# Research Papers — 2026-05 · bundle page 266 of 326 > Full-text markdown of papers #13251–13300 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/266 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/265.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/267.md) ### Distributional Spectral Diagnostics for Localizing Grokking Transitions Authors: Ziyue Wang, Yufeng Ying, Takafumi Kanamori Published: 2026-05-07 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2605.08237v1 In grokking, a model first fits the training data while test accuracy remains low, and only later begins to generalize. We ask whether this transition can be localized from observed training trajectories before the test accuracy rises, and formulate grokking transition localization as a diagnostic problem with an explicit threshold/FPR/lead-time trade-off. Task-dependent observables are summarized as empirical distributions, mapped to Wasserstein/quantile coordinates, and analyzed by Hankel dynamic mode decomposition (DMD); the resulting reconstruction residual, together with spectrum and effective rank, forms the diagnostic output. On held-out modular-addition Transformer runs, the residual achieves AUROC \(\approx \) 0.93 for grokking-vs-non-grokking discrimination at the run level; under a fixed sustained-threshold operating rule, true-positive alarms can precede onset, with lead time reported jointly with false-alarm rate and uncertainty intervals. Perturbation experiments show that, in the tested \(wd=1\) pool, high-residual windows exhibit about \(3\times\) larger short-horizon perturbation deviation than low-residual windows. In a same-data norm-window control, perturbation sensitivity aligns with the residual ordering rather than total-parameter-norm ordering, suggesting that the residual is not merely a total-norm proxy at the window level in the studied \(wd=1\) dynamics. Norm signals remain strong run-level regime indicators, and log-probability performs best among the observables tested under the current protocol. We position the residual as a window-level monitoring and localization signal in the studied modular-arithmetic Transformer settings, not a universal early-warning predictor or an intervention rule. --- ### When Does Value-Aware KV Eviction Help? A Fixed-Contract Diagnostic for Non-Monotone Cache Compression Authors: Ruijie Zhang, Haozhe Liang, Da Chang, Li Hu, Fanqi Kong, Huaxiao Yin, Yu Li Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08234v1 Long-context LLM inference is bottlenecked by the memory and bandwidth cost of reading large KV caches during decoding. KV compression reduces this cost by keeping only part of the cache, but task accuracy alone does not identify why a selector succeeds or fails. A selector can fail at three steps: it may miss the evidence future decoding needs, give high scores to tokens that do not affect the output, or break related evidence when fitting scores into a small cache. We introduce a fixed-contract diagnostic that holds the selector's setup fixed and changes one decision slot at a time. For value ranking, the probe combines a block's attention mass with the estimated output change from removing it. On LongBench across three models and two budgets, the probe is positive on 72.6% of positive-margin cells and 32.4% of nonpositive-margin cells. NeedleBench M-RT at 32k and a RULER 8k check probe support closure under branched retrieval, and a 264-cell sign evaluation separates support recovery and output-value ranking from leverage effects near the boundary. The resulting order is to recover decode-side evidence, rank its output value, and preserve coupled evidence during projection. --- ### Towards Differentially Private Reinforcement Learning with General Function Approximation Authors: Yi He, Xingyu Zhou Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.07049v1 We present the first theoretical guarantees for differentially private online reinforcement learning (RL) with general function approximation, extending beyond prior work restricted to tabular and linear settings. Our approach combines a batched policy update scheme with the exponential mechanism, together with a novel regret analysis. We show that, even under general function approximation, the regret in the model-free setting under differential privacy matches the state of the art for the linear case, scaling as $\widetilde{O}(K^{3/5})$, where $K$ denotes the number of episodes. As an important by-product, we also establish the first regret bound for online RL with batch update that depends on the standard complexity measure of coverability, complementing existing results based on a newly introduced Eluder-Condition class. In addition, we uncover fundamental gaps in recent results for private RL with linear function approximation, thereby clarifying its landscape. --- ### Unlocking High-Fidelity Molecular Generation from Mass Spectra via Dual-Stream Line Graph Diffusion Authors: Xujun Che, Xiuxia Du, Depeng Xu Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.07048v1 De novo molecular generation from tandem mass spectra is a challenging inverse problem whose core difficulty lies in the circular dependency between atom-level and bond-level reasoning: determining a bond's type requires knowing its endpoint atoms' chemical environment, yet an atom's environment is in turn defined by its incident bonds. Existing graph diffusion methods process atoms and bonds within a single computation stream, where atom-bond information synchronization can only occur implicitly across layers. We argue that this single-stream paradigm, rather than the choice of any particular aggregation kernel, is a key architectural bottleneck. We propose DualLGD (Dual-stream Line Graph Diffusion), which reformulates molecular graph denoising as the alternating solution of two coupled subproblems: atom-level reasoning and bond-level reasoning, each operating in its own dedicated representation space. The line graph provides a natural mathematical construction for the bond space, in which bond angles, dihedrals, conjugation chains, and rings correspond to local topological motifs between bonds. Incidence-constrained bidirectional cross-attention synchronizes the two streams at every layer, ensuring that each atom attends only to its incident bonds and vice versa, respecting the fundamental chemical principle that an atom's environment is determined by its bonding context. On the NPLIB1 and MassSpecGym benchmarks, DualLGD achieves top-1 accuracy of 34.37\% and 23.89\%, approximately $3\times$ the previous state of the art. Ablation studies confirm the architecture as the primary source of improvement: DualLGD without any pre-training already surpasses the previous best fully pretrained model. --- ### An Interpretable and Scalable Framework for Evaluating Large Language Models Authors: Xinhao Qu, Qiang Heng, Hao Zeng, Xiaoqian Liu Published: 2026-05-07 Categories: stat.ML, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.07046v1 Evaluation of large language models (LLMs) is increasingly critical, yet standard benchmarking methods rely on average accuracy, overlooking both the inherent stochasticity of LLM outputs and the heterogeneity of benchmark items. Item Response Theory (IRT) offers a principled framework for modeling latent model abilities and item characteristics, but conventional methods are computationally expensive and numerically unstable, limiting large-scale implementations. To address these challenges, we propose an interpretable and scalable framework for LLM evaluation based on the majorization-minimization principle. Our approach reformulates the problem as a sequence of constrained matrix factorization subproblems, enabling stable and efficient parameter estimation with theoretical guarantees for identifiability and convergence. Experiments on synthetic and real-world datasets, including MATH-500 and six Open LLM Leaderboard benchmarks, demonstrate that our method achieves superior scalability and interpretability. It delivers orders-of-magnitude speedups over competing methods while maintaining comparable or even higher estimation accuracy. Our results align with established scaling laws and offer insights into item difficulty and discrimination, informing more principled benchmark design. --- ### The Context Gathering Decision Process: A POMDP Framework for Agentic Search Authors: Chinmaya Kausik, Adith Swaminathan, Nathan Kallus Published: 2026-05-07 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.07042v1 Large Language Model (LLM) agents are deployed in complex environments -- such as massive codebases, enterprise databases, and conversational histories -- where the relevant state far exceeds their context windows. To navigate these spaces, an agent must iteratively explore the environment to find relevant information. However, without explicit infrastructure, an agent's working memory can degrade into lossy representations of the search state, resulting in redundant work (e.g. repetitive looping) and premature stopping. In this work, we formalize this challenge as the Context Gathering Decision Process (CGDP), a specialized Partially Observable Markov Decision Process, where an agent's objective is to adaptively refine its belief state to isolate the necessary information for a task. We model an LLM's behavior as approximate Thompson Sampling within this CGDP, and introduce a predicate-based method that decomposes an LLM's implicit search into explicit and modular operations. We then derive two plug-and-play interventions for iterative LLM agents: a persistent, predicate-based belief state that bounds context while preserving multi-hop reasoning, and a programmatic exhaustion gate that halts unproductive search without premature stopping. Across four methods and three question-answering domains, we empirically validate that replacing an LLM's implicit state with our CGDP-motivated belief state improves multi-hop reasoning by up to $11.4\%$; while the modular programmatic exhaustion detection saves up to $39\%$ of tokens without any degradation in agent performance. Ultimately, we argue that framing the LLM agent loop as a CGDP can guide the design of modular, non-interfering improvements to agentic search harnesses. --- ### Dr-BA: Separable Optimization for Direct Radar Bundle Adjustment & Localization Authors: Daniil Lisus, Cedric Le Gentil, Timothy D. Barfoot Published: 2026-05-07 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2605.07041v1 This paper introduces Dr-BA, a first-of-its-kind radar bundle adjustment (BA) framework that operates directly on 2D spinning radar intensity images. Unlike camera or lidar sensors, radar is largely unaffected by precipitation, making it a critical modality for autonomous systems that require all-weather robustness. Existing state estimation approaches using spinning radar typically extract sparse point clouds from range-azimuth-intensity measurements and apply point cloud alignment techniques to estimate vehicle motion, scene structure, or to localize within an existing map. In contrast, Dr-BA uses the full radar returns from multiple scans to jointly estimate dense maps and sensor poses. By formulating the problem as a separable optimization, we derive an efficient and general solution that decouples pose estimation from mapping. In addition to solving the BA problem, this formulation naturally extends to direct radar-only localization (DRL) within a previously built map. Dr-BA achieves state-of-the-art radar-based BA and cross-session localization performance, demonstrated on more than 200 km of on-road data across five distinct routes. Our implementation is publicly available at https://github.com/utiasASRL/dr_ba. --- ### Cognitive Agent Compilation for Explicit Problem Solver Modeling Authors: Hyeongdon Moon, Carolyn Rosé, John Stamper Published: 2026-05-07 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2605.07040v1 Large language models (LLMs) are widely used for tutoring, feedback generation, and content creation, but their broad pretraining makes them hard to constrain and poor substitutes for controllable learners. Educational systems often require inspectable and editable knowledge states: educators want to know what a system assumes the learner knows, and learners benefit when the system can justify actions in terms of explicit skills, misconceptions, and strategies. Inspired by cognitive architectures, we propose Cognitive Agent Compilation (CAC), a framework that uses a strong teacher LLM to compile problem-solving knowledge into an explicit target agent. CAC separates (i) knowledge representation, (ii) problem-solving policy, and (iii) verification and update rules, with the goal of making bounded problem solving more inspectable and editable in educational settings. We present an early proof of concept implemented with Small Language Models that surfaces key design trade-offs, particularly between explicit control and scalable generalization, and positions CAC as an initial step toward bounded-knowledge AI for educational applications. --- ### PACEvolve++: Improving Test-time Learning for Evolutionary Search Agents Authors: Minghao Yan, Bo Peng, Benjamin Coleman, Ziqi Chen, Zhouhang Xie, Shuo Chen, Zhankui He, Noveen Sachdeva, Weili Wang, Ed H. Chi, Shivaram Venkataraman, Wang-Cheng Kang et al. Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.07039v1 Large language models have become drivers of evolutionary search, but most systems rely on a fixed, prompt-elicited policy to sample next candidates. This limits adaptation in practical engineering and research tasks, where evaluations are expensive, and progress depends on learning task-specific search dynamics. We introduce PACEvolve++, an advisor-model reinforcement learning framework for test-time policy adaptation in evolutionary search agents. PACEvolve++ decouples strategic search decisions from implementation: a trainable advisor generates, assesses, and selects hypotheses, while a stronger frontier model translates selected hypotheses into executable candidates. To train the advisor under non-stationary feedback, we propose a phase-adaptive approach that adapts its optimization strategy to different phases of the evolutionary process. Early in evolution, it uses group-relative feedback to learn broad search preferences; later, as reward gaps compress, it emphasizes best-of-$k$ frontier contribution to support stable refinement. Across expert-parallel load balancing, sequential recommendation, and protein fitness extrapolation, PACEvolve++ outperforms the state-of-the-art evolutionary search framework with frontier models, achieving faster convergence and stabilizing test-time training during evolutionary search. --- ### Learning Material-Aware Hamiltonian Risk Fields for Safe Navigation Authors: Aditya Sai Ellendula, Yi Wang, Chandrajit Bajaj Published: 2026-05-07 Categories: cs.LG, cs.MA, cs.RO arXiv: https://arxiv.org/abs/2605.07038v1 Risk-aware navigation should be selective: a policy should expose evasive degrees of freedom only when the local scene admits a lower-risk feasible maneuver, and suppress them when no safer alternative exists. We show that adding one context-energy term to a port-Hamiltonian navigation policy produces a learned force channel with exactly this falsifiable signature. When the local risk field contains a feasible lower-risk direction, the induced context force activates toward it; when the apparent escape is blocked or not yet available, a route-aware gate suppresses lateral force rather than hallucinating an unsafe maneuver. A CVaR tail-risk objective focuses gradient updates on rare but consequential risk transitions. We validate the selectivity signature across four settings. In the primary delayed-required-escape benchmark, route-aware CVaR reduces premature force activation from 0.950 to 0.180 versus DWA while raising success from 0.480 to 0.810 with zero replans. On real off-road terrain (RELLIS-3D), route-aware enrichment achieves correct activation rate 0.837 and false activation rate 0.114, compared to 0.378/0.752 for scalar risk gradients. On static semantic maps (DFC2018), enrichment reduces catastrophic failure from 0.60 to 0.10 and oscillation by 90.7% while preserving path efficiency. In highway traffic, collisions drop from 100% to 0% when a lane escape is feasible; when no escape exists, the policy suppresses the lateral maneuver. The selectivity property follows from the gradient structure of the context energy rather than from training-time tuning. --- ### Beyond the Wrapper: Identifying Artifact Reliance in Static Malware Classifiers using TRUSTEE Authors: Riyazuddin Mohammed, Lan Zhang Published: 2026-05-07 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2605.07034v1 Modern cybersecurity relies heavily on static machine-learning-based malware classifiers. However, transformations such as packing and other non-semantic modifications applied to executable files limit their reliability. Malware classifiers often learn these unnecessary artifacts rather than the true binary behavior because of the high association between maliciousness and packing. Moreover, these malware classifiers are black boxes, making it difficult to understand what they learn. To address this issue, we proposed a two-part framework using the post-hoc interpretability XAI tool TRUSTEE, followed by a manual analysis of the top features. We conducted several controlled experiments by varying the dataset composition ratios to understand their impact on the results. The top-ranked features across all experiments, identified by TRUSTEE, were predominantly packing artifacts, portable executable(PE) metadata, and n-grams at the string level, rather than malicious semantics. These results suggest that these malware classifiers are highly sensitive to dataset composition and can misinterpret packing as malicious behavior. Our proposed framework allows for the reproducible diagnosis of such biases and forms a guideline for building more robust and semantically meaningful malware detection models --- ### A Systematic Investigation of RL-Jailbreaking in LLMs Authors: Montaser Mohammedalamen, Kevin Roice, Reginald McLean, Alyssa Lefaivre Škopac Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.07032v3 The evolution of generative models from next-token predictors to autonomous engines of complex systems necessitates rigorous safety hardening. Adversarial jailbreaking, the strategic manipulation of models to elicit harmful output, remains a primary threat to safe deployment. While Reinforcement Learning (RL) frames jailbreaking as a multi-step attack through sequential optimization, a mechanistic understanding of why the framework succeeds remains incomplete. To fill this gap, we present the first systematic decomposition of RL jailbreaking. We deconstruct the framework into problem formalization (reward function, action space, episode length), and algorithmic measures (RL algorithm, training data, reward-shaping) to identify the structural determinants of adversarial success. Our results reveal that the RL-jailbreaker successfully compromised all targeted models and safeguards. Through this first-of-its-kind analysis, we demonstrate that environment formalization, specifically dense rewards and extended episode lengths, is the primary driver of jailbreaking success. This work provides a tool for improving RL-jailbreaker efficiency and, ultimately, harden generative models resistant to RL-based attacks. --- ### BGM-IV: AI-Powered Bayesian Generative Modeling for Instrumental Variable Regression with High-Dimensional Covariates Authors: Guyue Luo, Haidong Lu, Andrew J. Loza, Qiao Liu Published: 2026-05-07 Categories: stat.ML, cs.AI, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2605.07029v2 Instrumental-variable (IV) regression enables causal estimation under endogeneity, but modern IV problems often involve nonlinear structural effects and high-dimensional covariates. Existing methods typically operate in observed or generic learned feature spaces, and they often yield point estimates without uncertainty quantification. We introduce BGM-IV, a Bayesian generative modeling approach that performs nonlinear IV regression through posterior inference in a causally structured latent space. BGM-IV separates covariate variation by the role in the treatment and outcome mechanism, and accounts for endogeneity through an IV-integrated pseudo-likelihood that averages over instrument-induced treatment variation. The resulting model provides both structural-function estimates and predictive intervals for outcomes under intervention. Across various benchmark datasets, BGM-IV outperforms existing nonlinear IV methods overall, with significant gains in high-dimensional settings, while achieving near-nominal predictive coverage. These results highlight structured latent generative modeling as a flexible approach to uncertainty-aware IV inference with rich covariates. The code of BGM-IV is available at https://github.com/liuq-lab/BGM-IV. --- ### BGM-IV: an AI-powered Bayesian generative modeling approach for instrumental variable analysis Authors: Guyue Luo, Qiao Liu Published: 2026-05-07 Categories: stat.ML, cs.AI, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2605.07029v1 Instrumental-variable (IV) regression enables causal estimation under endogeneity, but modern IV problems often involve nonlinear structural effects and high-dimensional covariates. Existing nonlinear IV methods directly learn the causal relation in observed feature space or rely on learned representations within two-stage or moment-based procedures, which can struggle when the causal information is embedded in a high-dimensional representation. We propose BGM-IV, a latent Bayesian generative modeling approach that reframes nonlinear IV regression as posterior inference in a causally structured latent space. BGM-IV infers latent components that separately capture shared confounding structure, outcome-specific variation, treatment-specific variation, and covariate-only nuisance information. To account for endogeneity, BGM-IV replaces the confounded outcome likelihood with an IV-integrated pseudo-likelihood that averages over instrument-induced treatment values within the latent model. Across various benchmark datasets, BGM-IV remains competitive in the classical low-dimensional regime and performs best in high-dimensional covariate regimes. Together, these results show that structured latent generative modeling provides a principled and effective strategy to nonlinear IV estimation with rich covariates. The code of BGM-IV is available at https://github.com/liuq-lab/BGM-IV. --- ### Learning Cross-Atlas Consistent Brain Disorder Representations via Disentangled Multi-Atlas Functional Connectivity Learning Authors: Minheng Chen, Chao Cao, Jing Zhang, Tianming Liu, Dajiang Zhu Published: 2026-05-07 Categories: q-bio.NC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.07026v1 Functional connectivity (FC) derived from resting-state fMRI is widely used to characterize large-scale brain network alterations in neurological and psychiatric disorders. However, FC construction critically depends on the choice of brain atlas, and different parcellations may emphasize distinct organizational features, leading to heterogeneous and sometimes inconsistent representations. Existing multi-atlas approaches partially alleviate this issue but often fuse atlas-derived features or predictions at a relatively shallow level, while single-atlas disentanglement methods do not explicitly address cross-atlas heterogeneity. We propose Multi-Atlas Disentangled Connectivity LEarning (MADCLE), a multi-branch representation learning framework that jointly encodes FC matrices derived from different brain atlases. Rather than introducing a single explicitly shared latent variable across parcellations, MADCLE learns atlas-wise disease-related representations and encourages them to be cross-atlas consistent through distributional alignment. Meanwhile, covariate-related and atlas-dependent residual factors are modeled separately using covariate similarity supervision, atlas-specific reconstruction, and decorrelation constraints, thereby reducing the leakage of non-disease and parcellation-dependent information into the disease-related embeddings. Experiments on the ADNI and ADHD-200 datasets suggest that MADCLE achieves competitive or improved performance compared with single-atlas baselines, multi-atlas GNN/Transformer models, and recent multi-atlas consistency frameworks. These results support the potential value of structured disentanglement for FC-based disorder identification under heterogeneous parcellation schemes. --- ### Delulu: A Verified Multi-Lingual Benchmark for Code Hallucination Detection in Fill-in-the-Middle Tasks Authors: Mahdi Erfanian, Nelson Daniel Troncoso, Aashna Garg, Amabel Gale, Xiaoyu Liu, Pareesa Ameneh Golnari, Shengyu Fu Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.07024v1 Large Language Models for code generation frequently produce hallucinations in Fill-in-the-Middle (FIM) tasks -- plausible but incorrect completions such as invented API methods, invalid parameters, undefined variables, or non-existent imports. These failures pass superficial review yet introduce runtime errors. We introduce Delulu, a verified multi-lingual benchmark of 1,951 FIM samples across 7 languages and 4 hallucination types. Samples are curated through an adversarial pipeline: a frontier LLM generates plausible hallucinations, four diverse judge models evaluate them, embedding-based clustering mines progressively harder examples, self-contained Docker containers verify that golden completions compile while hallucinated variants produce the expected runtime error, and a final human-expert review removes any remaining biased or trivially decidable samples. We evaluate 11 open-weight FIM models from five families spanning 0.5B-32B parameters: a six-point Qwen2.5-Coder scaling slate, plus a cross-family slate (CodeLlama, DeepSeek-Coder-V2, StarCoder2). The strongest model reaches only 84.5% pass@1, no family exceeds 0.77 Edit Similarity, and every family produces hallucination-aligned completions on a non-trivial share of samples, confirming that the difficulty exposed by Delulu is task-intrinsic rather than family-specific. We release the benchmark, containers, and evaluation framework at https://github.com/microsoft/delulu. --- ### Semantic Prior Guided One-View 6D Pose Estimation for Novel Objects Authors: Yang Luo, Yan Gong, Yongsheng Gao Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.07023v2 In many practical 6D object pose estimation scenarios, we often have access to only a single real-world RGB-D reference view per object, typically without CAD models. Existing methods largely rely on explicit 3D models or multi-view data, which limits their scalability. To address this challenging single-reference model-free setting, we propose \textbf{OneViewAll}, a semantic-prior-guided framework that performs pose estimation via a novel Project-and-Compare paradigm. Instead of relying on computationally expensive CAD-based rendering, our method directly aligns reference and query observations within a projection-equivariant space. OneViewAll progressively integrates hierarchical semantic priors across three levels: (1) \textit{category- and scene-level} priors for efficient hypothesis initialization; (2) \textit{object-level symmetry} priors for geometry completion via mirror fusion; and (3) \textit{patch-level} priors for discriminative refinement. Extensive experiments demonstrate that OneViewAll achieves \textbf{92.5\%} ADD-0.1 accuracy on the LINEMOD dataset using only one real reference view -- significantly outperforming the CVPR 2025 baseline One2Any (52.6\%). It also yields consistent improvements on YCB-V, Real275, and Toyota-Light while maintaining low inference latency. Our results underscore the efficacy of symmetry-aware projection in handling symmetric, texture-less, and occluded objects. --- ### Self-Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale Authors: Haydn Jones, Yimeng Zeng, Alden Rose, Li S. Yifei, Yining Huang, Kaiwen Wu, Jiaming Liang, Maggie Ziyu Huan, Yoseph Barash, Cesar de la Fuente-Nunez, Osbert Bastani, Zachary Ives et al. Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.07022v3 Manually curated biomedical repositories -- spanning bioactivity, genomics, and chemistry -- are expensive to maintain, lag behind primary literature, and discard experimental context, obscuring nuances needed to assess data correctness and coverage. We show that PubMed itself can be autonomously and cost-effectively turned into structured datasets that are larger, more nuanced, and more accurate than the curated databases they replace. We present three coupled contributions: (1) an LLM-based entity-tagging pipeline, grounded in nine biomedical ontologies, that tags 4.5B entities across 19 categories in a 22.5M-paper, 2.5T-token PubMed corpus; (2) hybrid sparse-dense retrieval supporting entity-filtered semantic queries over the tagged corpus; and (3) Starling, a multi-agent deep research system that, given only a natural-language task description, designs precision- and recall-targeted retrieval filters, induces an extraction schema, and emits structured records with nuance-rich fields and supporting passages. Across six tasks -- blood-brain barrier permeability, oral bioavailability, acute toxicity (LD50), gene-disease associations, protein subcellular localization, and chemical reactions -- Starling produces ~6.3M records (91K-3M per task); several are, to our knowledge, the largest public datasets for their property. Frontier-model rejection of our extractions is 0.6-7.7% across tasks, far below error rates we measure on widely used curated counterparts (e.g., 16.5% on BBB_Martins, 7.3% on Bioavailability_Ma). Beyond scale and accuracy, the supporting passages carry nuance tabular databases discard -- e.g., oral bioavailability may depend on fed vs. fasted state. Together, the corpus, retrieval, and agent establish a foundation for AI-driven therapeutic design. Code and datasets: https://github.com/starling-labs/starling. --- ### Behavior Cue Reasoning: Monitorable Reasoning Improves Efficiency and Safety through Oversight Authors: Christopher Z. Cui, Taylor W. Killian, Prithviraj Ammanabrolu Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.07021v2 Reasoning in Large Language Models (LLMs) poses a challenge for oversight as many misaligned behaviors do not surface until reasoning concludes. To address this, we introduce Behavior Cue Reasoning for making LLM reasoning more controllable and monitorable. Behavior Cues are special token sequences that a model is trained to emit immediately before specific implicit and explicit behaviors, acting as dual purpose signal and control levers. When fine-tuning a weaker external monitor with Reinforcement Learning for reasoning oversight, a compressed view of only information surfaced by Behavior Cues is sufficient signal for the monitor to prune up to 50% of otherwise wasted reasoning tokens in complex math problem solving. When leveraged by an almost optimal rule-based monitor in an environment where excessive constraint violations results in failure, Behavior Cues allows for the recovery of safe actions from 80% of reasoning traces that would otherwise end with the proposal of an unsafe action, more than doubling the success rate from 46% to 96%. Through evaluation across two model families and three domains, we show that Behavior Cue Reasoning improves reasoning monitorability and controllability with no cost to performance. More broadly, our work progresses scalable oversight by demonstrating how the monitored model itself can be trained to reason more tractably to oversight. Code: https://github.com/christopherzc/behavior-cues --- ### FlashMol: High-Quality Molecule Generation in as Few as Four Steps Authors: Xinyuan Wei, Zian Li, Shaoheng Yan, Cai Zhou, Muhan Zhang Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.07020v1 Generating chemically valid 3D molecular conformations is critical for computational drug discovery. Classical diffusion-based models like GeoLDM perform well but require hundreds of steps, making large-scale in silico screening impractical. Recent efforts on few-step molecular generation have accelerated this process to 12-50 steps, but they often largely sacrifice sample stability. In this work, we present FlashMol, an ultra-fast molecule generative model producing high-quality molecular conformations in as few as 4 steps. To achieve this, we adapt distribution matching distillation (DMD) - a reverse KL-divergence minimization objective - to the molecular domain for effective distillation. Considering the local minimization behavior of DMD, we respace the molecule generation timesteps, providing the generator with much better initialization and enables effective distillation. Additionally, to mitigate the mode-seeking behavior of DMD and improve diversity, we further regularize it with a Jensen-Shannon divergence term, which incorporates the mean-seeking behavior of the forward KL divergence. Extensive experiments on QM9 and GEOM-DRUG datasets demonstrate that FlashMol matches and even surpasses the original 1000-step teacher, achieving up to 250$\times$ acceleration in sampling speed while maintaining high molecular quality. --- ### LensVLM: Selective Context Expansion for Compressed Visual Representation of Text Authors: Roy Xie, Dan Friedman, Donghan Yu, Bowen Pan, Christopher Fifty, Jang-Hyun Kim, Xianzhi Du, Zhe Gan, Vivek Rathod, Bhuwan Dhingra Published: 2026-05-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.07019v1 Vision Language Models (VLMs) offer the exciting possibility of processing text as rendered images, bypassing the need for tokenizing the text into long token sequences. Since VLM image encoders map fixed-size images to a fixed number of visual tokens, varying rendering resolution provides a fine-grained compression knob. However, accuracy deteriorates quickly as compression increases: characters shrink below the vision encoder's effective resolution, making them indistinguishable. To address this, we propose LensVLM, an inference framework and post-training recipe that enables VLMs to scan compressed images, then selectively expand only the relevant images to their uncompressed form via learned tools. Building on Qwen3.5-9B-Base, LensVLM maintains accuracy comparable to the full-text upper bound at 4.3x effective compression and outperforms retrieval-based, text- and visual-compression baselines up to 10.1x effective compression across seven text QA benchmarks. LensVLM also generalizes to multimodal document and code understanding tasks, with the accuracy gain over baselines growing as compression increases. Our analysis validates this approach: training makes visual compression robust to rendering choices, and as compression grows the model increasingly relies on expanded content rather than unreliable visual reading. The analysis also yields practical tool-choice guidance: text expansion is preferable for rendered text, while high-resolution image expansion suits native documents whose layout cues carry task-relevant information. --- ### CoBit: Language Modeling with Bitstream Diffusion Authors: Georgios Batzolis, Mark Girolami, Luca Ambrogioni Published: 2026-05-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.07013v2 Diffusion language models (DLMs) promise parallel, order-agnostic generation, but on standard benchmarks they have historically lagged behind autoregressive models in sample quality and diversity. Recent continuous flow and diffusion approaches have narrowed this gap. In this work, we further close the autoregressive gap by modeling text as a continuous diffusion process over fixed-width binary bitstreams. We refer to the resulting model as CoBit (Continuous Bitstream Diffusion). Our approach represents semantic tokens as analog bit sequences and uses a matched-filter residual parameterization to isolate contextual learning from analytic independent-bit posteriors. Crucially, we adopt a stochastic sampler that applies Langevin-type corrections gated by the entropy-rate profile, concentrating stochasticity in high-information regions while remaining nearly deterministic elsewhere. On LM1B, our 130M-parameter model reaches a generative perplexity (GenPPL) of 59.76 at matched real-data entropy (4.31) using 256 neural function evaluations (NFEs), outperforming prior DLM baselines and reaching the autoregressive reference. On OpenWebText (OWT), our sampler establishes a new continuous-DLM Pareto frontier, achieving GenPPL 27.06 at entropy 5.26 using 4x fewer steps than previous 1024-NFE baselines. Scaling the same recipe to a 462M-parameter model (CoBit-M) further improves the OWT GenPPL-entropy frontier over the 130M model (CoBit-S) and over medium-scale continuous and discrete DLM baselines, reaching GenPPL 19.5 at entropy 5.40, near real-data entropy (5.44), and approaching pretrained GPT-2 Medium over the high-quality region. As an additional benefit, bitstream diffusion removes the O(V) vocabulary scaling bottleneck of standard DLMs: by predicting O(log V) bitwise logits via semantic bit-patching, it lowers memory and raises throughput, a scalable paradigm as vocabulary sizes grow. --- ### Dual-Agent Co-Training for Health Coaching via Implicit Adversarial Preference Optimization Authors: Da Long, Lingyi Fu, Diya Michelle Rao, Jasmine Ruales Carrera, Yang Bai, Shandian Zhe Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.07011v1 Motivational-interviewing-based health coaching is an effective approach for improving mental health and promoting healthy behavior change. However, the scarcity of trained human coaches and the high cost of coaching services make such support inaccessible to many people who could benefit from it. This motivates the development of AI health coaches that can provide scalable and affordable support. Existing methods typically optimize only one side of the interaction: they either train a dialogue agent against a fixed client environment or train a client simulator against a fixed assistant. This one-sided setup can limit exploration of the interaction space and may be inefficient at developing the capabilities required by the target agent and pushing its performance boundaries. In this paper, we propose a dual-agent framework that interactively co-trains both the health coach agent and the client simulator. The coach is optimized with DPO using Pareto-dominant response pairs identified by a multi-dimensional LLM judge. In turn, the client is trained adversarially by reversing these preferences, inducing an implicit adversarial training dynamic. We further show that this co-training process admits a natural stochastic-game interpretation. Extensive experiments demonstrate that our method effectively improves coaching quality across several important dimensions. --- ### Inductive Power Grid Cascading Failure Analysis with GRU-Gated Graph Attention Authors: Tianxin Zhou, Xiang Li, Haibing Lu Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.07010v1 Identifying vulnerable transmission lines in power grids before a cascading failure occurs is challenging: existing methods can learn inter-line failure correlations from cascade data, but they are trained and evaluated on a single grid, and transferring the learned knowledge to an unseen grid remains an open problem. We address this by training a single Gated Recurrent Unit (GRU)-gated Graph Attention Network on combined cascading failure data from limited training grids and applying it directly to any unseen grid without retraining. A GRU gate controls what information each node retains or discards at each cascade iteration. Empirical evaluation shows that the model transfers zero-shot to multiple new grids spanning inter-time and inter-domain settings. Using information extracted from the trained model, we consistently identify more vulnerable lines than established structural and electrical baselines. --- ### Equivalence of Coarse and Fine-Grained Models for Learning with Distribution Shift Authors: Adam R. Klivans, Shyamal Patel, Konstantinos Stavropoulos, Arsen Vasilyan Published: 2026-05-07 Categories: cs.DS, cs.LG arXiv: https://arxiv.org/abs/2605.07005v2 Recent work on provably efficient algorithms for learning with distribution shift has focused on two models: PQ learning (Goldwasser et al. (2020)) and TDS learning (Klivans et al. (2024)). Algorithms for TDS learning are allowed to reject a test set entirely if distribution shift is detected. In contrast, PQ learners may only reject points that are deemed out-of-distribution on an individual basis. Our main result is a surprising equivalence between these two models in the distribution-free setting. In particular, we give an efficient black-box reduction from PQ learning to TDS learning for any Boolean concept class. This equivalence implies the first hardness results for distribution-free TDS learning of basic classes such as halfspaces. The main technical contribution underlying our equivalence is a method for boosting, via branching programs, the weak distinguishing power of TDS learners that have rejected the target domain. We also show that giving a learner access to membership queries sidesteps these hardness results and allows for efficient, distribution-free PQ learnability of halfspaces. Our algorithm iteratively recovers large-margin separators obtained by applying successive Forster transforms on the training data. --- ### Adaptive auditing of AI systems with anytime-valid guarantees Authors: Siyu Zhou, Patrick Vossler, Venkatesh Sivaraman, Yifan Mai, Jean Feng Published: 2026-05-07 Categories: cs.AI, math.ST, stat.ML arXiv: https://arxiv.org/abs/2605.07002v1 A major bottleneck in characterizing the failure modes of generative AI systems is the cost and time of annotation and evaluation. Consequently, adaptive testing paradigms have gained popularity, where one opportunistically decides which cases and how many to annotate based on past results. While this framework is highly practical, its extreme flexibility makes it difficult to draw statistically rigorous conclusions, as it violates classical assumptions: the number of observations is typically limited (often 10 to 50 cases) and decisions regarding sampling and stopping are made in the midst of data collection rather than based a pre-specified rule. To characterize what statistical inferences can be drawn from highly adaptive audits, we introduce a hypothesis testing framework from two 'dueling' perspectives: (i) the model's null that asserts there is no failure mode with performance below a target threshold versus (ii) the auditor's null that asserts they have a sampling strategy that will uncover a failure mode. Leveraging Safe Anytime-Valid Inference (SAVI), we formalize the auditor as conducting 'testing by betting', which translates into simultaneous e-processes for testing the dueling null hypotheses. Furthermore, if the auditor is sufficiently powerful, we prove that these two hypotheses are asymptotically inverses of each other, in that passage of a stringent audit does in fact certify the AI system as being globally robust. Empirically, we demonstrate that our proposed testing procedures maintain anytime-valid type-I error control, outperform pre-specified testing methods, and can reach statistically rigorous conclusions sometimes with as few as 20 observations. --- ### SmellBench: Evaluating LLM Agents on Architectural Code Smell Repair Authors: Ion George Dinu, Marian Cristian Mihăescu, Traian Rebedea Published: 2026-05-07 Categories: cs.SE, cs.CL arXiv: https://arxiv.org/abs/2605.07001v2 Architectural code smells erode software maintainability and are costly to repair manually, yet unlike localized bugs, they require cross-module reasoning about design intent that challenges both developers and automated tools. While large language model agents excel at bug fixing and code-level refactoring, their ability to repair architectural code smells remains unexplored. We present the first empirical evaluation of LLM agents on architectural code smell repair. We contribute SmellBench, a task orchestration framework that incorporates smell-type-specific optimized prompts and supports iterative multi-step execution, together with a scoring methodology that separately evaluates repair effectiveness, false positive identification, and net codebase impact. We evaluate 11 agent configurations from four model families (GPT, Claude, Gemini, Mistral) on 65 hard-severity architectural smells detected by PyExamine in the Python project scikit-learn, validated against expert judgments. Expert validation reveals that 63.1% of detected smells are false positives, while the best agent achieves a 47.7% resolution rate. Agents identify false positives with up to $κ= 0.94$ expert agreement, but repair aggressiveness and net codebase quality are inversely related: the most aggressive agent introduces 140 new smells. These findings expose a gap between current LLM capabilities in localized code transformations and the architectural understanding needed for cross-module refactoring. SmellBench provides reusable infrastructure for tracking progress on this underexplored dimension of automated software engineering. We release our code and data at https://doi.org/10.5281/zenodo.19247588. --- ### Echo: KV-Cache-Free Associative Recall with Spectral Koopman Operators Authors: Anupama Sridhar, Alexander Johansen Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.06997v1 Long chain-of-thought reasoning and agentic tool-calling produce traces spanning tens of thousands of tokens, yet Transformer KV caches grow linearly with sequence length, creating a memory bottleneck on commodity hardware. State-space models offer constant-memory recurrence but suffer a memory cliff: retrieval accuracy collapses once the gap between a stored fact and its query exceeds the effective horizon of the recurrent state. We introduce Echo, a KV-cache-free associative recall architecture built around Spectral Koopman Attention (SKA); a drop-in replacement for attention layers that augments SSM blocks with a closed-form dynamical operator whose sufficient statistics are accumulated in constant memory with no KV cache. Echo fits a spectral linear system to the key and value history via kernel ridge regression and retrieves through a learned power-iterated filter, all from $O(r^{2})$ streaming state where $r$ is a small projection rank. On the Multi-Query Associative Recall benchmark, a pure Mamba-2 SSM fails to exceed chance accuracy (${\sim}3\%$) across all gap lengths and KV-pair counts, while at the 50M parameter scale SKA-augmented models achieve $100\%$ retrieval accuracy on every configuration tested, including distractor gaps of $4{,}096$ tokens with $32$ KV pairs. Across five additional transfer benchmarks including needle-in-a-haystack, tool-trace, and multi-hop retrieval, SKA consistently outperforms both pure SSM and SSM+Attention hybrids while maintaining constant inference memory. Ablations confirm that the spectral operator, not the prefix masking strategy, drives the retrieval gain. --- ### Optimal Experiments for Partial Causal Effect Identification Authors: Tobias Maringgele, Jalal Etesami Published: 2026-05-07 Categories: cs.AI, stat.ML arXiv: https://arxiv.org/abs/2605.06993v2 Causal queries are often only partially identifiable from observational data, and experiments that could tighten the resulting bounds are typically costly. We study the problem of selecting, prior to observing experimental outcomes, a cost-constrained subset of experiments that maximally tightens bounds on a target query. We formalize this as the max-potency problem, where epistemic potency measures the worst-case reduction in bound width guaranteed by an experiment, and show that this problem is NP-hard via a reduction from 0-1 knapsack. Building on the polynomial-programming framework of Duarte et al. (2023), we give a general procedure for evaluating epistemic potency in discrete settings. To control the super-exponential search space, we introduce two graphical pruning criteria that depend only on the causal graph and the query: a novel path-interception rule that exploits district structure to certify zero potency in linear time, and an identifiability check based on the ID algorithm. On Erdos-Renyi random graphs and 11 bnlearn benchmark networks, the two criteria together prune 50-88% of candidate experiments on average without solving a single polynomial program. For the general subset search, we show that ID-pruned experiments are combinatorially inert, yielding a super-exponential reduction in the number of subsets evaluated. We close with an end-to-end demonstration on observational NHANES data, selecting optimal experiments for estimating the effect of physical activity on diabetes. --- ### Optimal Experiments for Partial Causal Effect Identification Authors: Tobias Maringgele, Jalal Etesami Published: 2026-05-07 Categories: cs.AI, stat.ML arXiv: https://arxiv.org/abs/2605.06993v1 Causal queries are often only partially identifiable from observational data, and experiments that could tighten the resulting bounds are typically costly. We study the problem of selecting, prior to observing experimental outcomes, a cost-constrained subset of experiments that maximally tightens bounds on a target query. We formalize this as the max-potency problem, where epistemic potency measures the worst-case reduction in bound width guaranteed by an experiment, and show that this problem is NP-hard via a reduction from 0-1 knapsack. Building on the polynomial-programming framework of Duarte et al. (2023), we give a general procedure for evaluating epistemic potency in discrete settings. To control the super-exponential search space, we introduce two graphical pruning criteria that depend only on the causal graph and the query: a novel path-interception rule that exploits district structure to certify zero potency in linear time, and an identifiability check based on the ID algorithm. On Erdos-Renyi random graphs and 11 bnlearn benchmark networks, the two criteria together prune 50-88% of candidate experiments on average without solving a single polynomial program. For the general subset search, we show that ID-pruned experiments are combinatorially inert, yielding a super-exponential reduction in the number of subsets evaluated. We close with an end-to-end demonstration on observational NHANES data, selecting optimal experiments for estimating the effect of physical activity on diabetes. --- ### Why Does Agentic Safety Fail to Generalize Across Tasks? Authors: Yonatan Slutzky, Yotam Alexander, Tomer Slor, Yoav Nagel, Nadav Cohen Published: 2026-05-07 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2605.06992v2 AI agents are increasingly deployed in multi-task settings, where the task to perform is specified at test time, and the agent must generalize to unseen tasks. A major concern in such settings is safety: often, an agent must not only execute unseen tasks, but do so while avoiding risks and handling ones that materialize. Empirical evidence suggests that even when the ability to execute generalizes to unseen tasks, the ability to do so safely frequently does not. This paper provides theory and experiments indicating that failures of agentic safety to generalize across tasks are not merely due to limitations of training methods, but can also reflect an inherent property of safety itself: the relationship between a task and its safe execution is more complex than the relationship between a task and its execution alone. Theoretically, we analyze linear-quadratic control with $H_{\infty}$-robustness, and prove that in many cases, the mapping from task specification to an optimal controller has higher Lipschitz constant with safety requirements than without, yielding a Lipschitz bound of independent interest. Empirically, we demonstrate our conclusions in simulated quadcopter navigation with a neural network agent and in CRM with an LLM agent, showing that imitating safe and unsafe teachers is similarly straightforward on tasks seen in training, yet generalizing across tasks is considerably more difficult with the safe teacher. Our findings suggest that current efforts to enhance agentic safety may be insufficient, and point to a need for fundamentally different approaches. --- ### Why Does Agentic Safety Fail to Generalize Across Tasks? Authors: Yonatan Slutzky, Yotam Alexander, Tomer Slor, Yoav Nagel, Nadav Cohen Published: 2026-05-07 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2605.06992v1 AI agents are increasingly deployed in multi-task settings, where the task to perform is specified at test time, and the agent must generalize to unseen tasks. A major concern in such settings is safety: often, an agent must not only execute unseen tasks, but do so while avoiding risks and handling ones that materialize. Empirical evidence suggests that even when the ability to execute generalizes to unseen tasks, the ability to do so safely frequently does not. This paper provides theory and experiments indicating that failures of agentic safety to generalize across tasks are not merely due to limitations of training methods, but reflect an inherent property of safety itself: the relationship between a task and its safe execution is more complex than the relationship between a task and its execution alone. Theoretically, we analyze linear-quadratic control with $H_{\infty}$-robustness, and prove that the mapping from task specification to an optimal controller has higher Lipschitz constant with safety requirements than without, yielding a Lipschitz bound of independent interest. Empirically, we demonstrate our conclusions in simulated quadcopter navigation with a neural network agent and in CRM with an LLM agent. Our findings suggest that current efforts to enhance agentic safety may be insufficient, and point to a need for fundamentally different approaches. --- ### TrajGANR: Trajectory-Centric Urban Multimodal Learning via Geospatially Aligned Neural Representations Authors: Maria Despoina Siampou, Gengchen Mai, Ni Lao, Jinmeng Rao, Neha Arora, Cyrus Shahabi, Shushman Choudhury Published: 2026-05-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.06990v2 Many urban prediction tasks depend not only on the static attributes of a location, such as the visual appearance of its built environment, but also on its dynamic function: how people navigate and use it over time. Predicting congestion, travel demand, or road safety risks, for instance, requires mobility information that imagery and geographic coordinates alone cannot provide. Trajectories carry this signal, yet current trajectory representations are difficult to align with static geospatial observations: sparse, irregular GPS samples rarely coincide with street-view image coordinates, and existing encoders produce either a single embedding per trajectory or road segment, or embeddings only at the observed GPS samples. We introduce TrajGANR, a trajectory-centric multimodal self-supervised learning framework that represents each trajectory as a continuous, path-conditioned neural field. Reconstructed from the observed GPS samples, this representation can be queried at arbitrary coordinates (including the off-path locations where street-view images were captured), yielding localized mobility embeddings that align with street-view image and location representations at the same coordinate. Across four fine-grained road-level tasks in San Francisco and Porto, TrajGANR consistently outperforms recent geospatial and trajectory foundation models, with ablations attributing the gains to both the mobility signal and the fine granularity at which it is aligned. TrajGANR's location encoder further generalizes beyond the road network to areas where neither street-view imagery nor trajectories are available. --- ### TRAJGANR: Trajectory-Centric Urban Multimodal Learning via Geospatially Aligned Neural Representations Authors: Maria Despoina Siampou, Gengchen Mai, Ni Lao, Jinmeng Rao, Neha Arora, Cyrus Shahabi, Shushman Choudhury Published: 2026-05-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.06990v1 Multimodal self-supervised learning (MSSL) has emerged as a key paradigm for pretraining geospatial foundation models. However, existing geospatial MSSL methods are mainly designed for static pairs of modalities, such as satellite imagery, street-view imagery, and text, where learning is driven by aligning observations from the same or nearby locations. This assumption breaks down for human mobility trajectories, which represent continuous movement along paths rather than discrete observations at individual locations. Although trajectories are important for urban understanding through their ability to capture human activity across roads, neighborhoods, and places over time, they remain largely underexplored in current geospatial MSSL frameworks. We present TrajGANR, a novel trajectory-centric geospatial MSSL framework that aligns continuous movement patterns with static, location-based observations. TrajGANR learns a continuous neural representation of trajectories at arbitrary points along each path, which enables fine-grained alignment with nearby street-view images, even when they are not co-located with any trajectory waypoints. We leverage this capability to introduce an MSSL objective that jointly aligns three modalities: trajectories, street-view images, and their geographic locations. We evaluate TrajGANR on four urban mobility and road understanding tasks. Across these tasks, TrajGANR consistently outperforms existing geospatial MSSL frameworks and a trajectory-specific foundation model. Ablation studies further demonstrate that our proposed MSSL objective and the multimodal learning framework are the primary drivers of these improvements, highlighting the importance of fine-grained geospatial alignment over coarser aggregation, as well as geospatial multimodal learning. --- ### Drawing Lines in Psychological Space: What K-means Clustering Reveals in Simulated and Real Psychometric Data Authors: Pedro Henrique Ramos Pinto, Maria Jullyanna Ferreira Marques, Luiz Carlos Serramo Lopez Published: 2026-05-07 Categories: stat.AP, cs.AI, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2605.06989v1 K-means clustering is widely used in psychological and psychometric research to identify profiles, subgroups, and potential typologies, yet its classical formulation does not test whether such groups exist as latent psychological categories. Instead, K-means partitions multidimensional space into regions around centroids, favoring compact, approximately spherical clusters defined by geometric distance. In this paper, we examine this limitation through a sequence of controlled simulated datasets. We then extend the analysis to the SMARVUS dataset, a large international psychometric dataset comprising survey responses from university students across 35 countries, to evaluate whether similar geometric partitioning patterns emerge in empirical psychological data. By contrasting simulated and empirical data, this paper argues that K-means can produce stable and visually coherent clustering solutions even in continuous Gaussian latent spaces without true subgroup structure. --- ### Response Time Enhances Alignment with Heterogeneous Preferences Authors: Federico Echenique, Alireza Fallah, Baihe Huang, Michael I. Jordan Published: 2026-05-07 Categories: cs.LG, cs.GT, econ.TH, stat.ML arXiv: https://arxiv.org/abs/2605.06987v1 Aligning large language models (LLMs) to human preferences typically relies on aggregating pooled feedback into a single reward model. However, this standard approach assumes that all labelers share the same underlying preferences, ignoring the fact that real-world labelers are highly heterogeneous and usually anonymous. Consequently, relying solely on binary choice data fundamentally distorts the learned policy, making the true population-average preference unidentifiable. To overcome this critical limitation, we demonstrate that augmenting preference datasets with a simple, secondary signal -- the user's response time -- can restore the identifiability of the population's average preference. By modeling each decision as a Drift-Diffusion Model (DDM), we introduce a novel, consistent estimator of heterogeneous preferences that successfully corrects the distortions of standard choice-only labels. We prove that our estimator asymptotically converges to the true average preference even in extreme cases where each anonymous labeler contributes only a single choice. Empirically, across both synthetic and real-world datasets, our method consistently outperforms standard baselines that otherwise fail and plateau at a bias floor. Because response times are essentially free to record and require zero user tracking or identification, our results bring promises and open up new opportunities for future data-collection pipelines to improve the social benefit without requiring user-level identifiers or repeated elicitations. --- ### FastOmniTMAE: Parallel Clause Learning for Scalable and Hardware-Efficient Tsetlin Embeddings Authors: Ahmed K. Kadhim, Lei Jiao, Rishad Shafik, Ole-Christoffer Granmo, Mayur Kishor Shende Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.06982v1 Embedding models in natural language processing (NLP) increasingly rely on deep architectures such as BERT, while simpler models such as Word2Vec provide efficient representations but limited interpretability. The Tsetlin Machine (TM) offers an alternative logic-based learning paradigm. Omni TM Autoencoder (Omni TM-AE) applies this paradigm to static embedding by exploiting automaton state distributions within a single clause layer, but its training process remains slow. In this work, we propose FastOmniTMAE, a reformulation of Omni TM-AE that replaces sequential training dependencies with a two-stage parallel process: evaluation and update. Using a Single-Run Multi-Environment Benchmark covering classification, similarity, and clustering, FastOmniTMAE achieves up to 5$\times$ faster training in classification while maintaining comparable embedding quality under both Spearman and Kendall similarity measures. To address the limited efficiency of TM training on conventional GPUs, we further implement FastOmniTMAE as a reusable accelerator on SoC-FPGA platforms. The Multi-Hardware Benchmark shows that FastOmniTMAE achieves similarity scores of 0.669 on a resource-constrained FPGA and 0.696 on an UltraScale+ SoC, demonstrating efficient logic-based embedding training with a small hardware footprint. --- ### Bridging Textual Profiles and Latent User Embeddings for Personalization Authors: Zhaoxuan Tan, Xiang Zhai, Yan Zhu, Meng Jiang, Mohamed Hammad Published: 2026-05-07 Categories: cs.IR, cs.CL arXiv: https://arxiv.org/abs/2605.06981v1 Personalized systems rely on user representations to connect behavioral history with downstream recommendation applications. Existing methods typically employ either supervised latent user embeddings, which are effective for retrieval but difficult to interpret, or textual user profiles, which are interpretable but challenging to optimize for downstream utility due to lack of direct supervision. To bridge this gap, we present BLUE, a reinforcement learning framework that unifies these two forms of user representation by aligning language-based user profiles with embedding-based recommendation objectives. Given a user interaction history, BLUE leverages a profiler Large Language Model (LLM) to generate textual profiles, while an embedding model provides reward signals. This encourages the resulting textual representations to move closer to positive items and farther from negative ones in the embedding space. We further introduce a text-space supervision signal based on next-item prediction, ensuring the learned profiles remain both semantically meaningful and highly effective for downstream retrieval. Experiments on Amazon Reviews 2023 and Google Local Reviews in zero-shot sequential recommendation settings demonstrate that BLUE consistently outperforms strong baselines under both frozen and trainable embedding conditions. Notably, BLUE achieves clear gains in cross-domain transfer, highlighting the strong generalization ability of the learned user profiles. Furthermore, these generated profiles provide superior personalized context for question answering compared to raw user histories or alternative profile optimization methods. Overall, these results show that BLUE provides an effective way to unify interpretable textual profiling with discriminative latent embeddings for personalization. --- ### PLOT: Progressive Localization via Optimal Transport in Neural Causal Abstraction Authors: Jonathn Chang, Arya Datla, Ziv Goldfeld Published: 2026-05-07 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2605.06979v1 Causal abstraction offers a principled framework for mechanistic interpretability, aligning a high-level causal model with the low-level computation realized by a neural network through counterfactual intervention analysis. Existing methods such as distributed alignment search (DAS) learn expressive subspace interventions, but the relevant neural site is unknown a priori, so finding a handle requires a computationally burdensome search over candidate sites. We introduce PLOT (Progressive Localization via Optimal Transport), a transport-based framework that localizes causal variables from the output effect geometry of abstract and neural interventions. PLOT fits an optimal transport coupling between abstract variables and candidate neural sites, yielding a global soft correspondence that can be calibrated into intervention handles. In simple settings, a single coupling over individual neurons suffices. In larger models, PLOT is applied progressively, moving from coarse sites such as tokens, timesteps, or layers to finer supports such as coordinate groups or PCA spans, and optionally guiding DAS based on the localized signal. Across experiments of increasing complexity, transport-only PLOT handles are exceedingly fast and competitive on accuracy, while PLOT-guided DAS reaches DAS-level accuracy at a fraction of full DAS runtime, providing an efficient localization engine for causal abstraction research at scale. --- ### Group of Skills: Group-Structured Skill Retrieval for Agent Skill Libraries Authors: Kun Zeng, Yu Huo, Siyu Zhang, Zi Ye, Yuecheng Zhuo, Haoyue Liu, Yuquan Lu, Junhao Wen, Xiaoying Tang Published: 2026-05-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.06978v1 Skill-augmented agents increasingly rely on large reusable skill libraries, but retrieving relevant skills is not the same as presenting usable context. Existing methods typically return atomic skills or dependency-aware bundles whose internal roles remain implicit, leaving the agent to infer the execution entry point, support skills, visible requirements, and failure-avoidance guidance. We introduce Group of Skills (GoSkills), an inference-time group-structured retrieval method that changes the agent-facing retrieval object from a flat skill list to a compact, role-labeled execution context. GoSkills builds anchor-centered skill groups from a typed skill graph, expands support groups through a group graph, bottlenecks the selected group plan into a bounded set of atomic skill payloads, and renders a fixed execution contract with Start, Support, Check, and Avoid fields, without changing the downstream agent, skill payloads, or execution environment. Experiments on SkillsBench and ALFWorld show that GoSkills preserves visible-requirement coverage under a small skill budget, improves over flat skill-access baselines, and often improves reward and agent-only runtime relative to structural retrieval references. --- ### $f$-Divergence Regularized RLHF: Two Tales of Sampling and Unified Analyses Authors: Di Wu, Chengshuai Shi, Jing Yang, Cong Shen Published: 2026-05-07 Categories: cs.LG, cs.AI, cs.IT, stat.ML arXiv: https://arxiv.org/abs/2605.06977v1 Reinforcement Learning from Human Feedback (RLHF) has become a cornerstone technique for post-training large language models. While most existing approaches rely on the reverse KL-regularization, recent empirical studies have begun exploring alternative divergences (e.g., forward KL, chi-squared) as regularizers in RLHF. However, a unified theoretical understanding of general $f$-divergence regularization remains under-explored. To fill this gap, this work develops a comprehensive theoretical framework for online RLHF with a general $f$-divergence regularized objective. Rather than treating each possible divergence function individually, we adopt a holistic perspective across the entire function class and propose two algorithms based on distinct sampling principles. The first extends the classical optimism principle with a carefully designed exploration bonus, while the second introduces a new method that exploits the sensitivity of the optimal policy to reward perturbations under $f$-divergence regularization. Theoretical analysis shows that $O(\log T)$ regret and $O(1/T)$ sub-optimality gap are achievable, establishing provable efficiency of both algorithms and, to the best of our knowledge, the first performance bounds for online RLHF under general $f$-divergence regularization. --- ### A Differentiable Bayesian Relaxation for Latent Partial-Order Inference Authors: Dongqing Li, Geoff K. Nicholls, Shiyi Sun, You Luo Published: 2026-05-07 Categories: stat.ML, cs.LG, stat.CO arXiv: https://arxiv.org/abs/2605.06976v1 Many ranking and agent trace datasets are recorded as linear orders even though their latent structure is only partially ordered. This is especially common in agent and workflow traces, where observed order may reflect arbitrary linearization rather than true prerequisites. We introduce a differentiable relaxation for latent partial-order inference from such traces. Starting from a hard frontier-constrained model of noisy linear extensions, we replace discontinuous product-order precedence and binary frontier feasibility with smooth surrogates, yielding a continuous posterior that preserves closure-level partial-order semantics and supports gradient-based MCMC and variational inference. We prove soft transitivity, sharp-limit frontier recovery, and convergence to the hard likelihood. Experiments on synthetic data, records of social dominance relations, and cloud-agent traces show close posterior fidelity to hard MCMC on small instances and improved runtime--accuracy trade-offs on larger problems. --- ### Decentralized Time-Varying Optimization for Streaming Data via Temporal Weighting Authors: Muhammad Faraz Ul Abrar, Nicolò Michelusi, Erik G. Larsson Published: 2026-05-07 Categories: eess.SP, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2605.06971v1 Classical optimization theory largely focuses on fixed objective functions, whereas many modern learning systems operate in dynamic environments where data arrive sequentially and decisions must be updated continuously. In this work, we study optimization with streaming data over a distributed network of agents. We adopt a structured, weight-based formulation that explicitly captures the streaming-data origin of the time-varying objective: at each time step, every agent receives a new sample, and the network seeks to track the minimizer of a temporally weighted objective formed from all samples observed across the network so far. We focus on decentralized gradient descent (DGD) with a limited communication/computation budget, where at each time step, only a limited number of DGD iterations can be performed before the objective changes again. For strongly convex and smooth losses, we analyze the tracking error with respect to the time-varying minimizer through a fixed-point theory lens. Our analysis reveals that the tracking error decomposes into a fixed-point tracking term and a bias term induced by data heterogeneity across agents. We specialize the analysis to two natural weighting strategies: uniform weights, which treat all samples equally, and exponentially discounted weights, which geometrically decay the influence of older data. Under uniform weighting, DGD tracks the fixed-point at a rate $\mathcal{O}(1/t)$, whereas discounted weighting yields a non-vanishing fixed-point tracking floor controlled by the discount factor. In both cases, decentralization induces an additional non-zero bias floor under a constant step size. We validate our theoretical findings through numerical simulations. --- ### Bringing Multimodal Large Language Models to Infrared-Visible Image Fusion Quality Assessment Authors: Yuchen Guo, Junli Gong, Yao Lu, Xintong Xu, Yiuming Cheung, Weifeng Su Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.06969v2 Infrared-Visible image fusion (IVIF) aims to integrate thermal information and detailed spatial structures into a single fused image to enhance perception. However, existing evaluation approaches tend to over-optimize both hand-crafted no-reference statistics and full-reference metrics that treat the source images as pseudo ground truths. Recent IVIF reward-modelling efforts learn from human ratings but use scalar regression on aggregated scores, neither leveraging the reasoning of Multimodal Large Language Models (MLLMs) nor encoding per-image perceptual ambiguity in their supervision, but naively introducing MLLMs with discrete one-hot supervision likewise collapses fused images of similar quality into different rating levels. To address this, we introduce FuScore, which utilizes an MLLM to mimic human visual perception by producing continuous quality score, rather than discrete level predictions, enabling fine-grained discrimination among fused images of similar quality. We exploit the agreement among four IVIF-specific sub-dimensions to construct a per-image soft label whose sharpness reflects how consensual the overall judgment is. We further introduce a tripartite objective combining per-image distributional supervision, within-source-pair Thurstone fidelity for method-level ordering, and cross-source-pair Thurstone fidelity for scene-level ordering across scenes. Extensive experiments demonstrate that FuScore achieves state-of-the-art correlation with human visual preferences. --- ### AI and Consciousness: Shifting Focus Towards Tractable Questions Authors: Iulia-Maria Comsa Published: 2026-05-07 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2605.06965v1 As language-based AI systems become more anthropomorphic, the question of whether they can have subjective experience is increasingly pressing. I focus here on the tractability of research questions in the space of AI consciousness. I argue that the fundamental problem of whether AI systems can be conscious is currently intractable in its direct form, given the absence of a universally accepted scientific theory of consciousness, as well as the historical open-endedness of the philosophical mind-body problem. In contrast, questions around the adjacent subject of perceived AI consciousness are tractable, timely, and highly consequential for society. The general public is increasingly open to the possibility of consciousness in AI systems and routinely adopts the vocabulary of human cognition and subjective experience to describe them. This phenomenon is already driving societal shifts across user experience, ethical standards, and linguistic norms. I therefore propose an increased research focus on uncovering the causes and effects of perceived AI consciousness, which ultimately shape how we see our own human subjective experience relative to artificial entities. To support this, I map the current landscape of AI consciousness perception and discuss its key potential drivers and societal consequences. Finally, I urge developers, decision-makers, and the broader scientific community to commit to clear and accurate communication regarding the topic of AI consciousness, explicitly acknowledging its inherent uncertainties. --- ### From Surface Learning to Deep Understanding: A Grounded AI Tutoring System for Moodle Authors: Anna Ostrowska, Michał Kukla, Gabriela Majstrak, Jan Opala, Sebastian Pergała, Jan Skwarek, Anna Wróblewska Published: 2026-05-07 Categories: cs.HC, cs.AI, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2605.06963v1 This demo paper describes the development of the AI Teaching \& Learning Assistant, a modular Moodle plugin that leverages Retrieval-Augmented Generation (RAG) to deliver high-quality, hallucination-free education. The system employs a dual-centric design, providing students with interactive, Socratic-based tutoring and educators with a "human-in-the-loop" workspace for supervised content generation. By grounding Large Language Model (LLM) responses in teacher-provided materials, the assistant addresses the risks of misinformation while encouraging deep conceptual mastery. Evaluation via the Ragas (LLM-as-a-Judge) framework and a preliminary user study confirms its effectiveness, achieving faithfulness scores up to 0.97 and a 4.00/5.00 recommendation rate. --- ### Locally Near Optimal Piecewise Linear Regression in High Dimensions via Difference of Max-Affine Functions Authors: Haitham Kanj, Kiryung Lee Published: 2026-05-07 Categories: stat.ML, cs.LG, math.ST arXiv: https://arxiv.org/abs/2605.06959v1 This paper presents a parametric solution to piecewise linear regression through the Adaptive Block Gradient Descent (ABGD) algorithm. The heart of the method is the parametrization of piecewise linear functions as the difference of max-affine (DoMA) functions. A non-asymptotic local convergence analysis for ABGD is provided under sub-Gaussian covariate and noise distributions. To initialize ABGD, we adapt a prior algorithm originally developed for the simpler setting of max-affine functions. When suitably initialized, ABGD converges linearly to an $ε$-accurate estimate given $\tilde{\mathcal{O}}(d\max(σ_z/ε,1)^2)$ observations where $σ_z^2$ denotes the noise variance. This implies exact recovery given $\tilde{\mathcal{O}}(d)$ samples in the noiseless case. Also, such a rate is shown to be minimax optimal up to logarithmic factors. Synthetic numerical results corroborate the theoretical guarantees for ABGD. We also observe competitive performance compared to the state-of-the-art methods on real-world datasets. --- ### Learning and Reusing Policy Decompositions for Hierarchical Generalized Planning with LLM Agents Authors: Shirin Sohrabi, Haritha Ananthakrishnan, Harsha Kokel, Kavitha Srinivas, Michael Katz Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.06957v1 We present a dynamic policy-learning approach that combines generalized planning and hierarchical task decomposition for LLM-based agents. Our method, Hierarchical Component Learning for Generalized Policies (HCL-GP ), learns parameterized policies that generalize across task instances and automatically extracts reusable components from successful executions, organizing them into a component library for compositional policy generation. We address three challenges: (1) learning components through automated decomposition, (2) generalizing components to maximize reuse, and (3) efficient retrieval via semantic search. Evaluated on the AppWorld benchmark, our approach achieves 98.2% accuracy on normal tasks and 97.8% on challenge tasks with unseen applications, improving 15.8 points over static synthesis on challenging scenarios. For open-source models, dynamic reuse enables 62.5% success versus near-zero without reuse. This demonstrates that classical planning concepts can be effectively integrated with LLM agents for improved accuracy and efficiency. --- ### Kurtosis-Guided Denoising Score Matching for Tabular Anomaly Detection Authors: Victor Livernoche, Jie Zan, Reihaneh Rabbany Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.06955v1 Denoising score matching (DSM) provides a way to learn data distributions by training a neural network to recover the score function, defined as the gradient of the log density, from noise-corrupted samples. Once trained, the score magnitude at a test point reflects how consistent that point is with the learned distribution, making it a natural anomaly signal. The key practical challenge is selecting the perturbation scale: too little noise yields unstable score estimates in sparse regions, while too much erases local structure and weakens anomaly sensitivity. This is compounded by the difficulty of hyperparameter tuning when anomalies are unknown and no validation set is available. We introduce kurtosis-based noise scaling (K-DSM), a per-feature scheme that sets noise levels from the shape of each marginal distribution, improving coverage of low-density regions and precision in high-density regions without extra model complexity. Contrary to prior claims that multi-scale or noise-conditioned training is necessary, we find that a carefully trained single-scale model is already a strong anomaly detector. On standard tabular anomaly detection benchmarks, K-DSM achieves state-of-the-art performance in the semi-supervised setting. When combined with a lightweight EMA-teacher filtering rule that removes low-density training points before each gradient step, it also achieves strong performance in the fully unsupervised (contaminated) setting, suggesting that simple, data-adaptive noise scaling enables robust anomaly detection while reducing reliance on hyperparameter tuning. --- ### Multi-Objective Constraint Inference using Inverse reinforcement learning Authors: Syed Ihtesham Hussain Shah, Floris den Hengst, Aneta Lisowska, Annette ten Teije Published: 2026-05-07 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2605.06951v1 Constraint inference is widely considered essential to align reinforcement learning agents with safety boundaries and operational guidelines by observing expert demonstrations. However, existing approaches typically assume homogeneous demonstrations (i.e., generated by a single expert or multiple experts with identical objectives). They also have limited ability to capture individual preferences and often suffer from computational inefficiencies. In this paper, we introduce Multi-Objective Constraint Inference (MOCI), a novel framework designed to jointly extract shared constraints and individual preferences from heterogeneous expert trajectories, where multiple experts pursue different objectives. MOCI effectively models and learns from diverse, and potentially conflicting, behaviors. Empirical evaluations demonstrate that MOCI significantly outperforms existing baselines, achieving improved predictive performance, and maintaining competitive computational efficiency on a standard grid-world benchmark. These results establish MOCI as an accurate, flexible, and computationally practical approach for real-world constraint inference and preference learning tasks. ---