# Research Papers — 2026-05 · bundle page 321 of 326 > Full-text markdown of papers #16001–16050 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-05/page/321 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/320.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/322.md) ### Psychologically Potent, Computationally Invisible: LLMs Generate Social-Comparison-Eliciting Posts They Fail to Detect Authors: Hua Zhao, Jiapei Gu, Michelle Mingyue Gu Published: 2026-05-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.01017v2 We introduce Xiaohongshu Social Comparison Reader Elicitation (XHS-SCoRE), a reader-grounded benchmark for detecting whether text-only Xiaohongshu (RedNote) posts elicit Upward, Downward, or Neutral/no clear social comparison from a first-person reader perspective. The task targets a socially meaningful relational, behaviorally real signal not reducible to sentiment. Across prompted LLM classifiers and supervised Chinese encoders, we find a consistent generation--detection mismatch: the signal is textually learnable in-domain, but not robustly accessible to prompt-based classification. Prompted LLM classifiers show stable failures, especially neutralization of comparison-eliciting posts and model-specific directional skew. A controlled pilot shows that LLM-generated Xiaohongshu-style posts can shift perceived standing and comparison-related affect even when prompt-based detection of the same construct remains fragile. XHS-SCoRE contributes a benchmark for reader-grounded comparison detection and a diagnostic framework for studying when socially meaningful relational cues remain only partially visible to prompt-based inference. --- ### CLEAR: Revealing How Noise and Ambiguity Degrade Reliability in LLMs for Medicine Authors: Kevin H. Guo, Chao Yan, Avinash Baidya, Katherine Brown, Xiang Gao, Juming Xiong, Zhijun Yin, Bradley A. Malin Published: 2026-05-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.01011v2 Medical large language model (LLM) evaluations rely on simplified, exam-style benchmarks that rarely reflect the ambiguity of real-world medical inquiries. We introduce the CLinical Evaluation of Ambiguity and Reliability (CLEAR) framework, which assesses how decision-space presentation, ambiguity, and uncertainty affect LLMs' reasoning on medical benchmarks. CLEAR systematically perturbs (1) the number of plausible answer options, (2) the presence of a ground truth or abstention option, and (3) the semantic framing of answer options. Applying CLEAR on three benchmarks evaluated across 17 LLMs reveals three notable limitations of existing evaluation methods. First, increasing the number of plausible answers degrades a model's ability to identify the correct answer and abstain against incorrect ones. Second, this lack of caution intensifies as the framing of abstention shifts from assertive rejection like "None of the Above" to uncertainty admission like "I don't know" (IDK). Notably, just including IDK in the answer space increases incorrect answer selections. Lastly, we formalize the performance gap between identifying the correct answer and abstaining from incorrect ones as the humility deficit, which worsens with model scale. Our findings reveal limitations in standard medical benchmarks and underscore that scaling alone does not resolve LLM reliability issues. --- ### Can AI Debias the News? LLM Interventions Improve Cross-Partisan Receptivity but LLMs Overestimate Their Own Effectiveness Authors: Faisal Feroz, Jonas R. Kunst Published: 2026-05-01 Categories: cs.CL, cs.CY arXiv: https://arxiv.org/abs/2605.01006v3 Partisan news media erode cross-partisan trust, but large language models (LLMs) offer the potential of debiasing such content at scale. Across two pre-registered experiments, we tested whether LLM-generated debiasing of liberal news headlines improves conservative readers' trust-relevant judgments. In Study 1, subtle lexical debiasing (replacing emotive words with moderate synonyms) had no effect on any outcome. Study 2 found that a more substantive reframing intervention significantly increased conservatives' perceived trustworthiness, completeness, and willingness to engage with liberal news headlines, without producing a backfire effect among liberals. In Study 1, the intervention produced robust effects across silicon participants simulated with six different models (o3-mini, o3, GPT-4o mini, GPT-4o, GPT-5 mini, and GPT-5), whereas it had no impact on human readers. In Study 2, the intervention's effects among silicon participants generally aligned directionally with human responses but were significantly larger for some outcomes, and three models (o3, GPT-4o mini, and GPT-4o) incorrectly predicted a liberal backfire effect absent in humans. Moderation analyses revealed that the models' implicit theory of who responds to debiasing diverged from the psychological profile that actually predicted human responsiveness. Most strikingly, in Study 2, participants simulated by each of the six models suggested that debiasing effects would be stronger among participants high in political in-group identification. Yet, no such moderation was observed among human participants. These findings demonstrate that LLM-based debiasing can improve cross-partisan receptivity when targeting ideological framing rather than surface-level language, but that current models lack both the quantitative accuracy and qualitative psychological fidelity to evaluate their own interventions without human oversight. --- ### Pi-Change: A Prior-Informed Multiple Change Point Detection Algorithm Authors: Jonathon Jacobs, Shanshan Chen Published: 2026-05-01 Categories: stat.ME, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2605.01003v1 Statistical change point (CP) detection methods typically rely on likelihood-based inference and ignore contextual information about plausible CP locations beyond the observed sequence. Although informative priors provide a natural way to incorporate such information, general and computationally efficient methods for doing so are lacking, especially for multiple CP detection. To address this gap, we propose a prior-informed CP detection algorithm (Pi-Change) that incorporates prior information on CP locations through a time-varying penalty term. We prove that the proposed penalty can be embedded in the Pruned Exact Linear Time framework while preserving the dynamic programming recursion and pruning rule required for efficient multiple CP detection. Across simulation studies and three time-series applications, Pi-Change discourages spurious CPs unsupported by prior information, remains robust to prior misspecification, and improves detection accuracy. More broadly, Pi-Change extends multiple CP detection beyond purely data-driven fitting by incorporating partial prior knowledge in a computationally efficient and interpretable way. It is particularly useful when CPs arise from heterogeneous mechanisms or are associated with known external events, helping quantify the delay between an event and the resulting structural change. --- ### Most Current Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives Authors: Mohammed Abu Baker, Luca Baroni, Dan Wilhelm Published: 2026-05-01 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.00994v2 Finetuning can significantly modify the behavior of large language models, including introducing harmful or unsafe behaviors. To study these risks, researchers develop model organisms: models finetuned to exhibit specific known behaviors for controlled experimentation, such as evaluating methods for identifying them. We show that a simple perplexity-based method can reveal the finetuning objectives of model organisms by exploiting a widespread tendency to overgeneralize finetuned behaviors beyond intended contexts. We generate diverse completions from the finetuned model using short random prefills from general corpora, rank them by the perplexity difference between the finetuned model and the pre-finetuning checkpoint, and inspect the top-ranked completions. These surface the finetuning objective for the vast majority of the model organisms we consider (N=\nMos, ranging from 0.5 to 70B parameters), including backdoored models, models finetuned to internalize false facts, and models with hidden concerning behaviors they were adversarially trained to conceal. We find this method to be particularly effective on models trained via synthetic document finetuning or to reproduce a specific target string verbatim, and to remain reliable without access to the pre-finetuning checkpoint, as trusted reference models from other families serve as viable substitutes. Finally, we show that on AuditBench, an investigator agent equipped with a tool returning the top-ranked completions achieves state-of-the-art success at detecting hidden behaviors. --- ### Democratizing the medieval English legal tradition Authors: Michael Zhang, Elise Wang, Charlotte Whatley, Seth Strickland, Dylan Bannon Published: 2026-05-01 Categories: cs.CV, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.00977v1 The record of the beginning of the most widespread legal system in the world is contained in millions of pages of handwritten text. Most of the records of the first centuries of the Anglo-American legal system are hand-written in a highly abbreviated form of medieval Latin which only a few dozen scholars in the world are trained to read. In this interdisciplinary project, we construct a dataset of 4029 lines of text across 193 medieval criminal and civil cases. We then use the dataset to train an open-source end-to-end pipeline for transcribing these manuscripts. We first train standard neural network architectures for line segmentation and handwriting recognition (R-Blla and CNN+LSTM with CTC decoding, respectively) and show that they can already achieve 79% word accuracy, despite the relatively small training set and the challenge of expanding abbreviations. We then demonstrate that simple post-processing significantly boosts accuracy: adding an n-gram language model to the CTC decoder improves word accuracy to 82%, while asking Gemini Pro 3 to correct mistakes boosts accuracy to 88%. Finally, we compare the CNN+LSTM architecture with TrOCR, a transformer-based OCR architecture, demonstrating that TrOCR shows comparable word accuracy but worse character accuracy due to its over-willingness to guess, making it harder for humans to infer the correct reading. We incorporated our pipeline into a web portal (glyphmachina.com), opening up the English legal tradition to legal scholars, medievalists, and students. --- ### SRTJ: Self-Evolving Rule-Driven Training-Free LLM Jailbreaking Authors: Jindong Li, Ying Liu, Yali Fu, Jinjing Zhu, Leyao Wang, Menglin Yang, Rex Ying Published: 2026-05-01 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2605.00974v1 LLMs are increasingly equipped with safety alignment mechanisms, yet recent studies demonstrate that they remain vulnerable to jailbreaking attacks that elicit harmful behaviors without explicit policy violations. While a growing body of work has explored automated jailbreak strategies, existing methods face several fundamental challenges, including the lack of systematic utilization of both successful and failed attack experiences, as well as the absence of principled mechanisms for composing and selecting reusable attack rules under diverse constraints. As a result, existing methods struggle to accumulate transferable knowledge over time and to reliably adapt attack strategies across different targets and evolving safety mechanisms. To address these issues, we propose a Self-Evolving Rule-Driven Training-Free Jailbreak (SRTJ) framework that systematically discovers, composes, and refines attack strategies through interaction and feedback, without updating model parameters. Specifically, SRTJ couples experience-driven attack generation with answer set programming (ASP)-based rule selection and constraint-aware composition, where iterative verifier feedback is leveraged to jointly refine successful strategies and analyze failure patterns. The resulting rule memory evolves in a hierarchical multi-level manner, explicitly organizing distilled attack knowledge into long-term, middle-term, and short-term rules, thereby capturing both stable transferable strategies and transient adaptive behaviors to effectively balance exploration and exploitation across attack attempts. Extensive experiments on mainstream jailbreak benchmark (HarmBench) demonstrate that SRTJ achieves strong and stable attack performance across different target LLMs, while exhibiting improved robustness and generalization compared to existing jailbreak methods. The code is available at https://github.com/TheSolkatt/SRTJ. --- ### Physiology-Aware Masked Cross-Modal Reconstruction for Biosignal Representation Learning Authors: Hao Zhou, Simon A. Lee, Cyrus Tanade, Keum San Chun, Juhyeon Lee, Migyeong Gwak, Megha Thukral, Justin Sung, Eugene Hwang, Mehrab Bin Morshed, Li Zhu, Viswam Nathan et al. Published: 2026-05-01 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2605.00973v1 Biosignals acquired from different locations on the body often provide temporally ordered views of the same underlying physiological process. However, most existing self supervised learning methods treat these signals as interchangeable views, overlooking the directional temporal dynamics that link them. A canonical example is the relationship between electrocardiography (ECG), which captures the electrical activation initiating each heartbeat, and photoplethysmography (PPG), which records the resulting peripheral pulse delayed by vascular dynamics. To capture this structured relationship, we introduce xMAE, a biosignal pretraining framework that leverages masked cross modal reconstruction across temporally ordered biosignals as a training time constraint to encourage physiologically meaningful timing structure in the learned representations. We show that pretraining with xMAE yields representations that outperform both unimodal and multimodal baselines on 15 of 19 downstream tasks, including cardiovascular outcome prediction, abnormal laboratory test detection, sleep staging, and demographic inference, while generalizing across devices, body locations, and acquisition settings. Further analysis suggests that the ECG PPG timing structure is reflected in the learned PPG representations. More broadly, xMAE demonstrates the effectiveness of incorporating temporal structure into multimodal pretraining when signals observe different stages of a shared underlying process. Code is available at https://github.com/hzhou3/xMAE. --- ### Toward a Scientific Discovery Engine for Weather and Climate Data: A Visual Analytics Workbench for Embedding-Based Exploration Authors: Nihanth W. Cherukuru, Matt Rehme, Kirsten J. Mayer, David John Gagne, John Schreck, John Clyne, Charlie Becker Published: 2026-05-01 Categories: physics.data-an, cs.AI, cs.CV, cs.IR arXiv: https://arxiv.org/abs/2605.00972v2 Earth system science is producing increasingly large, high-dimensional datasets from both physics-based and AI-driven models. While embedding-based representations make these data searchable and serve as foundational building blocks for AI-driven discovery engines, nearest neighbors in latent spaces are not automatically scientifically meaningful. They may reflect real meteorological structures, or simply artifacts of preprocessing, geography, or model bias. Researchers therefore need visual tools to inspect latent space organization, trace search results back to physical evidence, and evaluate candidate representations against one another. We present an open source visual analytics workbench designed to support this provenance-aware scientific retrieval workflow. The system links distinct embedding experiments to shared source data, metadata, spatial contexts, and model configurations. It enables interactive retrieval strategy design by allowing users to issue image-level and localized patch-level queries, apply multi-constraint filters, and inspect analogs through familiar meteorological views. This facilitates a discovery loop where scientists characterize a phenomenon in a well-understood dataset and use its latent signature to probe larger archives. While we demonstrate the workbench through a tropical cyclone retrieval scenario using a vision foundation model (DINOv3) on ERA5 data, the framework is model-agnostic and designed to integrate with other embedding architectures in the future. Finally, we evaluate its out-of-core retrieval backend, demonstrating that interactive visual search over tens of millions of embeddings is highly scalable on commodity hardware. --- ### Reconstruction Interval Z-Phase Dependence of AI Detection Sensitivity in CT Lung Nodule Screening Authors: Dan Soliman Published: 2026-05-01 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2605.00971v1 Background: Sensitivity of AI-assisted lung nodule detection systems is known to vary with CT acquisition parameters including radiation dose, reconstruction kernel, and slice thickness. However, the dependence of detection probability on nodule position within the reconstruction cycle -- the z-phase -- has not, to the author's knowledge, been characterized for deep learning-based detection systems. Methods: A retrospective analysis was performed using the LIDC-IDRI dataset. Detection results from a previously validated 154-case perturbation study were re-analyzed. For each consensus nodule (>=4-reader agreement), z-phase was defined as the fractional position of the nodule center within the reconstruction cycle, folded to (0, 0.5). Detection sensitivity was stratified by z-phase bin, reconstruction interval (1mm, 3mm, 5mm), and by the ratio of reconstruction interval to nodule diameter (d/D). Results: At 5mm reconstruction interval, sensitivity was 71.6% vs 84.8% at 1mm baseline. Within the 5mm condition, sensitivity varied by 17.6 percentage points across z-phase bins. Stratified by d/D ratio, sensitivity was 92.4% for d/D < 0.5, 78.0% for 0.5 <= d/D < 1.0, and 61.4% for d/D >= 1.0, with a systematic z-phase effect present only in the d/D >= 1.0 stratum. Conclusions: AI detection sensitivity depends on the ratio of reconstruction interval to nodule diameter. When this ratio approaches or exceeds 1.0 -- as occurs for 3-6mm nodules at 5mm reconstruction -- z-phase becomes the dominant source of per-study detection variance. This stochastic effect is invisible to protocol-level quality metrics and not reflected in AI confidence scores. --- ### MedMosaic: A Challenging Large Scale Benchmark of Diverse Medical Audio Authors: Harshit Rajgarhia, Shuubham Ojha, Asif Shaik, Akhil Pothanapalli, Rachuri Lokesh, Abhishek Mukherji, Prasanna Desikan Published: 2026-05-01 Categories: cs.SD, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.00969v2 Medical audio data is difficult to collect due to privacy regulations and high annotation costs arising from domain expertise. Thus, existing benchmarks tend to underrepresent complex medical audio scenarios. To address this challenge, we present MedMosaic, a medical audio question-answering dataset designed to benchmark language and audio reasoning models under realistic clinical constraints. MedMosaic features a diverse range of medical audio types, including condition-related physiological sounds, carefully constructed synthetic voices to mimic speech with artifacts as well as real short and long length clinical conversations to model varying context lengths. The dataset also features a total of 46,701 question-answer pairs, spanning categories such as multiple-choice, sequential multi-turn, and open-ended question-answers, enabling systematic evaluation of multi-hop reasoning and answer generation capabilities. Benchmarking 13 audio and multimodal reasoning models reveals that reasoning remains challenging for all evaluated systems, with substantial performance variation across question types. In particular, even state-of-the-art model like Gemini-2.5-pro can only achieve 68.1% accuracy approximately. These findings underscore persistent limitations in medical reasoning and highlight the need for more robust, domain-specific multimodal reasoning models. A sample of benchmark data is available here: https://shorturl.at/Lyp33 --- ### Adaptive 3D-RoPE: Physics-Aligned Rotary Positional Encoding for Wireless Foundation Models Authors: Chenyu Zhang, Xinchen Lyu, Chenshan Ren, Yanzhao Hou, Xuefei Zhang, Shuhan Liu, Qimei Cui Published: 2026-05-01 Categories: eess.SP, cs.AI arXiv: https://arxiv.org/abs/2605.00968v2 Wireless foundation models (WFMs) have emerged as a promising paradigm for unified channel state information (CSI) acquisition across diverse tasks in sixth-generation (6G) networks. Although WFMs significantly outperform task-specific small models, their zero-shot cross-scenario generalization still remains limited for real-world applications. Existing positional embeddings, the sole interface through which self-attention perceives the temporal-frequency-antenna 3D physical coordinates of CSI, fail to capture the highly dynamic and axis-dependent coherence inherent in wireless channels. This paper proposes Adaptive 3D-RoPE, a channel-driven 3D rotary positional embedding framework for WFMs to dynamically align the 3D positional embeddings with the instantaneous coherence state of heterogeneous CSI. The design proceeds in three stages: first, an axis-wise learnable rotary prior independently preserves the temporal, frequency, and antenna coordinate structures; second, a feature-guided rotary modulation module maps the feature-wise standard deviation of visible CSI tokens to compact, sample-adaptive scales; third, identical coordinate offsets induce dynamically adjusted query-key interactions tailored to the instantaneous channel state. Extensive experiments on both simulated and measured datasets validate the effectiveness of Adaptive 3D-RoPE across three complementary dimensions. It reduces NMSE by 10.14, 6.25, and 4.61 dB relative to baselines under antenna, temporal, and frequency scaling, respectively. It transfers effectively to real-world measured CSI and remains robust under imperfect CSI. Finally, it transfers to the independently designed LWM backbone and beam-prediction task, improving zero-shot Top-1 accuracy by 8.03 percentage points. --- ### Adaptive 3D-RoPE: Physics-Aligned Rotary Positional Encoding for Wireless Foundation Models Authors: Chenyu Zhang, Xinchen Lyu, Chenshan Ren, Shuhan Liu, Qimei Cui Published: 2026-05-01 Categories: eess.SP, cs.AI arXiv: https://arxiv.org/abs/2605.00968v1 Positional encoding plays a pivotal role in determin?ing the extrapolation and generalization performance of wireless foundation models for channel state information (CSI) modeling, latent characterization, and task-specific prediction. However, existing CSI models inherit static or one-dimensional positional priors from natural language and vision architectures, which fundamentally misalign with the intrinsic physics of wireless channels by lacking explicit relative decay, collapsing the 3D spatio-temporal-frequency structure, and remaining scenario?rigid. This paper proposes Adaptive 3D-RoPE, a physics-aligned rotary positional encoding that establishes the structural corner?stone for wireless foundation models. The framework integrates a learnable, axis-decoupled 3D frequency bank to explicitly disentangle multi-dimensional phase dependencies, coupled with a lightweight channel-conditioned controller that dynamically modulates the prior via compact global CSI descriptors. This sample-adaptive mechanism transforms positional encoding from a static transformer component into a dynamic, coherence-aware inductive bias to resolve heterogeneous channel physics. Extensive experiments across 100 datasets demonstrate the superiority of the proposed scheme in both scale extrapolation and zero-shot generalization. Compared to the state-of-the-art, our method achieves up to a 10.7 dB reduction in normalized mean square error (NMSE) under 8 times antenna scale extrapolation. Given the same CSI input scales, our method can also improve zero-shot NMSE by 1.07 dB across unseen mobility scenarios and 0.90 dB in low-frequency-to-millimeter-wave tasks. --- ### Robust volatility updates for Hierarchical Gaussian Filtering Authors: Christoph Mathys, Nicolas Legrand, Peter Thestrup Waade, Nace Mikus, Lilian Aline Weber Published: 2026-05-01 Categories: cs.LG, cs.NE, q-bio.NC, stat.ML arXiv: https://arxiv.org/abs/2605.00966v1 Hierarchical Gaussian Filtering (HGF) networks allow for efficient updating of posterior distributions (beliefs) about hidden states of an agent's environment. HGF parent nodes can target the mean or variance of their children. New information entering at input nodes leads to a cascade of belief updates across the network according to one-step update equations for each node's mean and precision (inverse variance). However, the original form of the update equations for variance-targeting parents(volatility coupling) can in some regions of parameter space lead to negative posterior precision, a logical impossibility which causes the updating algorithm to terminate with an error. In this report, we introduce a modified quadratic approximation to the variational energy of volatility-coupled nodes that avoids negative posterior precision. The key idea is to interpolate between two quadratic expansions of the variational energy: one at the prior prediction and one at a second mode whose location is obtained in closed form via the Lambert W function. The resulting update equations are robust across the entire parameter space and faithfully track the variational posterior even for large prediction errors. --- ### Seeking Information with RAG-Assistants: Does Model Size Matter in Human-AI Collaborations? Authors: Lennard C. Froma, Tom Kouwenhoven, Maaike H. T. de Boer, Catholijn M. Jonker, Max J. van Duijn Published: 2026-05-01 Categories: cs.IR, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2605.00964v1 Much research on LLMs has focused on increasing benchmark performance. However, the evaluation of such models in real-world collaborative human-AI workflows has stayed behind. This work evaluates a chatbot-style assistant based on Retrieval-Augmented Generation (RAG) in a realistic multi-turn information-seeking scenario inspired by workplace settings where compliance with local legislation and secure handling of sensitive data are often key. Specifically, we examine the performance of humans (N=112) assisted by RAG-assistants compared to LLM-only or LLM+RAG baselines. In this setting, we investigate how underlying model size (3B, 8B, and 70B) shapes the human-AI collaborative dynamic and how it influences perceived usability and satisfaction. Results show that the performance gain of human-AI collaboration over the model-only baselines is significant, irrespective of model size, suggesting that hybrid systems are beneficial in information-seeking scenarios. Interestingly, however, perceived usability and satisfaction among participants showed little difference across model sizes. This demonstrates a nuanced trade-off between model size, performance, and user perception. Our work highlights the added value of evaluating AI applications in actual multi-turn interactions with human users, looking at usability and satisfaction besides accuracy, rather than focusing on benchmark performance only. --- ### Ablation Study of Multimodal Perception, Language Grounding, and Control for Human-Robot Interaction in an Object Detection and Grasping Task Authors: Zi Tian, Guanting Shen Published: 2026-05-01 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2605.00963v1 This manuscript extends our previous multimodal human-robot interaction system by introducing a controlled ablation study of the three modules that most strongly influence end-to-end performance: the large language model used for action extraction, the perception system used for visual grounding, and the controller used for motion execution. The goal is not to redesign the full pipeline, but to isolate the contribution of each component under a common experimental protocol and then evaluate the best combinations end-to-end. We therefore compare three language models, five perception configurations, and three controllers, followed by a second-stage factorial study over the best candidates. The resulting analysis is intended to clarify which choices primarily affect execution time, which primarily affect success rate, and where the largest engineering gains are likely to come from in future revisions of the system. --- ### Energy-Based Constraint Networks: Learning Structural Coherence Across Modalities Authors: Chirag Shinde Published: 2026-05-01 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2605.00960v1 We introduce energy-based constraint networks -- a modality-agnostic architecture that learns structural coherence from contrastive pairs. The system processes frozen encoder embeddings through a state-space model with dual-head attention, producing a scalar energy measuring structural consistency alongside per-position energy scores that localize violations. Multiple independently trained branches detect different violation types and compose at inference without interference. We demonstrate the framework in two domains. In text, the system achieves 93.4% accuracy on trained corruption types and 87.2% on 9 unseen types, using frozen BERT and 7.4M trainable parameters. In vision, the same architecture achieves competitive deepfake detection: 0.959 AUC on FaceForensics++ Deepfakes and 0.870 on Celeb-DF without any Celeb-DF training data, using frozen DINOv2 and 3.6M parameters per branch. The framework supports flexible training: branches learn from designer-specified corruptions, real-world paired data, or both. Composable branches require representation compatibility -- a finding validated through extensive experimentation where five incompatible approaches failed before the compatible one succeeded. The architecture is encoder-agnostic and domain-agnostic: changing the domain requires only new corruption strategies; changing the encoder requires only a new input projection layer. To our knowledge, this is the first architecture to learn within-modality structural coherence as an explicit energy landscape with per-position decomposition, and to demonstrate that the same architecture transfers across modalities via corruption respecification alone. --- ### "I Don't Know" -- Towards Appropriate Trust with Certainty-Aware Retrieval Augmented Generation Authors: Daan Di Scala, Maaike de Boer, Pınar Yolum Published: 2026-05-01 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2605.00957v1 Achieving the right amount of trust in AI systems is important, but challenging. The problem is exacerbated with the rise of Large Language Models (LLMs) as they provide human-level communication capabilities, but potentially hallucinate in the content that they generate. Moreover, they express over-confidence in their answers, making it difficult for users to judge their truthfulness. An important human value that users seek is benevolence, which can be met by LLM's self-reflection leading to reliable and honest answers. Accordingly, this paper proposes conveying appropriate levels of self-reflected certainty to build appropriate trust. Our contributions are twofold: 1) We develop CERTA (Certainty Enhanced RAG for Trustworthy Answers), a specialized Retrieval Augmented Generation (RAG) system that incorporates the relevance between question, context, and answer to reflect its uncertainty in answering questions; 2) We create the Certainty Benchmark with 90 question-context pairs of non-objective questions, divided over four categories (factuality, preference, sycophancy, morality) and three types of contexts (relevant, incomplete, irrelevant). We run experiments with a baseline RAG system and three CERTA settings using two LLMs. Our evaluations indicate that CERTA helps identify answers that are uncertain, decreases the cases of over-agreeing, and provides cautious behavior when prompted for moral judgments. --- ### E-MIA: Exam-Style Black-Box Membership Inference Attacks against RAG Systems Authors: Zelin Guan, Shengda Zhuo, Zeyan Li, Jinchun He, Wangjie Qiu, Zhiming Zheng, Shuqiang Huang Published: 2026-05-01 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2605.00955v1 Retrieval-Augmented Generation (RAG) equips large language models (LLMs) with external evidence by retrieving documents at inference time, but it also turns the retrieval corpusinto a sensitive asset. Under a black-box setting, an adversary given a candidate document can infer whether it has been ingested into the RAG knowledge base (i.e., document-level membership inference) solely from query response interactions, thereby leaking corpus coverage and the existence of sensitive topics. Existing RAG MIA methods either rely on soft signals such as semantic similarity, which often yield overlapping member/non-member score distributions and unstable thresholds, or employ explicit confirmation probes whose intent is conspicuous and thus prone to refusal and detection. We propose E-MIA, which converts verifiable hard evidence in the target document (e.g., fine-grained details, proper nouns/technical terms, definitional statements, metadata cues, and causal/constraint relations) into an exam with four objectively gradable question types (FB/SC/MC/T/F), and uses the aggregated exam score across multiple evidence targeted questions as the membership signal. Experiments across multiple datasets and diverse RAG configurations demonstrate that E-MIA improves member/non-member separability in stringent settings while preserving natural, stealthy queries, and we further analyze the impact of question composition and exam length on attack effectiveness. --- ### Graph Rewiring in GNNs to Mitigate Over-Squashing and Over-Smoothing: A Survey Authors: Hugo Attali, Nathalie Pernelle, Davide Buscaldi, Fragkiskos D. Malliaros Published: 2026-05-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.00951v1 Graph Neural Networks are powerful models for learning from graph-structured data, yet their effectiveness is often limited by two critical challenges: over-squashing, where information from distant nodes is excessively compressed, and over-smoothing, where repeated propagation makes node representations indistinguishable. Both phenomena stem from the interaction between message passing and the input topology, ultimately degrading information flow and limiting the performance of GNNs. In this survey, we examine graph rewiring techniques, a class of methods designed to modify the graph topology to enhance information propagation in GNNs. We provide a comprehensive review of state-of-the-art rewiring approaches, delving into their theoretical underpinnings, practical implementations, and performance trade-offs. --- ### Equation-Free Digital Twins for Nonlinear Structural Dynamics Authors: Mohammad Mahdi Abaei, Ahmad BahooToroody, Arttu Polojärvi, Heikki Remes, Ulf Tyge Tygesen, Mikko Suominen, Michael Beer Published: 2026-05-01 Categories: eess.SP, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2605.00950v2 Monitoring high-dimensional engineering structures in extreme environments is limited by non-stationary excitation, nonlinear structural kinematics, and stochastic forcing. Traditional model-based and black-box data-driven methods often struggle to resolve these dynamics in real time, particularly under sensor failure or partial observability. This paper introduces a rank-optimized digital twin framework based on Koopman operator theory, Hankel-matrix embeddings, and dynamic mode decomposition. By lifting operational data into a linear invariant subspace, the method enables autonomous, input-blind reconstruction of structural states without requiring a priori mass or stiffness matrices. The framework is validated on an NREL 5MW spar-buoy floating offshore wind turbine, representing a challenging coupled aero-hydro-servo-elastic system. Results show that the rank-optimized Koopman-Hankel manifold separates structural resonances from deterministic 3P rotor harmonics under colored noise, where standard subspace identification can be unreliable. A rolling-horizon virtual sensing strategy achieves high-fidelity reconstruction at critical structural hotspots, with coefficient of determination greater than 0.95 at 1 Hz data assimilation and accuracy exceeding 0.99 at higher sampling rates. By estimating a physical Lyapunov time of approximately 1.0 s, the study defines the predictability horizon associated with the system information barrier. The proposed framework provides a computationally efficient and resilient digital twin approach for real-time identification and virtual sensing of complex structural dynamics. --- ### Co-Generative De Novo Functional Protein Design Authors: Xinrui Chen, Yizhen Luo, Siqi Fan, Zaiqing Nie Published: 2026-05-01 Categories: q-bio.QM, cs.AI arXiv: https://arxiv.org/abs/2605.00948v1 De novo functional protein design aims to generate protein sequences that realize specified biochemical functions without relying on evolutionary templates, enabling broad applications in biotechnology and medicine. Existing approaches adopt either direct function-to-sequence mapping or decoupled structure-sequence generation strategies but often fail to achieve functionality and foldability simultaneously. To address this, we propose CodeFP, a Co-generative protein language model for de novo Functional Protein design that simultaneously decodes sequence and structure tokens, thereby enabling superior simultaneous realization of functionality and foldability. CodeFP utilizes functional local structures to enrich functional semantic encodings, overcoming the suboptimal translation of flat encodings into structure tokens, while introducing auxiliary functional supervision to alleviate training ambiguity stemming from the one-to-many structure-to-token mapping. Extensive experiments show that CodeFP consistently achieves average improvements of 6.1% in functional consistency and 3.2% in foldability over the strongest baseline. --- ### SCARV: Structure-Constrained Aggregation for Stable Sample Ranking in Redundant NLP Datasets Authors: Xu Zheng, Feiyu Wu, Linhong Wu, Zhuocheng Wang, Hui Li Published: 2026-05-01 Categories: cs.IR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.00944v1 Sample-level rankings are increasingly used in data-centric NLP for analysis, filtering, debugging, and curation, yet existing pipelines typically score training examples pointwise and rank them as if they were independent. This assumption is fragile in the presence of exact duplicates, near-duplicates, paraphrases, and other redundant structure common in NLP corpora, where stochastic training can make highly similar examples receive unstable relative orderings across random seeds. We study stable sample-level ranking under redundancy and propose \textsc{SCARV}, a modular aggregation framework that operates on top of an existing scoring proxy. \textsc{SCARV} combines robust multi-seed aggregation with a structure-aware aggregation/allocation step over redundancy clusters. Across synthetic redundancy, naturally mined QQP redundancy, multiple proxy families, several NLP tasks, and end-to-end DistilBERT fine-tuning, \textsc{SCARV} substantially improves over bare proxy rankings in global and local stability and yields more reproducible ranking-based decisions such as subset selection and suspicious-example retrieval. Our decomposition and compute-aware frontier sharpen the mechanism: robust multi-seed aggregation is the dominant generic stabilizer, while the structure-aware component adds value mainly under low aggregation budgets or when redundancy clusters are informative, naturally occurring, or sufficiently covered. These results position \textsc{SCARV} not as a universal data selector or a universally dominant replacement for seed-only aggregation, but as a stability-oriented aggregation layer for proxy-induced rankings in redundant NLP datasets. --- ### PPO guided Agentic Pipeline for Adaptive Prompt Selection and Test Case Generation Authors: Gourisetty Venkata Sai Koushik, Dama Aditya, Mahankali Harish Sai, Peddi Siddarhta, Shadab Ahmad, Vivek Yelleti Published: 2026-05-01 Categories: cs.SE, cs.LG arXiv: https://arxiv.org/abs/2605.00942v1 Developing effective test cases capable of thoroughly exercising large-scale software systems is inherently difficult, especially if such systems have voluminous, complex, and deeply nested source codes. In this work, we present a novel approach for generating test cases using a reinforcement learning-driven agentic framework where Proximal Policy Optimization (PPO) is coupled with an LLM engine to guide prompt selection during test generation. Our approach consists of two phases. In Phase I, the ToT-guided optimization agent partitions and minimizes the source code by removing redundancies without changing the functional behavior of the source code. In Phase II, a PPO-based policy network is trained to solve the problem of selecting prompts among eight different prompting techniques, such as Boundary Value Analysis, Random Fuzzing, etc., based on the inputted 11-dimensional state vector representing the source code complexity metrics and live coverage metrics to direct the LLM engine towards exploring unvisited paths in the program. The PPO agent receives rewards based on a combination of increases in line and branch coverages, penalties for unexplored branches, and rewards for reducing source code length. From experiments conducted on twenty benchmark programs, it is evident that the proposed approach, PPO-LLM, outperforms CBMC, kS-LLM, and kS-LLM++ in terms of branch and line coverage in almost all cases, for various loop bound values ranging from BOUND~1 to BOUND~2000. While at BOUND~1, the coverage of branches is 100\% using PPO-LLM on the PALS suite, in comparison, it is around 86.8\% using kS-LLM++. This confirms that adaptive prompt selection driven by PPO substantially outperforms static prompting strategies on PALS type programs. --- ### Uncertainty in a Single Pass: A Closed-Form Identity for One-Step Flow Matching Authors: Jiarui Xing, Song Wang, Jian Wang Published: 2026-05-01 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2605.00941v5 Flow matching provides a highly effective framework for generative modeling, yet estimating the uncertainty of its generated samples remains a fundamental challenge. Existing methods rely on auxiliary variance heads, model ensembles, or iterative covariance propagation, which invariably introduce accumulated errors and structural interference along the noise-to-image trajectory. We resolve these limitations by adapting Tweedie's formula to the linear flow matching interpolant, yielding an exact, closed-form identity for the posterior covariance. Because this identity relies exclusively on the divergence of the learned velocity field, it can be evaluated post hoc on pretrained models without any architectural modifications. This analytical formulation proves exceptionally powerful for single-step generative frameworks. By computing the end-to-end posterior covariance in one forward pass, our approach completely bypasses the compounding errors inherent to sequential integration. This mechanism fundamentally enhances robustness, providing highly accurate and stable uncertainty estimates while preventing the numerical degradation typical of multi-step approximations. We validate our framework across benchmark datasets, including CIFAR-10 and real brain MRI scans. The resulting divergence-based uncertainty maps are highly interpretable and tightly correlated with empirical reconstruction errors. Crucially, in medical imaging, this precise localization of structural ambiguity provides essential decision support for high-stakes clinical tasks such as tumor boundary delineation and neurosurgical planning. --- ### Divergence is Uncertainty: A Closed-Form Posterior Covariance for Flow Matching Authors: Jiarui Xing, Song Wang, Jian Wang Published: 2026-05-01 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2605.00941v4 Flow matching has become a leading framework for generative modeling, but quantifying the uncertainty of its samples remains an open problem. Existing approaches retrain the model with auxiliary variance heads, maintain costly ensembles, or propagate approximate covariance through many integration steps, trading off training cost, inference cost, or accuracy. We show that none of these trade-offs is necessary. By extending Tweedie's formula from the denoising setting to the flow matching interpolant, we derive an exact, closed-form expression for the posterior covariance at every point along the generative trajectory. The result depends on a single quantity, namely the divergence of the learned velocity field, which can be computed post-hoc on any pre-trained flow matching model, requiring no retraining and no architectural modification. For one-step generators such as MeanFlow, the same formula yields the end-to-end generation uncertainty in a single forward pass, eliminating the multi-step variance propagation required by all prior methods. Experiments on MNIST confirm that the resulting per-pixel uncertainty maps are semantically meaningful, concentrating on digit boundaries where inter-sample variation is highest, and that the scalar uncertainty score tracks actual prediction error, all at roughly $10^4 \times$ less total compute than ensembling or Monte Carlo dropout. --- ### Interpretable experiential learning based on state history and global feedback Authors: Anton Kolonin Published: 2026-05-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.00940v1 A new interpretable experiential learning model based on state history and global feedback is presented. It is capable of learning a behavioral model represented by a transition graph between sets of states, with transitions attributed with utility and evidence count. This model is expected to be suitable for solving reinforcement learning problem in resource-constrained environments. The model was thoroughly evaluated on the OpenAI Gym Atari Breakout benchmark, demonstrating performance comparable to some known neural network-based solutions. --- ### From Flat Facts to Sharp Hallucinations: Detecting Stubborn Errors via Gradient Sensitivity Authors: Yee Zhing Liew, Andrew Huey Ping Tan, Anwar P. P. Abdul Majeed Published: 2026-05-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.00939v2 Traditional hallucination detection fails on "Stubborn Hallucinations" - errors where LLMs are confidently wrong. We propose a geometric solution: Embedding-Perturbed Gradient Sensitivity (EPGS). We hypothesize that while robust facts reside in flat minima, stubborn hallucinations sit in sharp minima, supported by brittle memorization. EPGS detects this sharpness by perturbing input embeddings with Gaussian noise and measuring the resulting spike in gradient magnitude. This acts as an efficient proxy for the Hessian spectrum, differentiating stable knowledge from unstable memorization. Our experiments show that EPGS significantly outperforms entropy-based and representation-based baselines, providing a robust signal for detecting high-confidence factual errors. --- ### Fusing Urban Structure and Semantics: A Conditional Diffusion Model for Cross-City OD Matrix Generation Authors: Bin Chen, Zhuoya Meng, Fang Yang, Runkang Guo, Jingtao Ding, Yin Zhang, Chuan Ai, Zhengqiu Zhu Published: 2026-05-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.00938v1 Accurate modeling of commuting flows is important for urban governance, traffic planning, and resource allocation. However, the combined influence of individual intentions, geographic constraints, and social dynamics leads to considerable heterogeneity in commuting patterns, making it difficult to develop generation models that generalize across cities. To address this issue, we propose SEDAN, a Structure-Enhanced Diffusion model conditioned on Attributed Nodes for generalizable OD matrix generation. SEDAN models a city as an attributed graph. Each region is treated as a node with demographic and point-of-interest features, and commuting flows are modeled as weighted edges. Adjacency and distance matrices are incorporated to characterize spatial structure. Based on this representation, we design a fusion mechanism within SEDAN to jointly model semantic information and spatial information. Regional semantic attributes are used to model latent travel demand through graph-transformer-based node interactions, while spatial structure is injected into the generation process as explicit constraints. The adjacency matrix guides attention weights to strengthen interactions between neighboring regions. Meanwhile, the distance matrix serves as a diffusion condition to capture spatial proximity and travel impedance. The fusion of urban semantics and spatial constraints enables SEDAN to generate OD matrices that are both behaviorally plausible and geographically coherent. Experiments on real-world OD datasets from U.S. cities show that SEDAN achieves a 7.38\% improvement in RMSE over the state-of-the-art baseline, WEDAN. It also remains robust across heterogeneous urban scenarios and varying structural patterns. Our work provides an effective and generalizable solution for commuting OD matrix generation. The code is available at https://anonymous.4open.science/r/SEDAN. --- ### An ALE-Consistent Graph Neural Operator-Transformer Framework for Fluid-Structure Interaction Authors: Shihang Zhao, Martín Saravia, Haokui Jiang, Zhiyang Xue, Shunxiang Cao Published: 2026-05-01 Categories: physics.flu-dyn, cs.LG arXiv: https://arxiv.org/abs/2605.00937v1 We propose an arbitrary Lagrangian-Eulerian (ALE)-consistent machine learning framework for long-term fluid-structure interaction (FSI) prediction on deforming unstructured meshes. Specifically, the fluid dynamics are modeled by a surrogate that combines a graph neural operator (GNO) with a vision Transformer (ViT) for spatiotemporal prediction, while a lightweight long short-term memory (LSTM) network predicts structural kinematics at the interface. The two surrogates are coupled through a standard partitioned procedure. Most importantly, kinematic compatibility at the moving interface is enforced via an ALE-consistent boundary-correction step that updates the fluid-side interface velocity with the predicted structural velocity at each coupling update, thereby improving near-interface accuracy and long-term rollout stability. To mitigate autoregressive error accumulation, a two-stage training strategy is adopted, consisting of single-step supervised pretraining followed by long-term autoregressive fine-tuning. The proposed framework is validated on the benchmark problem of a flexible beam vibration in the wake of a cylinder. Results demonstrate accurate phase-consistent predictions over long rollouts and robust generalization under inlet-profile variations in both interpolation and extrapolation settings. Systematic ablation studies further assess the respective contributions of the ViT module, ALE-consistent boundary correction, and long-term training to predictive accuracy and rollout robustness. --- ### EventADL: Open-Box Anomaly Detection and Localization Framework for Events in Cloud-Based Service Systems Authors: Luan Pham, Victor Nicolet, Joey Dodds, Hui Guan, Daniel Kroening Published: 2026-05-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.00936v1 Anomaly detection and localization (ADL) is critical for maintaining reliability and availability in cloud systems. Recent ADL developments focus on metric and log data, leaving event data unexplored. To address this gap, we propose EventADL, the first open-box event-based ADL framework for cloud-based service systems. To motivate the design of our framework, we conduct a systematic analysis on 520 real-world incidents, and provide insights into how anomalies and their root causes manifest through event data. EventADL has three phases: offline training, online anomaly detection, and root cause localization. During the training phase, EventADL first learns Event Semantic Patterns (ESPs), which capture normal interactions between system entities using historical event data, and then learns Event Frequency Patterns (EFPs), which capture the normal frequency of known ESPs. In the online anomaly detection phase, any data in the event stream that deviates significantly from either pattern is identified as anomalous. For localization, EventADL constructs an Intervention Graph that models the relationships between recent system interactions and the detected anomalies for automatic root cause localization. The framework is designed to operate efficiently with unlabeled data and to produce interpretable anomalies with their corresponding root causes. Our evaluation on three real cloud service systems and two real-world incidents demonstrates that EventADL outperforms existing methods, achieving F1-scores of at least 90% for anomaly detection and 100% top-3 accuracy in root cause localization. --- ### Watch Your Step: Information Injection in Diffusion Models via Shadow Timestep Embedding Authors: An Huang, Junggab Son, Zuobin Xiong Published: 2026-05-01 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2605.00935v1 Diffusion models have become the foundation of modern generative systems, with most research focusing primarily on improving generation efficiency and output quality. The timestep embedding component is a crucial part of the diffusion pipeline, which provides a temporal conditioning signal to the denoising network, enabling it to adapt its predictions across different noise levels throughout the process. Despite their potential to contain substantial information, timestep embeddings remain underexplored in current research, especially for security risks and reliable provenance. To fill this gap, we introduce Shadow Timestep Embedding (STE), a novel mechanism that investigates the underutilized temporal space for malicious information injection into diffusion models. In particular, when zooming in on the timestep embedding space, we find that different timesteps exhibit distinct representational capabilities that can encode side-channel information. Moreover, such encoded information can be utilized for attack and defense purposes through the scheduler interface. We present a theoretical analysis of timestep embeddings as position-encoding mappings and derive a mutual coherence evaluation that explains the separability of disjoint timestep intervals. Our findings reveal the diffusion model's timestep as a powerful side channel for carrying dedicated information, motivating new directions for adversarial generative modeling by understanding the temporal dimension. --- ### Structured Analytic Coherent Point Drift for Non-Rigid Point Set Registration Authors: Wei Feng, Haiyong Zheng Published: 2026-05-01 Categories: cs.LG, cs.CV, stat.ML arXiv: https://arxiv.org/abs/2605.00934v2 Coherent Point Drift (CPD) is a representative probabilistic framework for unsupervised non-rigid point set registration. Its standard non-rigid M-step, however, relies on a point-indexed Gaussian-kernel system whose size grows with the number of moving points, making deformation estimation computationally heavy for large point sets and difficult to control in complexity during registration. To address these limitations, we propose Analytic-CPD, a new unsupervised non-rigid registration framework that gives CPD a structured analytic reformulation. Analytic-CPD preserves the CPD posterior correspondence layer, but lifts the M-step from point-indexed kernel displacement estimation to structured analytic mapping estimation. By coupling the Gaussian-mixture posterior mechanism of CPD with Structured Analytic Mappings (SAM), the method obtains a deformation model whose coefficient dimension is governed by the ambient dimension and analytic order rather than by the number of moving points. More importantly, deformation estimation is organized over an interpretable hierarchy of analytic function spaces, so the analytic order can be increased progressively as posterior correspondences become more reliable. We implement this idea through an increasing-degree continuation strategy with decreasing stage lengths: low-order analytic maps first stabilize the posterior correspondence structure, while higher-order modes later refine nonlinear residual deformation. Experiments on controlled model-matched, smooth model-mismatch, and registered human-shape data demonstrate the effectiveness and favorable accuracy--efficiency performance of Analytic-CPD. --- ### CGM-JEPA: Learning Consistent Continuous Glucose Monitor Representations via Predictive Self-Supervised Pretraining Authors: Hada Melino Muhammad, Zechen Li, Flora Salim, Ahmed A. Metwally Published: 2026-05-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.00933v1 Continuous Glucose Monitoring (CGM) can detect early metabolic subphenotypes (insulin resistance, IR; $β$-cell dysfunction), but population-scale deployment faces two coupled problems. First, the same physiological state appears through multiple views (CGM time series, venous OGTT, Glucodensity summaries), so single-view representations fail to transfer when deployment shifts the modality or setting. Second, baselines perform inconsistently across these shifts. Both problems point to one remedy: representations that abstract away from any single view to capture higher-level temporal and distributional structure. We propose CGM-JEPA, a self-supervised pretraining framework which predicts masked latent representations rather than raw values, yielding abstraction that transfers across modalities. X-CGM-JEPA adds a masked Glucodensity cross-view objective for complementary distributional information. We pretrain on $\sim$389k unlabeled CGM readings from 228 subjects and evaluate on two clinical cohorts ($N=27$ and $N=17$ public-release subsets) across three regimes (cohort generalization, venous-to-CGM transfer, home CGM) under 20-iteration $\times$ 2-fold cross-validation. X-CGM-JEPA ranks first or second on AUROC for both endpoints across all three regimes while no baseline does, exceeding the strongest baseline by up to $+6.5$ pp in cohort generalization and $+3.6$ pp in venous-to-CGM transfer (paired Wilcoxon, $p<0.001$). Under modality shift, it matches mean AUROC while redistributing toward weaker subgroups (ethnicity AUROC gap shrinks 25-54%); on sparse in-domain venous data, the distributional view lifts label-aware clustering (ARI $+39\%$, NMI $+40\%$). Code and weights: https://github.com/cruiseresearchgroup/CGM-JEPA --- ### Code World Model Preparedness Report Authors: Daniel Song, Peter Ney, Cristina Menghini, Faizan Ahmad, Aidan Boyd, Nathaniel Li, Ziwen Han, Jean-Christophe Testud, Saisuke Okabayashi, Maeve Ryan, Jinpeng Miao, Hamza Kwisaba et al. Published: 2026-05-01 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2605.00932v2 This report documents the preparedness assessment of Code World Model (CWM), a model for code generation and reasoning about code from Meta. We conducted pre-release testing across domains identified in our Frontier AI Framework as potentially presenting catastrophic risks, and also evaluated the model's misaligned propensities. Our assessment found that CWM does not pose additional frontier risks beyond those present in the current AI ecosystem. We therefore release it as an open-weight model. --- ### Hierarchical Federated Learning for Networked AI: From Communication Saving to Architecture-Aware Design Authors: Seyed Mohammad Azimi-Abarghouyi, Mehdi Bennis, Leandros Tassiulas Published: 2026-05-01 Categories: cs.LG, cs.DC, cs.IT arXiv: https://arxiv.org/abs/2605.00931v1 Federated learning (FL) is fundamentally a distributed optimization problem executed by communicating agents with local data, local computation, and partial system visibility. Once FL is viewed through that lens, hierarchy is not merely a scalability mechanism. It becomes the natural place to rethink how distributed optimization should be organized over real multi-tier networks. This article argues that hierarchical federated learning (HFL) should move beyond its common framing as a communication-saving protocol and instead be viewed as an architecture-aware design framework for networked AI. The framework is organized around three coupled design axes: architectural parameters, layer-wise optimization decomposition, and layer-wise communication realization. The first axis determines the coordination geometry of learning through hierarchy depth, layer asymmetry, and layered connectivity. The second determines how the global FL objective is decomposed across layers and highlights modular multi-layer optimization as a major opportunity beyond one dominant method everywhere. The third determines how the distributed optimization is physically realized under heterogeneous communication regimes, from interference-limited lower tiers to reliable upper tiers. A central message is that, in HFL, convergence becomes architecture-dependent: it is directly shaped by the chosen hierarchy, the assigned optimization roles, and the communication mechanisms that connect them. We develop this viewpoint using large-scale wireless edge intelligence as a flagship networked AI setting, then provide a comparative perspective on flat FL, two-tier HFL, and deep HFL together with a regime-oriented design map. The resulting perspective positions HFL as a practical methodology for designing future networked AI systems. --- ### Posterior Augmented Flow Matching Authors: George Stoica, Sayak Paul, Matthew Wallingford, Vivek Ramanujan, Abhay Nori, Winson Han, Ali Farhadi, Ranjay Krishna, Judy Hoffman Published: 2026-05-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00825v1 Flow matching (FM) trains a time-dependent vector field that transports samples from a simple prior to a complex data distribution. However, for high-dimensional images, each training sample supervises only a single trajectory and intermediate point, yielding an extremely sparse and high-variance training signal. This under-constrained supervision can cause flow collapse, where the learned dynamics memorize specific source-target pairings, mapping diverse inputs to overly similar outputs, failing to generalize. We introduce Posterior-Augmented Flow Matching (PAFM), a theoretically grounded generalization of FM that replaces single-target supervision with an expectation over an approximate posterior of valid target completions for a given intermediate state and condition. PAFM factorizes this intractable posterior into (i) the likelihood of the intermediate under a hypothesized endpoint and (ii) the prior probability of that endpoint under the condition, and uses an importance sampling scheme to construct a mixture over multiple candidate targets. We prove that PAFM yields an unbiased estimator of the original FM objective while substantially reducing gradient variance during training by aggregating information from many plausible continuation trajectories per intermediate. Finally, we show that PAFM improves over FM by up to 3.4 FID50K across different model scales (SiT-B/2 and SiT-XL/2), different architectures (SiT and MMDiT), and in both class and text conditioned benchmarks (ImageNet and CC12M), with a negligible increase in the compute overhead. Code: https://github.com/gstoica27/PAFM.git. --- ### HyCOP: Hybrid Composition Operators for Interpretable Learning of PDEs Authors: Jinpai Zhao, Nishant Panda, Yen Ting Lin, Eirik Valseth, Diane Oyen, Clint Dawson Published: 2026-05-01 Categories: cs.CE, cs.LG, math.NA arXiv: https://arxiv.org/abs/2605.00820v1 We introduce HyCOP, a modular framework that learns parametric PDE solution operators by composing simple modules (advection, diffusion, learned closures, boundary handling) in a query-conditioned way. Rather than learning a monolithic map, HyCOP learns a policy over short programs - which module to apply and for how long - conditioned on regime features and state statistics. Modules may be numerical sub-solvers or learned components, enabling hybrid surrogates evaluated at arbitrary query times without autoregressive rollout. Across diverse PDE benchmarks, HyCOP produces interpretable programs, delivers order-of-magnitude OOD improvements over monolithic neural operators, and supports modular transfer through dictionary updates (e.g., boundary swaps, residual enrichment). Our theory characterizes expressivity and gives an error decomposition that separates composition error from module error and doubles as a process-level diagnostic. --- ### When LLMs Stop Following Steps: A Diagnostic Study of Procedural Execution in Language Models Authors: Sailesh Panda, Pritam Kadasi, Abhishek Upperwal, Mayank Singh Published: 2026-05-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.00817v4 Large language models (LLMs) often achieve strong performance on reasoning benchmarks, but final-answer accuracy alone does not show whether they faithfully execute the procedure specified in a prompt. We introduce a controlled diagnostic benchmark for arithmetic procedural execution, where models are given a step-wise arithmetic procedure and two numeric inputs, and must return the final computed value. Complexity is varied through procedure length and look-back dependencies over intermediate variables. Average first-answer accuracy drops from 63% on 5-step procedures to 20\% on 95-step procedures. Generation-level analysis shows that failures often involve missing answers, premature answers, self-correction after an initial error and under-executed traces. These findings reveal a consistent decline in execution performance as arithmetic procedural complexity increases. --- ### Persistent Visual Memory: Sustaining Perception for Deep Generation in LVLMs Authors: Siyuan Huang, Xiaoye Qu, Yafu Li, Tong Zhu, Zefeng He, Muxin Fu, Daizong Liu, Wei-Long Zheng, Yu Cheng Published: 2026-05-01 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.00814v2 While autoregressive Large Vision-Language Models (LVLMs) demonstrate remarkable proficiency in multimodal tasks, they face a "Visual Signal Dilution" phenomenon, where the accumulation of textual history expands the attention partition function, causing visual attention to decay inversely with generated sequence length. To counteract this, we propose Persistent Visual Memory (PVM), a lightweight learnable module designed to strengthen sustained, on-demand access to visual evidence. Integrated as a parallel branch alongside the Feed-Forward Network (FFN) in LVLMs, PVM establishes a distance-agnostic retrieval pathway that directly provides visual embeddings for enhanced visual perception, thereby structurally mitigating the signal suppression inherent to deep generation. Extensive experiments on Qwen3-VL models demonstrate that PVM brings notable improvements with negligible parameter overhead, delivering consistent average accuracy gains across both 4B and 8B scales, particularly in complex reasoning tasks that demand persistent visual perception. Furthermore, in-depth analysis reveals that PVM shows improved robustness in longer generations and accelerates internal prediction convergence. --- ### Let ViT Speak: Generative Language-Image Pre-training Authors: Yan Fang, Mengcheng Lan, Zilong Huang, Weixian Lei, Yunqing Zhao, Yujie Zhong, Yingchen Yu, Qi She, Yao Zhao, Yunchao Wei Published: 2026-05-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00809v3 In this paper, we present \textbf{Gen}erative \textbf{L}anguage-\textbf{I}mage \textbf{P}re-training (GenLIP), a minimalist generative pretraining framework for Vision Transformers (ViTs) designed for multimodal large language models (MLLMs). To better align vision encoders with the autoregressive nature of LLMs, GenLIP trains a ViT to predict language tokens directly from visual tokens using a standard language modeling objective, without contrastive batch construction or an additional text decoder. This design offers three key advantages: (1) \textbf{Simplicity}: a single transformer jointly models visual and textual tokens; (2) \textbf{Scalability}: it scales effectively with both data and model size; and (3) \textbf{Performance}: it achieves competitive or superior results across diverse multimodal benchmarks. Trained on 8B samples from Recap-DataComp-1B, GenLIP matches or surpasses strong baselines despite using substantially less pretraining data. After continued pretraining on multi-resolution images at native aspect ratios, GenLIP further improves on detail-sensitive tasks such as OCR and chart understanding, making it a strong foundation for vision encoders in MLLMs. --- ### Let ViT Speak: Generative Language-Image Pre-training Authors: Yan Fang, Mengcheng Lan, Zilong Huang, Weixian Lei, Yunqing Zhao, Yujie Zhong, Yingchen Yu, Qi She, Yao Zhao, Yunchao Wei Published: 2026-05-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00809v2 In this paper, we present \textbf{Gen}erative \textbf{L}anguage-\textbf{I}mage \textbf{P}re-training (GenLIP), a minimalist generative pretraining framework for Vision Transformers (ViTs) designed for multimodal large language models (MLLMs). To better align vision encoders with the autoregressive nature of LLMs, GenLIP trains a ViT to predict language tokens directly from visual tokens using a standard language modeling objective, without contrastive batch construction or an additional text decoder. This design offers three key advantages: (1) \textbf{Simplicity}: a single transformer jointly models visual and textual tokens; (2) \textbf{Scalability}: it scales effectively with both data and model size; and (3) \textbf{Performance}: it achieves competitive or superior results across diverse multimodal benchmarks. Trained on 8B samples from Recap-DataComp-1B, GenLIP matches or surpasses strong baselines despite using substantially less pretraining data. After continued pretraining on multi-resolution images at native aspect ratios, GenLIP further improves on detail-sensitive tasks such as OCR and chart understanding, making it a strong foundation for vision encoders in MLLMs. --- ### Can Coding Agents Reproduce Findings in Computational Materials Science? Authors: Ziyang Huang, Yi Cao, Ali K. Shargh, Jing Luo, Ruidong Mei, Mohd Zaki, Zhan Liu, Wyatt Bunstine, William Jurayj, Somdatta Goswami, Tyrel McQueen, Michael Shields et al. Published: 2026-05-01 Categories: cs.CL, cond-mat.mtrl-sci, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2605.00803v2 Large language models are increasingly deployed as autonomous coding agents and have achieved remarkably strong performance on software engineering benchmarks. However, it is unclear whether such success transfers to computational scientific workflows, where tasks require not only strong coding ability, but also the ability to navigate complex, domain-specific procedures and to interpret results in the context of scientific claims. To address this question, we present AutoMat, a benchmark for evaluating LLM-based agents' ability to reproduce claims from computational materials science. AutoMat poses three interrelated challenges: recovering underspecified computational procedures, navigating specialized toolchains, and determining whether the resulting evidence supports a claim. By working closely with subject matter experts, we curate a set of claims from real materials science papers to test whether coding agents can recover and execute the end-to-end workflow needed to support (or undermine) such claims. We then evaluate multiple representative coding agent settings across several foundation models. Our results show that current LLM-based agents obtain low overall success rates on AutoMat, with the best-performing setting achieving a success rate of only 53%. Error analysis further reveals that agents perform worst when workflows must be reconstructed from paper text alone and that they fail primarily due to incomplete procedures, methodological deviations, and execution fragility. Taken together, these findings position AutoMat as both a benchmark for computational scientific reproducibility and a tool for diagnosing the current limitations of agentic systems in AI-for-science settings. --- ### Can Coding Agents Reproduce Findings in Computational Materials Science? Authors: Ziyang Huang, Yi Cao, Ali K. Shargh, Jing Luo, Ruidong Mei, Mohd Zaki, Zhan Liu, Wyatt Bunstine, William Jurayj, Somdatta Goswami, Tyrel McQueen, Michael Shields et al. Published: 2026-05-01 Categories: cs.SE, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.00803v1 Large language models are increasingly deployed as autonomous coding agents and have achieved remarkably strong performance on software engineering benchmarks. However, it is unclear whether such success transfers to computational scientific workflows, where tasks require not only strong coding ability, but also the ability to navigate complex, domain-specific procedures and to interpret results in the context of scientific claims. To address this question, we present AutoMat, a benchmark for evaluating LLM-based agents' ability to reproduce claims from computational materials science. AutoMat poses three interrelated challenges: recovering underspecified computational procedures, navigating specialized toolchains, and determining whether the resulting evidence supports a claim. By working closely with subject matter experts, we curate a set of claims from real materials science papers to test whether coding agents can recover and execute the end-to-end workflow needed to support (or undermine) such claims. We then evaluate multiple representative coding agent settings across several foundation models. Our results show that current LLM-based agents obtain low overall success rates on AutoMat, with the best-performing setting achieving a success rate of only 54.1%. Error analysis further reveals that agents perform worst when workflows must be reconstructed from paper text alone and that they fail primarily due to incomplete procedures, methodological deviations, and execution fragility. Taken together, these findings position AutoMat as both a benchmark for computational scientific reproducibility and a tool for diagnosing the current limitations of agentic systems in AI-for-science settings. --- ### Generating Statistical Charts with Validation-Driven LLM Workflows Authors: Pavlin G. Poličar, Andraž Pevcin, Blaž Zupan Published: 2026-05-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.00800v1 Generating diverse, readable statistical charts from tabular data remains challenging for LLMs, as many failures become apparent after rendering and are not detectable from data or code alone. Existing chart datasets also rarely provide fully aligned artifacts, such as executable code, dataset context, and question-answer pairs. We present a structured LLM-based workflow that decomposes chart generation into dataset screening, plot proposal, code synthesis, rendering, validation-driven refinement, description generation, and question-answer generation. By incorporating rendered-output validation, the workflow addresses visualization-specific failure modes such as readability and semantic mismatch. It treats chart generation as an inspectable process rather than a one-shot prompt-to-code task, retaining each chart with its code, dataset context, description, and question-answer pairs. Applied to UCI datasets, the workflow produces 1,500 charts from 74 datasets, spanning 24 chart families and paired with 30,003 question-answer pairs. We evaluate 16 multimodal LLMs (MLLMs) on these chart-question pairs. The results show that chart-syntax questions are nearly saturated, while value extraction, comparison, and reasoning remain more challenging, illustrating the workflow's utility for diagnostic studies of chart-grounded multimodal reasoning. --- ### GMGaze: MoE-Based Context-Aware Gaze Estimation with CLIP and Multiscale Transformer Authors: Xinyuan Zhao, Yihang Wu, Ahmad Chaddad, Sarah A. Alkhodair, Reem Kateb Published: 2026-05-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00799v1 Gaze estimation methods commonly use facial appearances to predict the direction of a person gaze. However, previous studies show three major challenges with convolutional neural network (CNN)-based, transformer-based, and contrastive language-image pre-training (CLIP)-based methods, including late fusion of image features, lack of factor-aware conditioning, and impractical capacity scaling. To address these challenges, we propose Globally-conditioned Multi-scale Gaze estimation (GMGaze), which leverages a multi-scale transformer architecture. Specifically, the model first introduces semantic prototype conditioning, which modulates the CLIP global image embedding using four learned prototype banks (i.e., illumination, background, head pose and appearance) to generate two complementary context-biased global tokens. These tokens, along with the CLIP patch and CNN tokens, are fused at the first layer. This early unified fusion prevents information loss common in late-stage merging. Finally, each token passes through sparse Mixture-of-Experts modules, providing conditional computational capacity without uniformly increasing dense parameters. For cross-domain adaptation, we incorporate an adversarial domain adaptation technique with a feature separation loss that encourages the two global tokens to remain de-correlated. Experiments using four public benchmarks (MPIIFaceGaze, EYEDIAP, Gaze360, and ETH-XGaze) show that GMGaze achieves mean angular errors of 2.49$^\circ$, 3.22$^\circ$, 10.16$^\circ$, and 1.44$^\circ$, respectively, outperforming previous baselines in all within-domain settings. In cross-domain evaluations, it provides state-of-the-art (SOTA) results on two standard transfer routes. --- ### RunAgent: Interpreting Natural-Language Plans with Constraint-Guided Execution Authors: Arunabh Srivastava, Mohammad A., Khojastepour, Srimat Chakradhar, Sennur Ulukus Published: 2026-05-01 Categories: cs.LG, cs.CL, cs.MA arXiv: https://arxiv.org/abs/2605.00798v1 Humans solve problems by executing targeted plans, yet large language models (LLMs) remain unreliable for structured workflow execution. We propose RunAgent, a multi-agent plan execution platform that interprets natural-language plans while enforcing stepwise execution through constraints and rubrics. RunAgent bridges the expressiveness of natural language with the determinism of programming via an agentic language with explicit control constructs (e.g., \texttt{IF}, \texttt{GOTO}, \texttt{FORALL}). Beyond verifying syntactic and semantic verification of the step output, which is performed based on the specific instruction of each step, RunAgent autonomously derives and validates constraints based on the description of the task and its instance at each step. RunAgent also dynamically selects among LLM-based reasoning, tool usage, and code generation and execution (e.g., in Python), and incorporates error correction mechanisms to ensure correctness. Finally, RunAgent filters the context history by retaining only relevant information during the execution of each step. Evaluations on Natural-plan and SciBench Datasets demonstrate that RunAgent outperforms baseline LLMs and state-of-the-art PlanGEN methods. --- ### When RAG Chatbots Expose Their Backend: An Anonymized Case Study of Privacy and Security Risks in Patient-Facing Medical AI Authors: Alfredo Madrid-García, Miguel Rujas Published: 2026-05-01 Categories: cs.CR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.00796v1 Background: Patient-facing medical chatbots based on retrieval-augmented generation (RAG) are increasingly promoted to deliver accessible, grounded health information. AI-assisted development lowers the barrier to building them, but they still demand rigorous security, privacy, and governance controls. Objective: To report an anonymized, non-destructive security assessment of a publicly accessible patient-facing medical RAG chatbot and identify governance lessons for safe deployment of generative AI in health. Methods: We used a two-stage strategy. First, Claude Opus 4.6 supported exploratory prompt-based testing and structured vulnerability hypotheses. Second, candidate findings were manually verified using Chrome Developer Tools, inspecting browser-visible network traffic, payloads, API schemas, configuration objects, and stored interaction data. Results: The LLM-assisted phase identified a critical vulnerability: sensitive system and RAG configuration appeared exposed through client-server communication rather than restricted server-side. Manual verification confirmed that ordinary browser inspection allowed collection of the system prompt, model and embedding configuration, retrieval parameters, backend endpoints, API schema, document and chunk metadata, knowledge-base content, and the 1,000 most recent patient-chatbot conversations. The deployment also contradicted its privacy assurances: full conversation records, including health-related queries, were retrievable without authentication. Conclusions: Serious privacy and security failures in patient-facing RAG chatbots can be identified with standard browser tools, without specialist skills or authentication; independent review should be a prerequisite for deployment. Commercial LLMs accelerated this assessment, including under a false developer persona; assistance available to auditors is equally available to adversaries. --- ### Unsupervised Denoising of Real Clinical Low Dose Liver CT with Perceptual Attention Networks Authors: Zhilin Guan, Wei Zhang Published: 2026-05-01 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2605.00793v2 With the development of deep learning, medical image processing has been widely used to assist clinical research. This paper focuses on the denoising problem of low-dose computed tomography using deep learning. Although low-dose computed tomography reduces radiation exposure to patients, it also introduces more noise, which may interfere with visual interpretation by physicians and affect diagnostic results. To address this problem, inspired by Cycle-GAN for unsupervised learning, this paper proposes an end-to-end unsupervised low-dose computed tomography denoising framework. The proposed framework combines a U-Net structure for multi-scale feature extraction, an attention mechanism for feature fusion, and a residual network for feature transformation. It also introduces perceptual loss to improve the network for the characteristics of medical images. In addition, we construct a real low-dose computed tomography dataset and design a large number of comparative experiments to validate the proposed method, using both image-based evaluation metrics and medical evaluation criteria. Compared with classical methods, the main advantage of this paper is that it addresses the limitation that real clinical data cannot be directly used for supervised learning, while still achieving excellent performance. The experimental results are also professionally evaluated by imaging physicians and meet clinical needs. --- ### Make Your LVLM KV Cache More Lightweight Authors: Xihao Chen, Yangyang Guo, Roger Zimmermann Published: 2026-05-01 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.00789v1 Key-Value (KV) cache has become a de facto component of modern Large Vision-Language Models (LVLMs) for inference. While it enhances decoding efficiency in Large Language Models (LLMs), its direct adoption in LVLMs introduces substantial GPU memory overhead due to the large number of vision tokens processed during the prefill stage. To tackle this problem, we propose LightKV, a novel approach that reduces KV cache size by exploiting the redundancy among vision-token embeddings. Guided by text prompts, LightKV employs cross-modality message passing to aggregate informative messages across vision tokens and progressively compress them during prefill. This prompt-aware guidance distinguishes our method from prior vision-only compression strategies. We evaluate LightKV on eight open-source LVLMs across eight public benchmark datasets, e.g., MME and SeedBench. Experimental results demonstrate that with only 55% of the original vision tokens, LightKV (a) halves the vision-token KV cache size, (b) reduces computation by up to 40%, and (c) preserves general-purpose performance while significantly outperforming existing baselines. ---