \emph{360-degree visual semantics}, invariant semantics under horizontal circular shifts. To probe CLIP's comprehension of these semantics, we then propose novel evaluation methodologies using keyword manipulation and horizontal circular shifts of varying magnitudes. Rigorous statistical analyses across popular CLIP configurations reveal that: (1) CLIP models effectively leverage explicit textual identifiers, demonstrating an understanding of 360-degree textual semantics; and (2) CLIP models fail to robustly preserve semantic alignment under horizontal circular shifts, indicating limited comprehension of 360-degree visual semantics. To address this limitation, we propose a LoRA-based fine-tuning framework that explicitly instills invariance to circular shifts. Our fine-tuned models exhibit improved comprehension of 360-degree visual semantics, though with a slight degradation in original semantic evaluation performance, highlighting a fundamental trade-off in adapting CLIP to 360-degree panoramic images. Code is available at https://github.com/littlewhitesea/360Semantics. --- ### Cortex-Inspired Continual Learning: Unsupervised Instantiation and Recovery of Functional Task Networks Authors: Kevin McKee, Thomas Hazy, Yicong Zheng, Zacharie Bugaud, Thomas Miconi Published: 2026-04-27 Categories: cs.LG, cs.AI, q-bio.NC arXiv: https://arxiv.org/abs/2604.24637v2 Block-sequential continual learning demands that a single model both protect prior solutions from catastrophic forgetting and efficiently infer at inference time which prior solution matches the current input without task labels. We present Functional Task Networks (FTN), a parameter-isolation method inspired by structural and dynamical motifs found in the mammalian neocortex. Similar to mixture-of-experts, this method uses a high dimensional, self-organizing binary mask over a large population of small but deep networks, inspired by dendritic models of pyramidal neurons. The mask is produced by a three-stage procedure: (1) gradient descent on a continuous mask identifies task-relevant neurons, (2) a smoothing kernel biases the result toward spatial contiguity, (3) and k-winner-take-all binarizes the resulting group at a fixed capacity budget. Like mixture-of-experts, each neuron is an independent deep network, so disjoint masks give exactly disjoint gradient updates, providing structural guarantees against catastrophic forgetting. This three-stage procedure recovers the sub-network of a previously-trained task in a single gradient step, providing unsupervised task segmentation at inference time. We test it on three continual-learning benchmarks: (1) a synthetic multi-task classification/regression generator, (2) MNIST with shuffled class labels (pure concept shift), and (3) Permuted MNIST (domain shift). On all three, FTN with fine grained smoothing (FTN-Slow) results in nearly zero forgetting. FTN with a large kernel and only 2 iterations of smoothing (FTN-Fast) trades off some retention for increased speed. We show that the spatial organization mechanism reduces the effective mask search from the combinatorial top-k subset problem in O(C(H,K)) to the complexity of a near-linear scan in O(H) over compact cortical neighborhoods, which is parallelized by the gradient-based update. --- ### Less Is More: Engineering Challenges of On-Device Small Language Model Integration in a Mobile Application Authors: William Oliveira Published: 2026-04-27 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.24636v2 On-device Small Language Models (SLMs) promise fully offline, private AI experiences for mobile users (no cloud dependency, no data leaving the device). But is this promise achievable in practice? This paper presents a longitudinal practitioner case study documenting the engineering challenges of integrating SLMs (Gemma 4 E2B, 2.6B parameters; Qwen3 0.6B, 600M parameters) into Palabrita, a production Android word-guessing game. Over a 5-day development sprint comprising 204 commits (~90 directly AI-related), the system underwent a radical transformation: from an ambitious design where the LLM generated complete structured puzzles (word, category, difficulty, and five hints as JSON) to a pragmatic architecture where curated word lists provide the words and the LLM generates only three short hints, with a deterministic fallback if it fails. We identify five categories of failures specific to on-device SLM integration: output format violations, constraint violations, context quality degradation, latency incompatibility, and model selection instability. For each failure category, we document the observed symptoms, root causes, and the prompt engineering and architectural strategies that effectively mitigated them, including multi-layer defensive parsing, contextual retry with failure feedback, session rotation, progressive prompt hardening, and systematic responsibility reduction. Our findings demonstrate that on-device SLMs are viable for production mobile applications, but only when the developer accepts a fundamental constraint: the most reliable on-device LLM feature is one where the LLM does the least. We distill our experience into eight actionable design heuristics for practitioners integrating SLMs into mobile apps. --- ### Computational Design and Experimental Validation of Photoactive PARP1 Inhibitors Authors: Simon Axelrod, Miroslav Kašpar, Kristýna Jelínková, Markéta Šmídková, Erika Bartůňková, Sille Štěpánová, Eugene Shakhnovich, Václav Kašička, Martin Dračínský, Zlatko Janeba, Rafael Gómez-Bombarelli Published: 2026-04-27 Categories: physics.chem-ph, cs.LG arXiv: https://arxiv.org/abs/2604.24634v1 Light-activated drugs are a promising way to treat localized diseases for which existing treatments have severe side effects. However, their development is complicated by the set of photophysical and biological properties that must be simultaneously optimized. Here we used computational techniques to find a set of promising candidates for the photoactive inhibition of the poly(ADP-ribose) polymerase 1 (PARP1) cancer target. Using our recently developed methods based on atomistic simulation and machine learning (ML), we screened a set of 5 million hypothetical photoactive ligands. Our workflow used protein-ligand docking to identify candidates with differential PARP1 binding under light and dark conditions; ML force fields and quantum chemistry calculations to predict p$K_\mathrm{a}$, absorption spectra, and thermal half-lives; graph-based surrogate models to screen additional compounds; excited-state nonadiabatic dynamics with ML force fields to estimate quantum yields; and free energy perturbation (FEP) to refine binding predictions. From these predictions, we prioritized a small set of synthetically feasible candidates expected to have red-shifted absorption spectra, thermal half-lives on the order of seconds to minutes, and isomer-dependent PARP1 binding under visible-light control. We synthesized 10 candidates and experimentally characterized their photobehavior and PARP1 inhibition constants. Among the validated compounds, \textbf{1} showed a 15-fold increase in inhibition of PARP1 upon green-light irradiation at 519 nm (208.8 $\pm$ 28.3 $μ$M vs 14.4 $\pm$ 1.9 $μ$M). These results validate the computation-guided screening strategy for identifying red-shifted PARP1 photoinhibitors, while also underscoring current limitations such as rapid thermal relaxation in aqueous media. --- ### Meta-CoT: Enhancing Granularity and Generalization in Image Editing Authors: Shiyi Zhang, Yiji Cheng, Tiankai Hang, Zijin Yin, Runze He, Yu Xu, Wenxun Dai, Yunlong Lin, Chunyu Wang, Qinglin Lu, Yansong Tang Published: 2026-04-27 Categories: cs.CV, cs.AI, cs.LG, cs.MM arXiv: https://arxiv.org/abs/2604.24625v1 Unified multi-modal understanding/generative models have shown improved image editing performance by incorporating fine-grained understanding into their Chain-of-Thought (CoT) process. However, a critical question remains underexplored: what forms of CoT and training strategy can jointly enhance both the understanding granularity and generalization? To address this, we propose Meta-CoT, a paradigm that performs a two-level decomposition of any single-image editing operation with two key properties: (1) Decomposability. We observe that any editing intention can be represented as a triplet - (task, target, required understanding ability). Inspired by this, Meta-CoT decomposes both the editing task and the target, generating task-specific CoT and traversing editing operations on all targets. This decomposition enhances the model's understanding granularity of editing operations and guides it to learn each element of the triplet during training, substantially improving the editing capability. (2) Generalizability. In the second decomposition level, we further break down editing tasks into five fundamental meta-tasks. We find that training on these five meta-tasks, together with the other two elements of the triplet, is sufficient to achieve strong generalization across diverse, unseen editing tasks. To further align the model's editing behavior with its CoT reasoning, we introduce the CoT-Editing Consistency Reward, which encourages more accurate and effective utilization of CoT information during editing. Experiments demonstrate that our method achieves an overall 15.8% improvement across 21 editing tasks, and generalizes effectively to unseen editing tasks when trained on only a small set of meta-tasks. Our code, benchmark, and model are released at https://shiyi-zh0408.github.io/projectpages/Meta-CoT/ --- ### XGRAG: A Graph-Native Framework for Explaining KG-based Retrieval-Augmented Generation Authors: Zhuoling Li, Ha Linh Hong Tran Nguyen, Valeria Bladinieres, Maxim Romanovsky Published: 2026-04-27 Categories: cs.AI, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2604.24623v1 Graph-based Retrieval-Augmented Generation (GraphRAG) extends traditional RAG by using knowledge graphs (KGs) to give large language models (LLMs) a structured, semantically coherent context, yielding more grounded answers. However, GraphRAG reasoning process remains a black-box, limiting our ability to understand how specific pieces of structured knowledge influence the final output. Existing explainability (XAI) methods for RAG systems, designed for text-based retrieval, are limited to interpreting an LLM response through the relational structures among knowledge components, creating a critical gap in transparency and trustworthiness. To address this, we introduce XGRAG, a novel framework that generates causally grounded explanations for GraphRAG systems by employing graph-based perturbation strategies, to quantify the contribution of individual graph components on the model answer. We conduct extensive experiments comparing XGRAG against RAG-Ex, an XAI baseline for standard RAG, and evaluate its robustness across various question types, narrative structures and LLMs. Our results demonstrate a 14.81% improvement in explanation quality over the baseline RAG-Ex across NarrativeQA, FairyTaleQA, and TriviaQA, evaluated by F1-score measuring alignment between generated explanations and original answers. Furthermore, XGRAG explanations exhibit a strong correlation with graph centrality measures, validating its ability to capture graph structure. XGRAG provides a scalable and generalizable approach towards trustworthy AI through transparent, graph-based explanations that enhance the interpretability of RAG systems. --- ### CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies Authors: Fan Du, Feng Yan, Jianxiong Wu, Xinrun Xu, Weiye Zhang, Weinong Wang, Yu Guo, Bin Qian, Zhihai He, Fei Wang, Heng Yang Published: 2026-04-27 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.24622v4 Flow-based vision-language-action (VLA) policies offer strong expressivity for action generation, but suffer from a fundamental inefficiency: multi-step inference is required to recover action structure from uninformative Gaussian noise, leading to a poor efficiency-quality trade-off under real-time constraints. We address this issue by rethinking the role of the starting point in generative action modeling. Instead of shortening the sampling trajectory, we propose CF-VLA, a coarse-to-fine two-stage formulation that restructures action generation into a coarse initialization step that constructs an action-aware starting point, followed by a single-step local refinement that corrects residual errors. Concretely, the coarse stage learns a conditional posterior over endpoint velocity to transform Gaussian noise into a structured initialization, while the fine stage performs a fixed-time refinement from this initialization. To stabilize training, we introduce a stepwise strategy that first learns a controlled coarse predictor and then performs joint optimization. Experiments on CALVIN and LIBERO show that our method establishes a strong efficiency-performance frontier under low-NFE (Number of Function Evaluations) regimes: it consistently outperforms existing NFE=2 methods, matches or surpasses the NFE=10 $π_{0.5}$ baseline on several metrics, reduces action sampling latency by 75.4%, and achieves the best average real-robot success rate of 83.0%, outperforming MIP by 19.5 points and $π_{0.5}$ by 4.0 points. These results suggest that structured, coarse-to-fine generation enables both strong performance and efficient inference. Our code is available at https://github.com/EmbodiedAI-RoboTron/CF-VLA. --- ### CF-VLA: Efficient Coarse-to-Fine Action Generation for Vision-Language-Action Policies Authors: Fan Du, Feng Yan, Jianxiong Wu, Xinrun Xu, Weiye Zhang, Weinong Wang, Yu Guo, Bin Qian, Zhihai He, Fei Wang, Heng Yang Published: 2026-04-27 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.24622v2 Flow-based vision-language-action (VLA) policies offer strong expressivity for action generation, but suffer from a fundamental inefficiency: multi-step inference is required to recover action structure from uninformative Gaussian noise, leading to a poor efficiency-quality trade-off under real-time constraints. We address this issue by rethinking the role of the starting point in generative action modeling. Instead of shortening the sampling trajectory, we propose CF-VLA, a coarse-to-fine two-stage formulation that restructures action generation into a coarse initialization step that constructs an action-aware starting point, followed by a single-step local refinement that corrects residual errors. Concretely, the coarse stage learns a conditional posterior over endpoint velocity to transform Gaussian noise into a structured initialization, while the fine stage performs a fixed-time refinement from this initialization. To stabilize training, we introduce a stepwise strategy that first learns a controlled coarse predictor and then performs joint optimization. Experiments on CALVIN and LIBERO show that our method establishes a strong efficiency-performance frontier under low-NFE (Number of Function Evaluations) regimes: it consistently outperforms existing NFE=2 methods, matches or surpasses the NFE=10 $π_{0.5}$ baseline on several metrics, reduces action sampling latency by 75.4%, and achieves the best average real-robot success rate of 83.0%, outperforming MIP by 19.5 points and $π_{0.5}$ by 4.0 points. These results suggest that structured, coarse-to-fine generation enables both strong performance and efficient inference. Our code is available at https://github.com/EmbodiedAI-RoboTron/CF-VLA. --- ### Looking for the Bottleneck in Fine-grained Temporal Relation Classification Authors: Hugo Sousa, Ricardo Campos, Alípio Jorge Published: 2026-04-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.24620v1 Temporal relation classification is the task of determining the temporal relation between pairs of temporal entities in a text. Despite recent advancements in natural language processing, temporal relation classification remains a considerable challenge. Early attempts framed this task using a comprehensive set of temporal relations between events and temporal expressions. However, due to the task complexity, datasets have been progressively simplified, leading recent approaches to focus on the relations between event pairs and to use only a subset of relations. In this work, we revisit the broader goal of classifying interval relations between temporal entities by considering the full set of relations that can hold between two time intervals. The proposed approach, Interval from Point, involves first classifying the point relations between the endpoints of the temporal entities and then decoding these point relations into an interval relation. Evaluation on the TempEval-3 dataset shows that this approach can yield effective results, achieving a temporal awareness score of $70.1$ percent, a new state-of-the-art on this benchmark. --- ### Evaluating whether AI models would sabotage AI safety research Authors: Robert Kirk, Alexandra Souly, Kai Fronsdal, Abby D'Cruz, Xander Davies Published: 2026-04-27 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.24618v1 We evaluate the propensity of frontier models to sabotage or refuse to assist with safety research when deployed as AI research agents within a frontier AI company. We apply two complementary evaluations to four Claude models (Mythos Preview, Opus 4.7 Preview, Opus 4.6, and Sonnet 4.6): an unprompted sabotage evaluation testing model behaviour with opportunities to sabotage safety research, and a sabotage continuation evaluation testing whether models continue to sabotage when placed in trajectories where prior actions have started undermining research. We find no instances of unprompted sabotage across any model, with refusal rates close to zero for Mythos Preview and Opus 4.7 Preview, though all models sometimes only partially completed tasks. In the continuation evaluation, Mythos Preview actively continues sabotage in 7% of cases (versus 3% for Opus 4.6, 4% for Sonnet 4.6, and 0% for Opus 4.7 Preview), and exhibits reasoning-output discrepancy in the majority of these cases, indicating covert sabotage reasoning. Our evaluation framework builds on Petri, an open-source LLM auditing tool, with a custom scaffold running models inside Claude Code, alongside an iterative pipeline for generating realistic sabotage trajectories. We measure both evaluation awareness and a new form of situational awareness termed "prefill awareness", the capability to recognise that prior trajectory content was not self-generated. Opus 4.7 Preview shows notably elevated unprompted evaluation awareness, while prefill awareness remains low across all models. Finally, we discuss limitations including evaluation awareness confounds, limited scenario coverage, and untested pathways to risk beyond safety research sabotage. --- ### Infrastructure-Guided Connectivity-Enhanced Road Crack Detection and Estimation Authors: Haosong Xiao, Yamini Ramesh, Rishabh Shukla, Swarat Sarkar, Chaozhe R. He Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24616v1 In this paper, we report the world's first infrastructure-guided communication-enhanced road crack detection pipeline that is effective and implementable on passenger vehicles. We first design a customized communication protocol to transmit the region of interest from the infrastructure to the vehicle. With proper camera image processing (e.g., dynamic cropping and frame selection), the focused images are provided to the crack detection model. Leveraging state-of-the-art crack detection model backbones and a carefully prepared dataset comprising a forward-facing view with a crack, we train the model to improve crack-detection performance. We demonstrate the full detection pipeline on an experimental vehicle platform, showcase the detection effectiveness, and project future research directions. --- ### NeSyCat: A Monad-Based Categorical Semantics of the Neurosymbolic ULLER Framework Authors: Daniel Romero Schellhorn, Till Mossakowski Published: 2026-04-27 Categories: cs.AI, cs.LO, math.CT, math.LO arXiv: https://arxiv.org/abs/2604.24612v1 ULLER (Unified Language for LEarning and Reasoning) offers a unified first-order logic (FOL) syntax, enabling its knowledge bases to be used directly across a wide range of neurosymbolic systems. The original specification endows this syntax with three pairwise independent semantics: classical, fuzzy, and probabilistic, each accompanied by dedicated semantic rules. We show that these seemingly disparate semantics are all instances of one categorical framework based on monads, the very construct that models side effects in functional programming. This enables the modular addition of new semantics and systematic translations between them. As example, we outline the addition of generalised quantification in Logic Tensor Networks (LTN) to arbitrary (also infinite) domains by extending the Giry monad to probability spaces. In particular, our approach allows a modular implementation of ULLER in Python and Haskell, of which we have published initial versions on GitHub. --- ### Uncovering Latent Patterns in Social Media Usage and Mental Health: A Clustering-Based Approach Using Unsupervised Machine Learning Authors: Md All Shahria, Sanjeda Dewan Mithila, Touhid Alam, Mohammad Sakib Mahmood, Mahfuza Khatun Published: 2026-04-27 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.24611v1 The widespread adoption of social media has heightened interest in its psychological effects, particularly on mental health indicators such as anxiety, depression, loneliness, and sleep quality, as these platforms increasingly influence social interactions and well-being. Although previous research has examined correlations between social media use and mental health, few studies have utilized unsupervised machine learning to segment users based on behavioral and psychological patterns, leaving a gap in identifying distinct risk profiles across diverse groups. This study seeks to address this by segmenting individuals according to their social media usage and psychological well-being, employing clustering to reveal hidden patterns and evaluate their mental health implications. Data from 551 participants, collected via an online survey, were preprocessed using KNN imputation for missing values, one-hot encoding for categorical variables like Gender with 5 unique values, and outlier detection via IQR and Z-score methods. K-Means clustering, optimized at 6 clusters using the Elbow Method and a Silhouette Score of 0.32, was applied, with PCA reducing 22 dimensions for visualization and a correlation heatmap highlighting relationships, such as a 0.28 correlation between social media hours and anxiety. --- ### Evaluation of Pose Estimation Systems for Sign Language Translation Authors: Catherine O'Brien, Gerard Sant, Mathias Müller, Sarah Ebling Published: 2026-04-27 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.24609v1 Many sign language translation (SLT) systems operate on pose sequences instead of raw video to reduce input dimensionality, improve portability, and partially anonymize signers. The choice of pose estimator is often treated as an implementation detail, with systems defaulting to widely available tools such as MediaPipe Holistic or OpenPose. We present a systematic comparison of pose estimators for pose-based SLT, covering widely used baselines (MediaPipe Holistic, OpenPose) and newer whole-body/high-capacity models (MMPose WholeBody, OpenPifPaf, AlphaPose, SDPose, Sapiens, SMPLest-X). We quantify downstream impact by training a controlled SLT pipeline on RWTH-PHOENIX-Weather 2014 where only the pose representation varies, evaluating with BLEU and BLEURT. To contextualize translation outcomes, we analyze temporal stability, missing hand keypoints, and robustness to occlusion using higher-resolution videos from the Signsuisse dataset. SDPose and Sapiens achieve the best translation performance (BLEU ~11.5), outperforming the common MediaPipe baseline (BLEU ~10). In occlusion cases, Sapiens is correct in all tested instances (15/15), while OpenPifPaf fails in nearly all (1/15) and also yields the weakest translation scores. Estimators that frequently leave out hand keypoints are associated with lower BLEU/BLEURT. We release code that can be used not only to reproduce our experiments, but also considerably lowers the barrier for other researchers to use alternative pose estimators. --- ### Learning to Route Queries to Heads for Attention-based Re-ranking with Large Language Models Authors: Yuxing Tian, Fengran Mo, Zhiqi Huang, Weixu Zhang, Jian-Yun Nie Published: 2026-04-27 Categories: cs.IR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.24608v1 Large Language Models (LLMs) have recently been explored as fine-grained zero-shot re-rankers by leveraging attention signals to estimate document relevance. However, existing methods either aggregate attention signals across all heads or rely on a statically selected subset identified by heuristic rules. This solution can be suboptimal because the informative heads can vary across queries or domains. Moreover, naively combining multiple heads can degrade performance due to redundancy or conflicting ranking signals. In this paper, we propose a query-dependent head selection method, RouteHead, for attention-based re-ranking with LLMs. Specifically, we learn a lightweight router that can map each query to an optimal head set, and relevance scores are computed by aggregating attention signals only from these heads. Since query-to-head optimal labels are unavailable, we first construct pseudo labels via an offline search. The router represents each head with a learnable embedding and represents each query using an embedding extracted from the hidden states of the frozen LLM. Then it is trained on the pseudo labels with a sparsity regularizer. Experiments on diverse benchmarks and multiple LLM backbones show that the proposed method consistently outperforms strong baselines. --- ### Majorization-Guided Test-Time Adaptation for Vision-Language Models under Modality-Specific Shift Authors: Lixian Chen, Mingxuan Huang, Yanhui Chen, Junyi Lin, Yang Shi Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24602v4 Vision--language models can face asymmetric visual and textual shifts at deployment. These shifts expose a multimodal failure mode in which an unreliable branch remains overconfident, dominates fusion, and causes entropy-based test-time adaptation to sharpen an incorrect prediction. We model this behavior as doubly stochastic posterior mixing and cast adaptation as constrained de-mixing. Majorization-Guided Multimodal Test-Time Adaptation (MG-MTTA) freezes both encoders and updates only a lightweight fusion module. Running-anchor consistency estimates relative branch drift, while cross-modal conflict regulates modality dominance before entropy sharpening. The analysis gives sufficient conditions for entropy reduction to preserve the clean decision and an explicit threshold at which a biased modality reverses the fused ranking. Across visual, textual, and joint shifts, MG-MTTA improves ImageNet top-1 accuracy from 57.97\% to 66.51\% under textual shift and from 21.68\% to 26.27\% under joint shift, while reducing wrong-more-confident failures. The largest gains occur under textual and joint shifts, where the two branches differ more in reliability. Project page: https://mg-mtta.github.io/. --- ### Majorization-Guided Test-Time Adaptation for Vision-Language Models under Modality-Specific Shift Authors: Lixian Chen, Yanhui Chen, Junyi Lin Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24602v2 Vision-language models transfer well in zero-shot settings, but at deployment the visual and textual branches often shift asymmetrically. Under this condition, entropy-based test-time adaptation can sharpen the fused posterior while increasing error, because an unreliable modality may still dominate fusion. We study this failure mode through a majorization view of multimodal posteriors and cast adaptation as a constrained de-mixing problem on the fused prediction. Based on this view, we propose MG-MTTA, which keeps the backbone frozen and updates only a lightweight gate or adapter. The objective combines fused-posterior entropy minimization with a reliability-aware gate prior built from anchor-based modality consistency and cross-modal conflict. Our analysis gives conditions under which entropy reduction preserves the correct ranking and a threshold that characterizes modality-dominance failure. On the ImageNet-based benchmark, MG-MTTA improves top-1 accuracy from 57.97 to 66.51 under semantics-preserving textual shift and from 21.68 to 26.27 under joint visual-textual shift, while remaining competitive in the visual-only benchmark. These results show that multimodal test-time adaptation should control modality reliability, not just prediction entropy. --- ### Quantum Kernel Advantage over Classical Collapse in Medical Foundation Model Embeddings Authors: Sebastian Cajas Ordóñez, Felipe Ocampo Osorio, Dax Enshan Koh, Rafi Al Attrach, Aldo Marzullo, Ariel Guerra-Adames, J. Alejandro Andrade, Siong Thye Goh, Chi-Yu Chen, Rahul Gorijavolu, Xue Yang, Noah Dane Hebdon et al. Published: 2026-04-27 Categories: quant-ph, cs.AI arXiv: https://arxiv.org/abs/2604.24597v1 We provide evidence of quantum kernel advantage under noiseless simulation in binary insurance classification on MIMIC-CXR chest radiographs using quantum support vector machines (QSVM) with frozen embeddings from three medical foundation models (MedSigLIP-448, RAD-DINO, ViT-patch32). We propose a two-tier fair comparison framework in which both classifiers receive identical PCA-q features. At Tier 1 (untuned QSVM vs. untuned linear SVM, C = 1 both sides), QSVM wins minority-class F1 in all 18 tested configurations (17 at p < 0.001, 1 at p < 0.01). The classical linear kernel collapses to majority-class prediction on 90-100% of seeds at every qubit count, while QSVM maintains non-trivial recall. At q = 11 (MedSigLIP-448 plateau center), QSVM achieves mean F1 = 0.343 vs. classical F1 = 0.050 (F1 gain = +0.293, p < 0.001) without hyperparameter tuning. Under Tier 2 (untuned QSVM vs. C-tuned RBF SVM), QSVM wins all seven tested configurations (mean gain +0.068, max +0.112). Eigenspectrum analysis reveals quantum kernel effective rank reaches 69.80 at q = 11, far exceeding linear kernel rank, while classical collapse remains C-invariant. A full qubit sweep reveals architecture-dependent concentration onset across models. Code: https://github.com/sebasmos/qml-medimage --- ### Skill Retrieval Augmentation for Agentic AI Authors: Weihang Su, Jianming Long, Qingyao Ai, Qiaozhi He, Yichen Tang, Changyue Wang, Yiteng Tu, Yingbo Wang, Yiqun Liu Published: 2026-04-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.24594v3 As large language models (LLMs) evolve into agentic problem solvers, they increasingly rely on external, reusable skills to handle tasks beyond their native parametric capabilities. In existing agent systems, the dominant strategy for incorporating skills is to explicitly enumerate available skills within the context window. However, this strategy fails to scale: as skill corpora expand, context budgets are consumed rapidly, and the agent becomes markedly less accurate in identifying the right skill. To this end, this paper formulates Skill Retrieval Augmentation (SRA), a new paradigm in which agents dynamically retrieve, incorporate, and apply relevant skills from large external skill corpora on demand. To make this problem measurable, we construct a large-scale skill corpus and introduce SRA-Bench, the first benchmark for decomposed evaluation of the full SRA pipeline, covering skill retrieval, skill incorporation, and end-task execution. SRA-Bench contains 5,400 capability-intensive test instances and 636 manually constructed gold skills, which are mixed with web-collected distractor skills to form a large-scale corpus of 26,262 skills. Extensive experiments show that retrieval-based skill augmentation can substantially improve agent performance, validating the promise of the paradigm. At the same time, we uncover a fundamental gap in skill incorporation: current LLM agents tend to load skills at similar rates, regardless of whether a gold skill is retrieved or whether the task actually requires external capabilities. This shows that the bottleneck in skill augmentation lies not only in retrieval but also in the base model's ability to determine which skill to load and when external loading is actually needed. These findings position SRA as a distinct research problem and establish a foundation for the scalable augmentation of capabilities in future agent systems. --- ### Fraud Detection in Cryptocurrency Markets with Spatio-Temporal Graph Neural Networks Authors: Lidia Losavio, Luca Persia, Madan Sathe, Dimosthenis Pasadakis Published: 2026-04-27 Categories: cs.LG, cs.CE arXiv: https://arxiv.org/abs/2604.24590v1 Technological advancements in cryptocurrency markets have increased accessibility for investors, but concurrently exposed them to the risks of market manipulations. Existing fraud detection mechanisms typically rely on machine learning methods that treat each financial asset (i.e., token) and its related transactions independently. However, market manipulation strategies are rarely isolated events, but are rather characterized by coordination, repetition, and frequent transfers among related assets. This suggests that relational structure constitutes an integral component of the signal and can be effectively represented through graphical means. In this paper, we propose three graph construction methods that rely on aggregated hourly market data. The proposed graphs are processed by a unified spatio-temporal Graph Neural Network (GNN) architecture that combines attention-based spatial aggregation with temporal Transformer encoding. We evaluate our methodology on a real-world dataset comprised of pump-and-dump schemes in cryptocurrency markets, spanning a period of over three years. Our comparative results showcase that our graph-based models achieve significant improvements over standard machine learning baselines in detecting anomalous events. Our work highlights that learned market connectivity provides substantial gains for detecting coordinated market manipulation schemes. --- ### A systematic evaluation of vision-language models for observational astronomical reasoning tasks Authors: Wenke Ren, Hengxiao Guo, Wenwen Zuo, Xiaoman Zhang Published: 2026-04-27 Categories: cs.AI, astro-ph.GA, astro-ph.IM arXiv: https://arxiv.org/abs/2604.24589v1 Vision-language models (VLMs) are increasingly proposed as general-purpose tools for scientific data interpretation, yet their reliability on real astronomical observations across diverse modalities remains untested. We present AstroVLBench, a comprehensive benchmark comprising over 4,100 expert-verified instances across five tasks spanning optical imaging, radio interferometry, multi-wavelength photometry, time-domain light curves, and optical spectroscopy. Evaluating six frontier models, we find that performance is strongly modality-dependent: while one model (Gemini 3 Pro) emerges as the most consistently capable across tasks, task-specific strengths vary, and all models substantially underperform domain-specialized methods. Mechanistic ablations reveal that performance depends not only on directing attention to salient visual features but also on grounding those features in physical knowledge. Phenomenological prompts describing what to look for improve accuracy by sharpening model focus, but physical prompts explaining why those features matter perform better overall and yield more balanced classifications with reduced class-specific bias. Consistent with this picture, presenting the underlying one-dimensional measurements directly as numerical tables instead of rendered plots yields up to 13 percentage points improvement. Reasoning quality analysis further demonstrates that, without explicit physical grounding, models may reach correct predictions from phenomenologically plausible cues while providing physically imprecise justifications, establishing that accuracy alone is insufficient for trustworthy scientific deployment. These findings provide the first systematic, multi-modal baselines for VLMs in observational astronomy and identify the specific representation, grounding, and reasoning bottlenecks where current models fail. --- ### Point-MF: One-step Point Cloud Generation from a Single Image via Mean Flows Authors: Yuta Baba, Keiji Yanai Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24586v1 Single-image point cloud reconstruction must infer complete 3D geometry, including occluded parts, from a single RGB image. While diffusion-based reconstructors achieve high accuracy, they typically require many denoising iterations, resulting in slow and expensive inference. We propose Point-MF, a Mean-Flow-based framework for low-NFE single-image point cloud reconstruction that couples a Mean-Flow-compatible architecture with an auxiliary loss. Specifically, Point-MF operates directly in point-cloud space to learn the mean velocity field and enables one-step reconstruction with a single network function evaluation (1-NFE), without relying on VAE-based latent representations. To make Mean Flow effective under large interval jumps, Point-MF employs a Diffusion Transformer tailored to the Mean-Flow setting, conditioned on frozen DINOv3 image features via a lightweight token adapter and equipped with explicit interval/time conditioning. Moreover, we introduce Denoised Space Anchor, a set-distance auxiliary loss on the denoised-space estimate $x_θ$ induced by the predicted velocity field, to stabilize large-step generation and reduce outliers and density artifacts. On ShapeNet-R2N2 and Pix3D, Point-MF strikes a strong balance between reconstruction quality and inference speed compared to multi-step diffusion baselines and competitive feedforward models, while generating high-quality point clouds with millisecond-level latency. --- ### Improving Vision-language Models with Perception-centric Process Reward Models Authors: Yingqian Min, Kun Zhou, Yifan Li, Yuhuan Wu, Han Peng, Yifan Du, Wayne Xin Zhao, Min Yang, Ji-Rong Wen Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24583v1 Recent advancements in reinforcement learning with verifiable rewards (RLVR) have significantly improved the complex reasoning ability of vision-language models (VLMs). However, its outcome-level supervision is too coarse to diagnose and correct errors within the reasoning chain. To this end, we propose Perceval, a process reward model (PRM) that enables token-level error grounding, which can extract image-related claims from the response and compare them one by one with the visual evidence in the image, ultimately returning claims that contain perceptual errors. Perceval is trained with perception-intensive supervised training data. We then integrate Perceval into the RL training process to train the policy models. Specifically, compared to traditional GRPO, which applies sequence-level advantages, we apply token-level advantages by targeting penalties on hallucinated spans identified by Perceval, thus enabling fine-grained supervision signals. In addition to augmenting the training process, Perceval can also assist VLMs during the inference stage. Using Perceval, we can truncate the erroneous portions of the model's response, and then either have the model regenerate the response directly or induce the model to reflect on its previous output. This process can be repeated multiple times to achieve test-time scaling. Experiments show significant improvements on benchmarks from various domains across multiple reasoning VLMs trained with RL, highlighting the promise of perception-centric supervision as a general-purpose strategy. For test-time scaling, it also demonstrates consistent performance gains over other strategies, such as major voting. Our code and data will be publicly released at https://github.com/RUCAIBox/Perceval. --- ### Diffusion Model as a Generalist Segmentation Learner Authors: Haoxiao Wang, Antao Xiang, Haiyang Sun, Peilin Sun, Changhao Pan, Yifu Chen, Minjie Hong, Weijie Wang, Shuang Chen, Yue Chen, Zhou Zhao Published: 2026-04-27 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.24575v1 Diffusion models are primarily trained for image synthesis, yet their denoising trajectories encode rich, spatially aligned visual priors. In this paper, we demonstrate that these priors can be utilized for text-conditioned semantic and open-vocabulary segmentation, and this approach can be generalized to various downstream tasks to make a general-purpose diffusion segmentation framework. Concretely, we introduce DiGSeg (Diffusion Models as a Generalist Segmentation Learner), which repurposes a pretrained diffusion model into a unified segmentation framework. Our approach encodes the input image and ground-truth mask into the latent space and concatenates them as conditioning signals for the diffusion U-Net. A parallel CLIP-aligned text pathway injects language features across multiple scales, enabling the model to align textual queries with evolving visual representations. This design transforms an off-the-shelf diffusion backbone into a universal interface that produces structured segmentation masks conditioned on both appearance and arbitrary text prompts. Extensive experiments demonstrate state-of-the-art performance on standard semantic segmentation benchmarks, as well as strong open-vocabulary generalization and cross-domain transfer to medical, remote sensing, and agricultural scenarios-without domain-specific architectural customization. These results indicate that modern diffusion backbones can serve as generalist segmentation learners rather than pure generators, narrowing the gap between visual generation and visual understanding. --- ### FastOMOP: A Foundational Architecture for Reliable Agentic Real-World Evidence Generation on OMOP CDM data Authors: Niko Moeller-Grell, Shihao Shenzhang, Zhangshu Joshua Jiang, Richard JB Dobson, Vishnu V Chandrabalan Published: 2026-04-27 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2604.24572v1 The Observational Medical Outcomes Partnership Common Data Model (OMOP CDM), maintained by the Observational Health Data Sciences and Informatics (OHDSI) collaboration, enabled the harmonisation of electronic health records data of nearly one billion patients in 83 countries. Yet generating real-world evidence (RWE) from these repositories remains a manual process requiring clinical, epidemiological and technical expertise. LLMs and multi-agent systems have shown promise for clinical tasks, but RWE automation exposes a fundamental challenge: agentic systems introduce emergent behaviours, coordination failures and safety risks that existing approaches fail to govern. No infrastructure exists to ensure agentic RWE generation is flexible, safe and auditable across the lifecycle. We introduce FastOMOP, an open-source multi-agent architecture that addresses this gap by separating three infrastructure layers, governance, observability and orchestration, from pluggable agent-teams. Governance is enforced at the process boundary through deterministic validation independent of agent reasoning, ensuring no compromised or hallucinating agent can bypass safety controls. Agent teams for phenotyping, study design and statistical analysis inherit these guarantees through controlled tool exposure. We validated FastOMOP using a natural-language-to-SQL agent team across three OMOP CDM datasets: synthetic data from Synthea, MIMIC-IV and a real-world NHS dataset from Lancashire Teaching Hospitals (IDRIL). FastOMOP achieved reliability scores of 0.84-0.94 with perfect adversarial and out-of-scope block rates, demonstrating process-boundary governance delivers safety guarantees independent of model choice. These results indicate that the reliability gap in RWE deployment is architectural rather than model capability, and establish FastOMOP as a governed architecture for progressive RWE automation. --- ### MEG-RAG: Quantifying Multi-modal Evidence Grounding for Evidence Selection in RAG Authors: Xihang Wang, Zihan Wang, Chengkai Huang, Quan Z. Sheng, Lina Yao Published: 2026-04-27 Categories: cs.CL, cs.IR, cs.IT arXiv: https://arxiv.org/abs/2604.24564v2 Multimodal Retrieval-Augmented Generation (MRAG) addresses key limitations of Multimodal Large Language Models (MLLMs), such as hallucination and outdated knowledge. However, current MRAG systems struggle to distinguish whether retrieved multimodal data truly supports the semantic core of an answer or merely provides superficial relevance. Existing metrics often rely on heuristic position-based confidence, which fails to capture the informational density of multimodal entities. To address this, we propose Multi-modal Evidence Grounding (MEG), a semantic-aware metric that quantifies the contribution of retrieved evidence. Unlike standard confidence measures, MEG utilizes Semantic Certainty Anchoring, focusing on high-IDF information-bearing tokens that better capture the semantic core of the answer. Building on MEG, we introduce MEG-RAG, a framework that trains a multimodal reranker to align retrieved evidence with the semantic anchors of the ground truth. By prioritizing high-value content based on semantic grounding rather than token probability distributions, MEG-RAG improves the accuracy and multimodal consistency of generated outputs. Extensive experiments on the M$^2$RAG benchmark show that MEG-RAG consistently outperforms strong baselines and demonstrates robust generalization across different teacher models. --- ### Enhancing molecular dynamics with equivariant machine-learned densities Authors: Mihail Bogojeski, Muhammad R. Hasyim, Leslie Vogt-Maranto, Klaus-Robert Müller, Kieron Burke, Mark E. Tuckerman Published: 2026-04-27 Categories: physics.chem-ph, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2604.24563v1 Machine-learning interatomic potentials (MLIPs) have enabled molecular dynamics at near ab initio accuracy, yet remain limited to energies and forces by construction, leaving electronic observables such as dipole moments and polarizabilities inaccessible. We introduce DenSNet, a density-first approach to machine-learned electronic structure that learns the Hohenberg--Kohn map from nuclear configurations to the ground-state electron density. Our approach employs an SE(3)-equivariant neural network to predict density coefficients of a flexible atom-centered Gaussian basis, combined with a $Δ$-learning strategy that uses superposed atomic densities as a prior to accelerate training. A second equivariant network then maps the predicted density to the total energy, providing a unified framework for molecular dynamics and electronic structure. We validate DenSNet on ethanol, ethanethiol, and resorcinol, where infrared spectra from machine-learned trajectories show excellent agreement with experimental gas-phase measurements. To test scalability, we train on polythiophene oligomers with 1--6 monomers and extrapolate to chains of up to 12 monomers, generating stable long-time trajectories whose infrared spectra agree with reference density functional theory calculations. Here, we show that reinstating the electron density as the central learned quantity opens a practical route to transferable prediction of spectroscopic and electronic observables in large-scale molecular simulations. --- ### Towards Lawful Autonomous Driving: Deriving Scenario-Aware Driving Requirements from Traffic Laws and Regulations Authors: Bowen Jian, Rongjie Yu, Hong Wang, Liqiang Wang, Zihang Zou Published: 2026-04-27 Categories: cs.AI, cs.CL, cs.CY arXiv: https://arxiv.org/abs/2604.24562v1 Driving in compliance with traffic laws and regulations is a basic requirement for human drivers, yet autonomous vehicles (AVs) can violate these requirements in diverse real-world scenarios. To encode law compliance into AV systems, conventional approaches use formal logic languages to explicitly specify behavioral constraints, but this process is labor-intensive, hard to scale, and costly to maintain. With recent advances in artificial intelligence, it is promising to leverage large language models (LLMs) to derive legal requirements from traffic laws and regulations. However, without explicitly grounding and reasoning in structured traffic scenarios, LLMs often retrieve irrelevant provisions or miss applicable ones, yielding imprecise requirements. To address this, we propose a novel pipeline that grounds LLM reasoning in a traffic scenario taxonomy through node-wise anchors that encode hierarchical semantics. On Chinese traffic laws and OnSite dataset (5,897 scenarios), our method improves law-scenario matching by 29.1\% and increases the accuracy of derived mandatory and prohibitive requirements by 36.9\% and 38.2\%, respectively. We further demonstrate real-world applicability by constructing a law-compliance layer for AV navigation and developing an onboard, real-time compliance monitor for in-field testing, providing a solid foundation for future AV development, deployment, and regulatory oversight. --- ### Aligned Multi-View Scripts for Universal Chart-to-Code Generation Authors: Zhihan Zhang, Lizi Liao Published: 2026-04-27 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.24559v1 Chart-to-code generation converts a chart image into an executable plotting script, enabling faithful reproduction and editable visualizations. Existing methods are largely Python-centric, limiting practical use and overlooking a critical source of supervision: the same chart can be expressed by semantically equivalent scripts in different plotting languages. To fill this gap, we introduce Chart2NCode, a dataset of 176K charts paired with aligned scripts in Python, R, and LaTeX that render visually equivalent outputs, constructed via a metadata-to-template pipeline with rendering verification and human quality checks. Building on a LLaVA-style architecture, we further propose CharLuMA, a parameter-efficient adaptation module that augments the multimodal projector with a language-conditioned mixture of low-rank subspaces, allowing the model to share core chart understanding while specializing code generation to the target language through lightweight routing. Extensive experiments show consistent gains in executability and visual fidelity across all languages, outperforming strong open-source baselines and remaining competitive with proprietary systems. Further analyses reveal that balanced multi-language supervision benefits all languages and that the adapter allocates a compact shared core plus language-specific capacity. Codes and data are available at https://github.com/Zhihan72/CharLuMA. ---