# Research Papers — 2026-05 · bundle page 251 of 326 > Full-text markdown of papers #12501–12550 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/251 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/250.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/252.md) ### When Child Inherits: Modeling and Exploiting Subagent Spawn in Multi-Agent Networks Authors: Ziwen Cai, Yihe Zhang, Xiali Hei Published: 2026-05-08 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2605.08460v1 Since the official release of ChatGPT in 2022, large language models (LLMs) have rapidly evolved from chatbot-style interfaces into agentic systems that can delegate work through tools and newly spawned subagents. While these capabilities improve automation and scalability, they also pose new security risks in multi-agent networks. Existing research has studied how individual LLM-based agents can be compromised through prompt injection, jailbreaking, poisoned retrieval data, or malicious extensions. Less is known about what happens after one agent is compromised inside a multi-agent network. In particular, inherited memory from parent agents can carry malicious instructions, outdated states, or unintended behavioral rules into newly created subagents, allowing a local compromise to spread across agent boundaries. In this paper, we model contemporary multi-agent networks through the lens of subagent inheritance. Our analysis shows that current frameworks can violate trust boundaries through insecure memory inheritance, weak resource control, stale post-spawn state, and improper termination authority. We demonstrate these risks in real agent frameworks and propose defenses based on explicit security invariants. Our findings show that inheritance is not merely an implementation detail, but a central component influencing the security of multi-agent systems. --- ### Neurally-plausible radial basis kernels using distributed Fourier embeddings Authors: Jakeb Chouinard Published: 2026-05-08 Categories: cs.LG, q-bio.NC arXiv: https://arxiv.org/abs/2605.08458v1 Coherent, continuous spatial representations are critical for synthesizing physical and perceptual phenomena into a single representational space. Radial basis kernels provide a path forward for this type of distributed representation. In this work, we aim to characterize and analyze common radial basis kernels realizable in the neurally-plausible framework of spatial semantic pointers. Further, we analyze previous radial basis kernel work based on grid cell-like representations and demonstrate that such representations are both capable of and optimal for realizing radial basis kernels. --- ### HEART: A High-Efficiency Adaptive Real-Time Telemonitoring Framework for Secure Electrocardiogram Signal Transmission Using Chaotic Encryption Authors: Beyazıt Bestami Yuksel Published: 2026-05-08 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2605.08456v1 The realtime analysis and secure transmission of electrocardiogram ECG signals are critical for accurate diagnosis and safeguarding patient privacy in telemedicine applications This study presents a novel realtime ECG monitoring system that employs a learnable key generator LKG derived from each patients own ECG signal characteristics to dynamically produce unique encryption keys These keys determine the parameters r and x0 of a logistic map used for chaotic encryption The system securely encrypts realtime ECG data immediately after acquisition ensuring confidential transmission and storage in the cloud For remote clinical access the encrypted data is downloaded and decrypted on the doctors side using the matching key generated at the source or securely stored in the cloud This approach eliminates the need for traditional key exchange and substantially raises the cost of exhaustive key search in practice through persegment biometric key refresh and combined permutation and XOR diffusion supported by minentropy evaluation Compared to statickey methods the learnable biometric key design offers greater unpredictability and individualization A comprehensive set of security assessments including Shannon entropy 7678 bits correlation and autocorrelation disruption histogram statistics NIST SP 80022 frequency testing plaintextkey sensitivity avalanche effect FFTbased spectral flatness and robustness to noise and occlusion confirms the methods strength Reconstruction fidelity MSE approximately 5x106 PSNR greater than 52 dB MAE approximately 0002 demonstrates nearlossless decryption and preserved diagnostic features Encryption latency remains low preserving realtime performance. --- ### CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging Authors: Shiyang Li, Haoyang Chen, Mattia Fazzini, Caiwen Ding Published: 2026-05-08 Categories: cs.LG, cs.PL, cs.SE arXiv: https://arxiv.org/abs/2605.08455v2 Debugging CUDA programs has long been challenging because failures often arise from subtle interactions among hardware behavior, compiler decisions, memory hierarchy, and asynchronous execution. More importantly, with the rapid expansion of GPU usage across scientific computing, machine learning, graphics, and systems workloads, CUDA debugging has become more challenging than ever. Current evaluations of LLM-based CUDA programming largely miss this setting: a model can pass correctness tests with repair by degeneration, simplifying the CUDA code into a safer but slower program that abandons the original optimization structure. We introduce CUDABEAVER, a benchmark for CUDA debugging from real failing workspaces produced during LLM-based CUDA generation. Each task provides the broken candidate, native build/test commands, raw error evidence, and a single editable file. CUDABEAVER evaluates whether a fixer truly repairs the failing CUDA code or merely finds a slower test-passing replacement, reporting results by failure category, debugging trajectory, stagnation mode, and performance preservation. We further propose pass@k(M,C,A), a protocol-conditional CUDA debugging metric by making the fixer M, corpus C, and protocol axes Aexplicit. Using this metric across 213 tasks and seven frontier LLMs, we show that protocol-aware evaluation gives a more faithful view of CUDA debugging ability: when performance-loss tolerance is high, fixers appear much stronger, but even a minor stricter performance requirement can sharply reduce measured success, shifting scores by up to 40 percentage points. --- ### Recovering Physical Dynamics from Discrete Observations via Intrinsic Differential Consistency Authors: Yuxiang Luo, Andrew Perrault Published: 2026-05-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08454v2 Recovering continuous-time dynamics from discrete observations is difficult because local supervision loses fidelity as the observation interval grows. We replace local supervision with a global structural constraint: any flow representing autonomous dynamics must satisfy the semi-group property under time translation. We train a time-conditioned secant velocity field whose deviation from this property---which we call Symmetry Rupture---serves two roles: as a training regularizer it confines the hypothesis space to flows that compose consistently across temporal scales; as an inference oracle it guides an adaptive solver to select the largest step size that preserves internal consistency, without relying on local truncation error estimates. On the diffusion-reaction and Navier-Stokes benchmarks, our method achieves the lowest rollout RMSE among all evaluated methods using $\leq 3$ function evaluations per step. On the near-conservative shallow water benchmark, our method matches or exceeds competing flow-based methods at far lower computational cost; the strongest Neural ODE baseline achieves lower absolute RMSE but at the expense of $>100$ function evaluations per step. In the direct auto-regressive setting, our solver maintains stable long-horizon rollouts where flow-matching baselines diverge and Neural ODE requires up to $12\times$ more function evaluations. The core contribution is a model that internalizes temporal consistency, removing the need for high-order external solvers to compensate for structural bias. --- ### Recovering Physical Dynamics from Discrete Observations via Intrinsic Differential Consistency Authors: Yuxiang Luo, Andrew Perrault Published: 2026-05-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08454v1 Recovering continuous-time dynamics from discrete observations is difficult because local supervision (e.g., pointwise regression targets, derivative approximations, or equation residuals) loses fidelity as the observation interval grows. We replace local supervision with a global structural constraint: any flow representing autonomous dynamics must satisfy the semi-group property under time translation. We train a time-conditioned secant velocity field whose deviation from this property, which we call Symmetry Rupture, serves two purposes. As a training regularizer, it confines the hypothesis space to flows that compose consistently across temporal scales. As an inference oracle, it lets the solver select the largest step size that preserves internal consistency, replacing the local truncation error that conventional adaptive solvers depend on. On the diffusion-reaction benchmark under time-informed inference, our method reduces rollout RMSE by 87\% while using 5x fewer function evaluations than a Neural ODE baseline. In the more demanding direct auto-regressive setting, where the model must predict distant future frames without intermediate temporal cues, our adaptive solver allocates compute based on local geometric complexity -- maintaining the lowest rollout RMSE on two of three PDE benchmarks while baselines either diverge or require up to an order of magnitude more function evaluations to remain stable. --- ### Sink vs. diagonal patterns as mechanisms for attention switch and oversmoothing prevention Authors: Peter Súkeník, Cristina López Amado, Christoph H. Lampert, Marco Mondelli Published: 2026-05-08 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2605.08453v1 This paper studies the role of sinks and diagonal patterns as attention switch and anti-oversmoothing mechanisms. We analyze geometric conditions under which sinks can be represented, showing a necessary alignment between the embedding of the sink and all other embeddings. Next, we refine the current understanding of the role of sinks in oversmoothing prevention: we specify the conditions under which dense attention provably smooths more than sparse attention, and empirically verify that such conditions are often satisfied in practice. We further prove an equivalence between sinks and hard attention switch, in which the output of the attention is identically 0. Finally, we relax the hard attention switch by allowing token self-communication: we provide a quantitative comparison of the costs of representing sinks vs.\ diagonal patterns, showing why sinks are favored in pretrained transformers. The introduction and analysis of diagonal patterns and the generalization of the attention switch close the gap between what oversmoothing prevention requires and what sinks provide, while also establishing when and why attention layers act like MLPs if token communication is not necessary. --- ### NICE FACT: Diagnosing and Calibrating VLMs in Quantitative Reasoning for Kinematic Physics Authors: Jian Lan, Zhicheng Liu, Xinpeng Wang, Yuhao Zhou, Haokun Chen, Jiancheng Lv, Barbara Plank, Thomas Seidl Published: 2026-05-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.08452v1 The ability to derive precise spatial and physical insights is a cornerstone of vision-language models (VLMs), yet their poor performances in related spatial intelligence tasks such as physical reasoning remain a fundamental barrier. The community critically lacks a scientific analysis revealing whether VLMs faithfully reach answers or plausibly make guesses. This work aims to provide a fundamental understanding of how VLMs perceive the physical world, and utilize physical laws, while assessing the reliability of model confidence. We propose NICE and FACT, a dual-diagnostic paradigm that explicitly decomposes quantitative reasoning for kinematic physics: FACT diagnoses visual fidelity, physical law comprehension, and temporal grounding. NICE studies our novel neighborhood-informed calibration method and novel metrics to evaluate and calibrate confidence reliability. Evaluated across 6 latest state-of-the-art VLMs, we uncover that models fail to identify visual preconditions or utilize necessary physical laws to reach answers. This work highlights and establishes a standardized diagnostic paradigm to guide the development of faithful, physically-grounded VLMs. --- ### RubiConv -- Efficient Boundary-Respecting Convolutions Authors: Linda Friso, Annie Marsden, Xinyi Chen, Arushi Gupta, Peter Bartlett, Mark Braverman, Elad Hazan Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08451v1 Convolutional architectures have emerged as powerful alternatives to Transformers for sequence modeling. The primary advantage is that they offer improved theoretical sequence length complexity by leveraging the Fast Fourier Transform (FFT). However, this theoretical improvement does not always meaningfully land in practice. One critical obstacle is that applying standard FFTs is not amenable to the large-scale training pipeline wherein data is packed from different sources into a single sequence for hardware efficiency. Indeed, standard FFT algorithms are not easily amenable to document packing. Existing workarounds suffer from severe inefficiencies, crippling the practical performance of convolutional architectures. We close this gap with RubiConv, a novel algorithm for performing hardware-efficient, boundary-respecting convolutions on packed sequences. Extensive experiments show that RubiConv achieves significant speedups over both attention and standard FFT-based baselines. This work makes the theoretical efficiency of long convolutional models a practical reality for large-scale, real-world data packing. --- ### Zero-shot Imitation Learning by Latent Topology Mapping Authors: Maxwell J. Jacobson, Yexiang Xue Published: 2026-05-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08450v1 Imitation learning is effective for training agents when expert demonstrations are available, but collecting demonstrations for every complex task in an environment is costly. We study the long-horizon, goal-conditioned setting where a fixed demonstration dataset contains useful behavior, but not complete examples for every task the agent must solve. Existing imitation learning methods can learn strong policies from demonstrations, but when solving long-horizon tasks, small errors accumulate over long primitive-action trajectories and make zero-shot adaptation to new tasks unreliable. We introduce Zero-shot Agents from Latent Topologies (ZALT), an imitation-learning method that solves unseen start-goal tasks beyond those demonstrated during training. ZALT identifies latent hub states where trajectories converge or diverge, learns policies and a dynamics model over hub-to-hub transitions, and plans over the hub topology to complete new tasks. This topology makes demonstrated behaviors explicitly composable while compressing long tasks into shorter sequences of abstract transitions -- combined, these enable ZALT to perform zero-shot adaptation. In a complex 3D maze environment, ZALT achieves 55% zero-shot success on unseen tasks, compared to 6% for the strongest baseline. --- ### LLM-guided Semi-Supervised Approaches for Social Media Crisis Data Classification Authors: Jacob Ativo, Bharaneeshwar Balasubramaniyam, Anh Tran, Khushboo Gupta, Hongmin Li, Doina Caragea, Cornelia Caragea Published: 2026-05-08 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.08448v1 Semi-supervised learning approaches have been investigated as a means to enhance the analysis of social media data in disaster management contexts. In this work, we present the first empirical evaluation of large language model (LLM) guided semi-supervised learning for crisis related tweet classification. We compare two recent LLM assisted semi-supervised methods, VerifyMatch and LLM guided Co-Training ( LG-CoTrain), against established semi-supervised baselines. Our results show that LG-CoTrain significantly outperforms classical semi-supervised approaches in low resource settings with 5, 10 and 25 labeled examples per class, achieving the highest averaged Macro F1 across events. VerifyMatch achieves competitive performance while also demonstrating strong calibration properties. As the number of labeled examples increases, the performance gap narrows and Self Training emerges as a strong baseline. We further observe that compact semi-supervised models can, in some cases, outperform very large LLMs operating in zero-shot settings. This finding highlights the potential of transferring knowledge from LLMs into smaller and more deployable models through LLM guided semi-supervised learning, offering a practical pathway for real world disaster response applications. Our project repository on Github is here. --- ### Revisiting the syntax of imperatives in Yemeni Arabic: An Agree across phases approach Authors: Mohammed Q. Shormani Published: 2026-05-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.08447v1 This article revisits the syntax of imperatives in Yemeni Arabic proposing an Agree acros phases (AAP) approach. I argue that the AAP approach successfully accounts for both simple and complex imperative constructions, including A'-chain structures, by establishing a close interactions between syntax and discourse. The study demonstrates that this interface is motivated by the interpretive and performative functions associated with imperatives, linking informational structure with propositional structure. It is also proposed that the thematic subject of imperatives is a 2-person pro, whereas any overt pronominal or nominal element occurring preverbally is not a subject, but rather a C-domain element, precisely aboutness topic. These topics serve as the logical subjects of imperatives and enter into a coreferentiality relationship with pro. This relation is analyzed as APP involving Match, yielding both local and non-local A'-chains. For core imperatives, viz., lacking an overt topic, I propose a null topic to (re)merge in Spec,TopP, whose interpretation depends on the discourse. --- ### Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks Authors: Pavel Prochazka Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08446v5 The standard training objectives of Bayesian deep learning are posterior-seeking: their optimum over the belief is the posterior of a fitted model, or its KL projection. We show that the shared target is a removable constraint on the belief-not an ideal that training can only approximate: posterior-seeking objectives define a binding map from model parameters to beliefs. From the local-normaliser structure of the Bethe/EP functional we derive a shared-cavity objective that carries this binding as an optional constraint, its per-observation data term a strictly proper predictive score for any likelihood. Our proposal is to drop the constraint. Free routing trains the belief as an optimisation variable of this objective; what trains on the predictive score is still a belief over weights, its prior and noise hyperparameters learned in the same gradient pass. We instantiate this at the Gaussian last layer, where exact inference is available: the freed belief departs from the posterior by a closed-form gap-the residual heteroscedasticity its variance family expresses. The instance, SCROLL, is single-pass, with no last-layer regularisation weight to cross-validate; it steps off the exact corner by a change of estimand (the shared cavity). SCROLL improves on the exact evidence corner at that corner's own learned features on seven of eight UCI datasets, matches or beats validation-tuned references on predictive likelihood and calibration-granted the ensembles' own five-member budget, it leads that tier on likelihood as well-and carries from UCI through frozen embeddings to end-to-end deep learning. --- ### Direct Bethe Free Energy Minimization for Bayesian Neural Networks Authors: Pavel Prochazka Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08446v4 Bayesian neural networks are typically trained on the evidence lower bound (ELBO), which keeps the joint likelihood but pays a Jensen gap at every observation. We train by local consistency instead: direct minimisation of the Bethe free energy, whose data term pays no gap-it scores each observation exactly by its predictive density, a strictly proper rule, for any likelihood with a tractable predictive convolution. Our departure is free routing: the beliefs are trained as free parameters of this objective, jointly with the backbone, rather than bound to the conjugate posterior computed in closed form (closed routing). Instantiated with a Gaussian last layer over a deterministic backbone, exact inference appears as the known, closed-routed corner-the neural-linear marginal likelihood, evidence-optimal, keeping the joint. A shared cavity instead trades it for a batchable per-plate predictive score, and free routing reaches that score's optimum-unattainable under the binding whenever the noise is heteroscedastic-improving NLL and calibration over the exact corner. This instance, SCROLL (Shared-Cavity fRee-rOuting Last-Layer), is a single-pass, any-likelihood Bayesian neural network, implicitly empirical-Bayes-prior precision, observation noise, covariance, and backbone fit in one gradient pass. At a single training run and forward pass per architecture-where the conventional references cross-validate their regularisation weight and ensembles pay 5-50x at inference-a fixed SCROLL variant is best-or-tied on NLL and calibration on 7/8 UCI regression benchmarks, and best on 4/5 across three large tabular datasets and two frozen text/vision embeddings. --- ### Measuring What Matters: Benchmarking Generative, Multimodal, and Agentic AI in Healthcare Authors: Prasanna Desikan, Harshit Rajgarhia, Shivali Dalmia, Ananya Mantravadi Published: 2026-05-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.08445v1 AI models are increasingly deployed in live clinical environments where they must perform reliably across complex, high-stakes workflows that standard training and validation datasets were never designed to capture. Evaluating these systems requires benchmarks: structured combinations of tasks, datasets, and metrics that enable reproducible, comparable measurement of what a model can do. The central challenge in healthcare AI is not performance alone, but the absence of systematic methods to measure reliability, safety, and clinical relevance under real-world conditions. Most existing benchmarks test what a model knows; too few test whether it can perform reliably and without failing across the full complexity of real clinical tasks. Current benchmarks have accumulated through ad hoc dataset construction optimized for narrow task performance: frontier models achieve near-perfect scores on medical licensing examinations, but when evaluated across real clinical tasks, performance degrades sharply, scoring 0.74--0.85 on documentation, 0.61--0.76 on clinical decision support, and only 0.53--0.63 on administrative and workflow tasks \cite{medhelm}. High benchmark scores give a false sense of deployment readiness, and the gap between performance and utility widens precisely as AI systems take on more consequential clinical roles. Without a principled framework for benchmark design, the field cannot determine whether poor clinical performance reflects model limitations or failures in how performance is being measured. --- ### Injection-Execution Dissociation: A Mechanistic Evaluation of Persistent Memory Attacks and Defenses in Stateful LLM Agents Authors: Jun Wen Leong Published: 2026-05-08 Categories: cs.CR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.08442v5 We discover that prompt-injection success and tool-execution success are separable safety properties: defenses that block injection do not necessarily block execution, and vice versa. We call this the injection-execution dissociation. In LLM agents with persistent memory, malicious instructions are stored at rates exceeding 97.5%, yet downstream execution ranges from 0% to 95% with no correlation to storage rate. This reframes the threat model: preventing storage alone is insufficient, and blocking execution requires structurally enforcing authority boundaries between memory ingestion and action execution. We substantiate this through a 5,040-run factorial experiment across nine open-source models (N=40 per condition), evaluating six defenses at four architectural layers against delayed-trigger attacks that persist across session boundaries via RAG retrieval. Defense effectiveness is governed by where a defense sits relative to the attack's authority boundary, not by classifier quality. Only Memory Sandbox -- a tool-layer defense that structurally isolates recalled memory from executable context -- reduces attack success to 0% for eight of nine models. A reasoning-mode ablation reveals a double dissociation: no single schema-layer intervention is safe across both reasoning and non-reasoning model classes. A loaded-corpus frontier evaluation (21 models, 3 providers; N=40 base, headline models topped up to N=172) reveals vendor-correlated patterns: Anthropic blocks predominantly at injection, OpenAI blocks at execution with variable generational hardening, and a pre-release Gemini endpoint exfiltrates in the majority of runs. Stored-but-dormant payloads constitute a compositional supply-chain risk in shared-memory deployments. --- ### DUET: Optimize Token-Budget Allocation for Reinforcement Learning with Verifiable Rewards Authors: Haoyu Hu, Xuandong Zhao, Xuhai "Orson'' Xu, Nori Jacoby Published: 2026-05-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08441v1 Reinforcement learning with verifiable rewards (RLVR) generates hundreds of thousands of tokens per training step, with rollout generation dominating the computational cost. The overall token budget can be controlled along two main dimensions: (i) deciding which prompts to allocate rollouts to, and (ii) deciding how long each rollout should be. Prior work has generally controlled only one of these dimensions at a time. We show that jointly tuning both decisions under a shared compute budget improves both reasoning quality and wall-clock training time. We instantiate this view as \textbf{DU}al-controlled tok\textbf{E}n alloca\textbf{T}ion (DUET), a computationally efficient layer over GRPO that uses a lightweight pre-rollout surrogate of prompt informativeness to set how many rollouts each prompt receives, and a marker-gated abort rule with importance reweighting to set when to stop them. On Qwen3-1.7B trained on MATH, DUET outperforms full-budget GRPO and the other three budget-aware baseline methods. DUET's advantage further generalizes to other benchmarks across math and coding, and is on par with the best baseline on the scientific Q\&A domain, while also achieving a $1.62\times$ wall-clock speedup. More notably, using only 50\% of the token budget, DUET still outperforms all baseline methods at their full budget, achieving an even higher $2.51\times$ speedup over full-budget GRPO. We verify the high performance of DUET on other backbone LLMs, including Qwen3-4B and Llama-3.2-3B-Instruct. Notably, the gap between DUET and the strongest baseline \emph{widens} as the budget tightens, contrary to the usual pattern in which efficient methods trade off quality as compute decreases. More broadly, these results suggest that DUET budget-aware control strategies are valuable not only for accelerating training, but also for improving the quality of the learning signal. --- ### TARO: Temporal Adversarial Rectification Optimization Using Diffusion Models as Purifiers Authors: Daniel Wesego, Pedram Rooshenas Published: 2026-05-08 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2605.08440v1 Adversarial purification with diffusion models seeks to project adversarial examples back toward the data manifold, but balancing semantic preservation and robustness against adaptive attacks remains challenging. Recent work shows that standard diffusion purification can fail under adaptive evaluation, while test-time score-based optimization is more resilient. Existing optimization defenses, however, typically rely on a single diffusion noise regime or treat timesteps uniformly, overlooking the distinct roles of coarse and fine denoising scales. We propose Temporal Adversarial Rectification Optimization (TARO), an inference-time purification method that builds a temporally guided score prior from multiple denoising views along the diffusion trajectory. TARO forms a coarse-to-fine residual target: high-noise experts provide globally smoothed structure with reduced adversarial sensitivity, while low-noise experts restore image-specific, class-relevant details. A guidance strength controls this temporal correction, allowing TARO to balance robust global rectification with semantic preservation. Empirically, TARO improves robust accuracy across datasets and adaptive threat models in a zero-shot setting, while remaining compatible with complementary adversarial-likelihood objectives for further robustness gains. --- ### Can Language Models Identify Side Effects of Breast Cancer Radiation Treatments? Authors: Natalie Seah, Danielle S. Bitterman, Daphna Spiegel, Thomas Hartvigsen Published: 2026-05-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2605.08439v2 Accurately communicating the side effects of cancer treatments to cancer survivors is critical, particularly in settings such as informed consent, where clinicians must clearly and comprehensively convey potential treatment toxicities. However, this task remains challenging due to clinical knowledge deficits about adverse treatment effects and fragmentation across electronic health record (EHR) systems. Large language models (LLMs) have the potential to assist in this task, though their reliability in oncology survivorship contexts remains poorly understood. We present a deployment-oriented stress-testing framework for evaluating LLM-generated radiation side effect lists in breast cancer treatment and survivorship care. Using 21 breast cancer patient profiles, we construct paired patient clinical scenarios that differ only in radiotherapy regimens to evaluate seven instruction-tuned LLMs under multiple prompting regimes. We then compare LLM outputs to a clinician-curated reference derived from informed consent documents at two major academic medical centers and developed by a team including more than seven breast radiation oncologists. The reference maps radiation dose-fractionation, fields, and locations to associated toxicities, broken down by frequency and temporal onset. Across models, we reveal sensitivity to minor documentation changes, trade-offs between precision and recall, and systematic under-recall of rare and long-term side effects. When used alone, constraints on the number of side effects generated reduce precision, and grounding outputs in clinician-curated side effect lists substantially improves reliability and robustness. These findings highlight important limitations of LLM use in oncology and suggest practical design choices for safer and more informative survivorship-focused applications. --- ### Magis-Bench: Evaluating LLMs on Magistrate-Level Legal Tasks Authors: Ramon Pires, Thales Sales Almeida, Celio Larcher Junior, Giovana Bonás, Hugo Abonizio, Marcos Piau, Roseval Malaquias Junior, Thiago Laitz, Rodrigo Nogueira Published: 2026-05-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.08437v1 Existing benchmarks for legal AI focus primarily on tasks where LLMs must produce legal arguments or documents, yet the capacity to \emph{judge} such arguments -- weighing competing claims, applying doctrine to facts, and rendering reasoned decisions -- is arguably as fundamental to a well-functioning legal system as advocacy itself. We introduce Magis-Bench, a benchmark for evaluating LLMs on magistrate-level writing tasks derived from recent Brazilian competitive examinations for judicial positions. Magis-Bench comprises 74 questions from eight examinations conducted between 2023 and 2025, including discursive legal analysis questions with multi-turn structure and practical exercises requiring the composition of complete civil and criminal judicial sentences. We evaluate 23 state-of-the-art LLMs using an LLM-as-a-judge methodology with four independent frontier models as evaluators. Our results show strong inter-judge agreement (Kendall's $W = 0.984$; pairwise Kendall's $τ\ge 0.897$), with Google's Gemini-3-Pro-Preview achieving the highest average score (6.97/10), followed by Gemini-3-Flash-Preview (6.67) and Claude-4.5-Opus (6.46). Even the best-performing models score below 70\% of the maximum, indicating that judicial-level legal reasoning and writing remain challenging for current LLMs. We release the complete benchmark, model outputs, and evaluation code to support further research on legal AI capabilities. --- ### A meshfree exterior calculus for generalizable and data-efficient learning of physics from point clouds Authors: Benjamin D. Shaffer, Brooks Kinch, M. Ani Hsieh, Nathaniel Trask Published: 2026-05-08 Categories: cs.LG, cs.AI, physics.comp-ph arXiv: https://arxiv.org/abs/2605.08436v1 We introduce a meshfree exterior calculus (MEEC) for learning structure-preserving descriptions of physics on point clouds, and use it to build MEEC-Net, a data-efficient surrogate that transfers across resolutions, geometries, and physical parameters. MEEC equips an $\varepsilon$-ball graph with virtual node and edge measures via a single sparse Schur complement solve; the resulting complex satisfies discrete conservation exactly, is end-to-end differentiable in the point positions, and exposes a direct geometry-to-physics link without the mesh-generation step required by conventional structure-preserving discretizations. MEEC-Net learns unknown physics as a shared edge-wise flux law in an SO($d$)-invariant local frame, so the same kernel produces compatible fluxes on any point cloud whose features lie in the training range. We prove a solution-error bound that splits into discretization and kernel-approximation terms which is independent of problem geometry, explaining the observed transfer from very few examples. We show that single-solution training transfers to unseen geometries, boundary conditions, and physical parameters. On five canonical PDE benchmarks MEEC-Net achieves 1-2 orders of magnitude lower out-of-distribution error than baseline neural-operator approaches. On the SimJEB structural-bracket benchmark it achieves competitive error while using substantially fewer training geometries. --- ### A Semantic-Sampling Framework for Evaluating Calibration in Open-Ended Question Answering Authors: Zhanliang Wang, Jiancong Xiao, Ruochen Jin, Shu Yang, Bojian Hou, Li Shen Published: 2026-05-08 Categories: cs.CL, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2605.08432v1 Calibration measures whether a model's predicted confidence aligns with its empirical accuracy, and is central to the reliable deployment of large language models (LLMs) in high-stakes domains such as medicine and law. While much recent work focuses on improving LLM calibration, the equally important question of how to evaluate it in realistic settings remains underdeveloped. Open-ended question answering (QA), the most common deployment setting for modern LLMs, is where existing evaluation methods fall short: logit-based metrics need restricted output formats and internal probabilities; verbalized confidence is self-reported and often overconfident; and sampling-based methods rely on task-specific extraction rules without a clear finite-sample target. We introduce Sem-ECE (Semantic-Sampling Expected Calibration Error), a calibration evaluation framework for open-ended QA that samples answers from the model, groups them into semantic classes, and uses the resulting frequencies as confidence. We study two estimators within this framework: Sem$_1$-ECE, the same-sample self-consistency score, and Sem$_2$-ECE, a held-out variant that separates answer selection from confidence evaluation. We prove both are asymptotically unbiased, and further show that they agree on easy questions but diverge on hard ones with Sem$_2$ achieving strictly smaller calibration error, so their gap also serves as a diagnostic for question difficulty. Experiments on three open-ended QA benchmarks across five leading commercial LLMs match our theoretical predictions and show that Sem-ECE outperforms verbalized confidence and existing sampling-based methods, while complementing logit-based evaluation when internal probabilities are unavailable. --- ### Active Multiple-Prediction-Powered Inference Authors: Nicholas Brawand, Nima Leclerc, Anhthy Ngo, Matthew Peterson, Sriram Vishwanath, Laith Alhussein, Ben Wellner Published: 2026-05-08 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2605.08429v1 Post-deployment monitoring of healthcare AI requires statistically valid, label-efficient methods, but gold-standard labels from clinician chart review are expensive. Prediction-powered inference (PPI) and active statistical inference (ASI) reduce label cost by combining a small labeled sample with abundant model predictions, but both are restricted to a single predictor, a poor fit for modern clinical pipelines that have multiple predictors of differing cost and accuracy available at inference time. We propose Active Multiple-Prediction-Powered Inference (AM-PPI), which routes each instance to a cost-appropriate predictor subset, samples gold-standard labels in proportion to the chosen subset's residual uncertainty, and reweights predictions to minimize estimator variance, all under a single deployment-time budget. AM-PPI generalizes ASI to leverage multiple predictors and extends Multiple-PPI from global per-predictor allocation to per-instance adaptive routing. We derive closed-form Karush-Kuhn-Tucker (KKT) conditions for all three decisions and prove, via biconvexity and strong duality, that the resulting fixed point is a global optimum despite the joint problem being non-jointly-convex. We establish asymptotic normality with valid coverage, minimum-variance unbiasedness within the linear-prediction augmented inverse propensity weighted (AIPW) class, and a closed-form criterion identifying when multiple predictors help. On synthetic data and three healthcare monitoring tasks, AM-PPI produces 10 to 40 percent narrower confidence intervals (CIs) than single-predictor ASI in the budget regime where routing matters, and matches the better baseline elsewhere. --- ### The Attacker in the Mirror: Breaking Self-Consistency in Safety via Anchored Bipolicy Self-Play Authors: Gabriele La Malfa, Emanuele La Malfa, Saar Cohen, Jie M. Zhang, Michael Luck, Michael Wooldridge, Elizabeth Black Published: 2026-05-08 Categories: cs.AI, cs.GT, cs.LG arXiv: https://arxiv.org/abs/2605.08427v1 Self-play red team is an established approach to improving AI safety in which different instances of the same model play attacker and defender roles in a zero-sum game, i.e., where the attacker tries to jailbreak the defender; if self-play converges to a Nash equilibrium, the model is guaranteed to respond safely within the settings of the game. Although the parameter sharing enforced by the use of the same model for the two roles improves stability and performance, it introduces fundamental theoretical and architectural limitations. We show that the set of Nash equilibria that can be reached corresponds to a broad class of behaviours that includes trivial always refuse strategies and oracle-like defenders, thus limiting practical applicability. We then show that when attacker and defender share and update the same base model, the dynamics collapse to self-consistency, so that attacks do not enforce adversarial pressure on the defender. In response, we propose Anchored Bipolicy Self-Play, which trains distinct role-specific LoRA adapters on top of a frozen base model, thereby maintaining stable optimisation while preserving adversarial pressure through explicit role separation. In relation to standard self-play, we show up to 100x greater parameter efficiency than finetuning and consistent improvements in safety compared to self-play fine-tuned models. We evaluate on Qwen2.5-{3B, 7B,14B}-IT models across widely used safety benchmarks, showing improved robustness without loss of reasoning ability. Cross-play experiments further show that our attacker and defender models are superior to self-play in terms of adversarial defence and safety. --- ### Mechanism Design Is Not Enough: Prosocial Agents for Cooperative AI Authors: Xuanqiang Angelo Huang, Charlie Tharas, Samuele Marro, Van Q. Truong, Bernhard Schölkopf, Emanuele La Malfa, Zhijing Jin Published: 2026-05-08 Categories: cs.GT, cs.AI arXiv: https://arxiv.org/abs/2605.08426v3 Ensuring that AI agents behave safely and beneficially when interacting with other parties has emerged as one of the central challenges of modern AI safety. While mechanism design, as the theory of designing rules to align individual and collective objectives, can incentivize cooperative behavior, it is still an open question whether it alone is sufficient to maximize LLM agents' social welfare. This work proves that the answer is negative: drawing from incomplete contract theory, we formally show that when contracts cannot distinguish all relevant future contingencies, there is a strictly positive welfare loss that no realistic mechanism can eliminate. We show that prosocial agents, who weigh others' welfare alongside their own, can close this gap and achieve outcomes that are socially superior and individually beneficial. Experimentally, we show that in multi-agent resource-allocation environments and canonical social dilemmas where agents are powered by large language models, prosociality is beneficial. The implication for AI safety is clear: to enable cooperative interactions at scale, designing adequate mechanisms is not sufficient; agents must be built to be intrinsically prosocial. --- ### Mechanism Design Is Not Enough: Prosocial Agents for Cooperative AI Authors: Xuanqiang Angelo Huang, Charlie Tharas, Samuele Marro, Van Q. Truong, Bernhard Schölkopf, Emanuele La Malfa, Zhijing Jin Published: 2026-05-08 Categories: cs.GT, cs.AI arXiv: https://arxiv.org/abs/2605.08426v2 Ensuring that AI agents behave safely and beneficially when interacting with other parties has emerged as one of the central challenges of modern AI safety. While mechanism design, as the theory of designing rules to align individual and collective objectives, can incentivize cooperative behavior, it is still an open question whether it alone is sufficient to maximize LLM agents' social welfare. This work proves that the answer is negative: drawing from incomplete contract theory, we formally show that when contracts cannot distinguish all relevant future contingencies, there is a strictly positive welfare loss that no realistic mechanism can eliminate. We show that prosocial agents, who weigh others' welfare alongside their own, can close this gap and achieve outcomes that are socially superior and individually beneficial. Experimentally, we show that in multi-agent resource-allocation environments and canonical social dilemmas where agents are powered by large language models, prosociality is beneficial. The implication for AI safety is clear: to enable cooperative interactions at scale, designing adequate mechanisms is not sufficient; agents must be built to be intrinsically prosocial. --- ### Generalized Wasserstein Flow Matching: Transport Plans, Everywhere, All at Once Authors: Moritz Piening, Richard Duong, Gabriele Steidl Published: 2026-05-08 Categories: cs.LG, math.OC, math.PR arXiv: https://arxiv.org/abs/2605.08424v1 Flow matching has recently emerged as a flexible and efficient framework for generative modelling by learning deterministic transport dynamics between probability measures. In this work, we extend flow matching to the space of probability measures over probability measures, introducing a Wasserstein-on-Wasserstein (WoW) formulation. Leveraging the nested Wasserstein geometry, we show that measures over transport plans naturally induce velocity fields that realize metameasure flows. This yields a principled generalization of Wasserstein flow matching via coupled outer and inner transport plans. To address the substantial computational cost of WoW transport, we propose scalable approximations based on sliced and linear Wasserstein distances, enabling efficient training while promoting numerically stable, near-straight trajectories. Our framework unifies and extends existing approaches to point cloud and set generation, providing a practical and theoretically grounded method for generative modelling in WoW spaces. --- ### Queryable LoRA: Instruction-Regularized Routing Over Shared Low-Rank Update Atoms Authors: Omatharv Bharat Vaidya, Connor T. Jerzak, Nhat Ho, Chandrajit Bajaj Published: 2026-05-08 Categories: cs.LG, cs.CL, stat.ML arXiv: https://arxiv.org/abs/2605.08423v1 We present a data-adaptive method for parameter-efficient fine-tuning of large neural networks. Standard low-rank adaptation methods improve efficiency by restricting each layer update to a fixed low-rank form, but this static parameterization can be too rigid when the appropriate correction depends on the input and on the evolving depth-wise computation of the network. Our approach replaces a purely layer-local adapter with a shared queryable memory of low-rank update atoms. For each block of layers, the model forms a query from the current low-rank state and a running summary of previous blocks, uses this query to retrieve a content-dependent combination of shared update components via attention, and applies the resulting routed operator within the low-rank bottleneck. In this way, the method retains the efficiency and scalability of low-rank adaptation while allowing the effective update to vary across inputs and to share reusable structure across layers. The resulting architecture provides a principled middle ground between static LoRA-style updates and fully generated parameter updates: it remains compact and parameter-efficient while supporting dynamic, context-sensitive adaptation. Further, we incorporate instruction-regularization by augmenting routing logits with a language-induced prior over update atoms, thereby biasing the selection of low-rank transformations toward semantically relevant directions without generating unconstrained parameter updates. Experiments on noisy non-linear regression tasks and LLM fine-tuning suggest that this queryable update-memory formulation can improve final test performance and training stability compared to standard low-rank adaptation, while using a comparable number of trainable parameters. --- ### Beyond Bag-of-Patches: Learning Global Layout via Textual Supervision for Late-Interaction Visual Document Retrieval Authors: Pascal Tilli, Mohsen Mesgar Published: 2026-05-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.08421v1 Visual Document Retrieval (VDR) models mostly rely on late interaction architectures, in which documents are represented by a set of local patch embeddings and then matched against query tokens. While efficient, this architecture prioritizes local similarity over global layout structure of documents to estimate relevancy between documents and query. In practice, this leads to errors as relevance originates from layout structure of documents with heterogeneous layouts combining figures, tables, and text. We make document layout learnable without changing inference. We propose a multimodal encoder that augments local patch representations with a global layout embedding, trained via textual descriptions encoding document layout information. Across four ViDoRe-v2 datasets, our model improves over the strongest architecturally comparable ColPali/ColQwen baseline by +2.4 nDCG@5 and +2.3 MAP@5, with statistically significant per-dataset gains over ColQwen. --- ### Central Limit Theorem for Two-Time-Scale Approximate Distributionally Robust RL Authors: Shengbo Wang, Zexi Zhang Published: 2026-05-08 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2605.08417v1 Designing model-free algorithms for distributionally robust reinforcement learning (DRRL) poses fundamental challenges. The robust Bellman operator is nonlinear in the transition kernel, which makes one-sample Bellman updates biased, while the adversarial optimization underlying robustness makes robust evaluation computationally demanding. To address these difficulties, we consider the natural small-ambiguity regime under Kullback--Leibler ambiguity sets and propose an approximate DRRL framework based on a first-order expansion of the relevant robust functional. This yields an approximate robust Bellman equation that removes the adversarial optimization while remaining first-order accurate in the ambiguity radius. To learn the fixed point of this approximate equation, we propose Mean-Variance Stochastic Approximation (MVSA), a model-free algorithm that uses only one-sample updates. This is achieved via a lifted stochastic approximation dynamics and a two-time-scale design. We then prove convergence and a central limit theorem for MVSA: its main iterate satisfies a central limit theorem at the canonical $n^{-1/2}$ scale, with explicitly characterized asymptotic covariances. Finally, we validate our theoretical findings with a numerical experiment. --- ### Alignment as Jurisprudence Authors: Nicholas Caputo Published: 2026-05-08 Categories: cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2605.08416v1 Jurisprudence, the study of how judges should properly decide cases, and alignment, the science of getting AI models to conform to human values, share a fundamental structure. These seemingly distant fields both seek to predict and shape how decisions by powerful actors, in one case judges and in the other increasingly powerful artificial intelligences, will be made in the unknown future. And they use similar tools of the specification and interpretation of language to try to accomplish those goals. The great debates of jurisprudence, about what the law is and what it should be, can provide insight into alignment, and lessons from what does and does not work in alignment can help make progress in jurisprudence. This essay puts the two fields directly into conversation. Drawing on leading accounts of jurisprudence, particularly Dworkin's principle-oriented interpretivism and Sunstein's positivist account of law as analogical reasoning, and on cutting-edge alignment approaches, namely Constitutional AI and case-based reasoning, it illustrates the value of a more sophisticated legally-inspired approach to the interplay of rules and cases in finetuning alignment and points to ways that AI can provide a better understanding of how the law works and how it can be improved by the introduction of AI. AI systems and the law should operate to empower people to act in the world, helping to expand their capabilities and the extent to which they are able to achieve their goals. As AI continues to improve in capacity, and as the constraints that legal theory places on human judges seem be coming undone, the conversation between these two fields will become increasingly essential and may help point to a better version of both. --- ### Political Plasticity: An Analysis of Ideological Adaptability in Large Language Models Authors: Bruno Bianchi, Diego Tiscornia, Matias Travizano, Ariel Futoransky Published: 2026-05-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.08415v1 Since the advent of Large Language Models (LLMs), a significant area of research has focused on their intrinsic biases, particularly in political discourse. This study investigates a different but related concept, "political plasticity", which is defined as the capacity of models to adapt their responses based on the user supplied context. To analyze this, a testing framework was developed using an expanded corpus of 200 politically-oriented questions across economic and personal freedom axes, based on a prior framework by Lester (1996). The study explored several methods to induce political bias, including simplified and topic-based system prompts, as well as user prompts with few-shot examples. The results show that while system prompts were largely ineffective, user prompts successfully elicited significant ideological shifts, particularly along the Economic Freedom axis in larger and newer models. Through a validation experiment, we examined whether models answer questionnaires by recognizing the underlying question format. Inverting the sense of the questions revealed unexpected, counter-intuitive shifts in most models, suggesting potential data leakage. Finally, we also analyzed how model plasticity varies when the experiment is conducted in different languages. The results reveal subtle yet notable shifts across each of the analyzed languages. Overall, our results indicate that small and older LLMs exhibit limited or unstable political plasticity, whereas newer frontier models display reliable, expected adaptability. --- ### SYNCR: A Cross-Video Reasoning Benchmark with Synthetic Grounding Authors: Sara Ghazanfari, Siddharth Garg, Prashanth Krishnamurthy, Farshad Khorrami Published: 2026-05-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.08412v1 Multimodal Large Language Models (MLLMs) have made rapid progress in single-video understanding, yet their ability to reason across multiple independent video streams remains poorly understood. Existing multi-video benchmarks rely largely on human-annotated real-world footage, limiting the precision of spatial, temporal, and physical ground truth and making it difficult to diagnose model failures. We introduce SYNCR, a controlled synthetic benchmark for cross-video reasoning with programmatically verified grounding. Built using Habitat, Kubric, and CLEVRER simulator engines, SYNCR contains 8,163 multi-video question-answer pairs grounded in 9,650 unique videos. It evaluates MLLMs across eight tasks spanning four diagnostic pillars: Temporal Alignment, Spatial Tracking, Comparative Reasoning, and Holistic Synthesis. Our zero-shot evaluation of leading open- and closed-weight MLLMs reveals a substantial gap between current models and humans: the best model achieves only 52.5% average accuracy, compared to an 89.5% human baseline. Models perform relatively well on temporal ordering but struggle with precise physical and spatial reasoning, with the best model reaching only 26.0% accuracy on Kinematic Comparison. We further find that parameter scaling and reasoning-specialized post-training improve temporal alignment capabilities, but do not reliably address fine-grained physical tracking or global spatial synthesis. Finally, an exploratory sim-to-real correlation analysis suggests that several SYNCR tasks track model-level trends on real-world multi-video benchmarks, while also exposing reasoning capabilities underrepresented by existing evaluations. Code available at https://github.com/SaraGhazanfari/SYNCR. --- ### Playing games with knowledge: AI-Induced delusions need game theoretic interventions Authors: Will Beaumaster, Paul Schrater Published: 2026-05-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.08409v1 Conversational AI has a fundamental flaw as a knowledge interface: sycophantic chatbots induce epistemic entrenchment and delusional belief spirals even in rational agents. We propose the problem does not stem from the AI model, rooted instead in a systemic consequence of the paradigm shift from user-driven knowledge search to users and agents engaged in strategic, repeated-play communication. We formalize the problem as a Crawford-Sobel cheap talk game, where costless user signals induce a pooling equilibrium. Agents optimized for user satisfaction produce sycophantic strategies that provide identical reinforcement across user types with opposite epistemic incentives: exploratory ``Growth-seekers'' ($θ_G$) and confirmatory ``Validation-seekers'' ($θ_V$). Under repeated play, this identification failure creates a coordination trap -- analogous to a Prisoner's Dilemma -- where locally rational feedback loops drive users toward pathologically certain false beliefs. We propose an inference-time mechanism design intervention called an Epistemic Mediator that breaks this pooling equilibrium by introducing a costly signal (epistemic friction), forcing type revelation based on users' asymmetric cognitive costs for processing resistance. A key contribution is Belief Versioning, a git-inspired epistemic meta-memory system that stores healthy beliefs and rollbacks when validation-seeking resistance is detected. In simulation, this intervention achieves a separating equilibrium achieving a $48\times$ differential in spiral rates while passing a learning preservation criterion), evidence that epistemic safety in AI is fundamentally a problem of strategic information environment design rather than simple model alignment. --- ### AdamFLIP: Adaptive Momentum Feedback Linearization Optimization for Hard Constrained PINN Training Authors: Binghang Lu, Runyu Zhang, Changhong Mou, Na Li, Guang Lin Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08408v1 Physics-informed neural networks (PINNs) provide a flexible framework for solving forward and inverse problems governed by partial differential equations (PDEs), but standard PINN training typically relies on soft penalty formulations that combine PDE residuals, data mismatch, and initial/boundary conditions using manually chosen weights. This often leads to ill-conditioning, sensitivity to loss weights, and poor constraint satisfaction. In this work, we reformulate PINN training as an equality-constrained optimization problem and propose a novel Adaptive Momentum Feedback Linearization Optimization for Hard Constrained PINN (AdamFLIP). The key idea is to view the constraint residuals as the output of a controlled dynamical system and to compute the Lagrange multiplier as a feedback input that locally drives these residuals toward stable linear contraction dynamics. AdamFLIP then applies Adam-style first- and second-moment adaptation to the resulting feedback-linearized Lagrangian gradient, combining principled constraint handling with the scalability and robustness of adaptive neural-network optimization. We test AdamFLIP on a range of benchmark forward and inverse PDE problem, and it consistently outperforms both the standard soft-constrained PINN and state-of-the-art constrained optimizers. Specifically, on the Navier--Stokes equations benchmark, AdamFLIP \textbf{reduces relative $L_2$ error by more than two thirds} for the predicted solution compared to the next best method. Our AdamFLIP framework provides an effective and computationally scalable hard constraint optimization method for PINN training. --- ### Effective Explanations Support Planning Under Uncertainty Authors: Hanqi Zhou, Britt Besch, Charley M. Wu, Tobias Gerstenberg Published: 2026-05-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.08406v1 Explaining how to get from A to B can be challenging. It requires mentally simulating what the listener will do based on what they are told. To capture this process, we propose a computational model that converts utterances into action plans: a large language model translates an explanation into program-like guidance (a policy prior and value map), and a planning agent executes it under partial observability. We score explanations by the efficiency and reliability of the resulting paths, penalizing replanning. Across four preregistered experiments, we collect a corpus of 1,200 explanations over 24 maps, elicit helpfulness judgments, measure baseline navigation, and test behavior with explanations of differing quality. Higher-scored explanations are judged more helpful and improve navigation: participants with explanations outperform those without, and high-scoring explanations help more than low-scoring ones. Together, these results show procedural explanation as utility-guided communication shaped by how language can be grounded into action under uncertainty. --- ### Belief or Circuitry? Causal Evidence for In-Context Graph Learning Authors: Katharine Kowalyshyn, Timothy Duggan, Daniel Little, Michael C Hughes Published: 2026-05-08 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.08405v1 How do LLMs learn in-context? Is it by pattern-matching recent tokens, or by inferring latent structure? We probe this question using a toy graph random-walk across two competing graph structures. This task's answer is, in principle, decidable: either the model tracks global topology, or it copies local transitions. We present two lines of evidence that neither account alone is sufficient. First, reconstructing the internal representation structure via PCA reveals that at intermediate mixture ratios, both graph topologies are encoded in orthogonal principal subspaces simultaneously. This pattern is difficult to reconcile with purely local transition copying. Second, residual-stream activation patching and graph-difference steering causally intervene on this graph-family signal: late-layer patching almost fully transfers the clean graph preference, while linear steering moves predictions in the intended direction and fails under norm-matched and label-shuffled controls. Taken together, our findings are most consistent with a dual-mechanism account in which genuine structure inference and induction circuits operate in parallel. --- ### Built Environment Reasoning from Remote Sensing Imagery Using Large Vision--Language Models Authors: Dongdong Wang, Deepak Balakrishnan, Ravi Srinivasan, Shenhao Wang Published: 2026-05-08 Categories: cs.CL, cs.AI, cs.CV, cs.ET arXiv: https://arxiv.org/abs/2605.08404v1 This work investigates the use of large language models (LLMs) for tasks in smart cities. The core idea is to leverage remote sensing imagery to characterize the built environment, including design suggestions, constructability assessment, landuse patterns, and risk identification. We examine remote sensing imagery at multiple spatial scales as inputs for multimodal language modeling and evaluate their effects on built-environment-related reasoning. In addition, we compare state-of-the-art LLMs, including InternVL and Qwen, in terms of accuracy and reliability when generating built environment recommendations. The results demonstrate the potential of integrating remote sensing imagery with large language models to assist smart cities and decision-making. --- ### AIPO: Learning to Reason from Active Interaction Authors: Junnan Liu, Linhao Luo, Thuy-Trang Vu, Gholamreza Haffari Published: 2026-05-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.08401v3 Recent advances in LLMs have demonstrated strong reasoning capabilities, largely stimulated by RLVR. However, the exploration of existing RLVR algorithms remains largely constrained by the knowledge and reasoning strategies already accessible to the policy model. Although recent methods introduce external expert demonstrations to broaden exploration, they typically rely on complete trajectory-level guidance, which can be sample-inefficient, information-sparse, and insufficiently adaptive to intermediate reasoning bottlenecks. Inspired by collaborative multi-agent systems, we propose AIPO, an enhanced reinforcement learning framework that improves LLM reasoning through active multi-agent interaction during exploration. Specifically, when encountering reasoning bottlenecks, AIPO enables the policy model to proactively consult three functional collaborative agents, namely the verify agent, knowledge agent, and reasoning agent, thereby obtaining fine-grained and state-dependent guidance during rollout. The resulting mixed-policy trajectories expose the policy to reasoning directions that may be difficult to discover through isolated on-policy exploration. To learn effectively from collaborator-provided tokens, we further introduce a corrected importance sampling coefficient together with a lower-bound clipping strategy to mitigate off-policy discrepancy and vanishing gradients. After training, the policy model reasons independently without relying on collaborative agents. Extensive experiments across mathematical, scientific, coding, and puzzle reasoning benchmarks show that AIPO consistently improves reasoning performance and generalizes across different policy models, collaborator backbones, and RLVR algorithms. --- ### AIPO: Learning to Reason from Active Interaction Authors: Junnan Liu, Linhao Luo, Thuy-Trang Vu, Gholamreza Haffari Published: 2026-05-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2605.08401v2 Recent advances in large language models (LLMs) have demonstrated remarkable reasoning capabilities, largely stimulated by Reinforcement Learning with Verifiable Rewards (RLVR). However, existing RL algorithms face a fundamental limitation: their exploration remains largely constrained by the inherent capability boundary of the policy model. Although recent methods introduce external expert demonstrations to extend this boundary, they typically rely on complete trajectory-level guidance, which is sample-inefficient, information-sparse, and may confine exploration to a static guidance space. Inspired by the potential of multi-agent systems, we propose $\textbf{AIPO}$, an enhanced reinforcement learning framework that improves LLM reasoning through active multi-agent interaction during exploration. Specifically, AIPO enables the policy model to proactively consult three functional collaborative agents, $\textit{Verify Agent}$, $\textit{Knowledge Agent}$, and $\textit{Reasoning Agent}$, when encountering reasoning bottlenecks, thereby receiving fine-grained and targeted guidance to actively expand its capability boundary during training. We further introduce a tailored importance sampling coefficient together with a clipping strategy to mitigate the off-policy bias and gradient vanishing issues that arise when learning from agent-provided feedback. After training, the policy model performs reasoning independently without relying on collaborative agents. Extensive experiments on diverse reasoning benchmarks, including AIME, MATH500, GPQA-Diamond, and LiveCodeBench, show that AIPO consistently improves reasoning performance, generalizes robustly across different policy models and RLVR algorithms, and effectively expands the reasoning capability boundary of the policy model. --- ### On Observation Time for Recovering Latent Hawkes Networks Authors: Jonas Linkerhägner, Michele Bortolasi, Lorenzo Baldassari, Maarten V. de Hoop, Ivan Dokmanić Published: 2026-05-08 Categories: math.ST, cs.IT, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2605.08400v1 Dynamics of interacting systems in engineering, society, and nature often evolve over latent networks that govern which entities can interact. We study the problem of inferring these networks from event-based observations, which arise naturally in finance, seismology, and neuroscience. While there is substantial algorithmic work addressing this important problem, theoretical results are scarce. In this paper we ask the following fundamental question: what is the minimum time that one must observe the dynamics in order to exactly recover the underlying network, as a function of the number $d$ of interacting entities? For a class of stationary Hawkes processes with sparse, weak interactions, we prove that an observation time of order $\log d$ is sufficient and necessary. For the upper bound we construct a two-stage estimator that uses clipped and binned event data for screening, followed by a least-squares refinement, and apply concentration bounds derived from the Poisson cluster representation. For the lower bound we combine Fano's inequality with Jacod's Girsanov formula for point processes on a suitable subclass of networks. --- ### CoCoDA: Co-evolving Compositional DAG for Tool-Augmented Agents Authors: Ziyang Yu, Qiyue Li, Liang Zhao Published: 2026-05-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.08399v1 Tool-augmented language models can extend small language models with external executable skills, but scaling the tool library creates a coupled challenge: the library must evolve with the planner as new reusable subroutines emerge, while retrieval from the growing library must remain within a fixed context budget. Existing tool-use and skill-library methods typically treat tools as flat or text-indexed memories, causing prompt cost to grow with library size and obscuring the typed, compositional structure of executable code. We propose CoCoDA, a framework that co-evolves the planner and tool library through a single code-native structure: a compositional code DAG. Nodes are primitive or composite tools, edges encode invocation dependencies, and each node stores a typed signature, description, pre/post-condition specification, and worked examples. At inference time, Typed DAG Retrieval prunes candidates by symbolic signature unification, ranks survivors by descriptions, filters them by behavioral specifications, and disambiguates with examples, keeping expensive context materialization on progressively smaller candidate sets. At training time, successful trajectories are folded into validated composite tools, while the planner is updated with a DAG-induced reward that credits composites by their primitive expansion size. We provide theoretical results showing retrieval cost reduction, sublinear retrieval time, compositional advantage under the shaped reward, monotone co-evolution under conservative updates, and DAG well-formedness. Across mathematical reasoning, tabular analysis, and code task benchmarks, CoCoDA enables an 8B student to match or exceed a 32B teacher on GSM8K and MATH and consistently improves over strong tool-use and library-learning baselines. --- ### Exploring and Exploiting Stability in Latent Flow Matching Authors: Rania Briq, Michael Kamp, Ohad Fried, Sarel Cohen, Stefan Kesselheim Published: 2026-05-08 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2605.08398v2 In this work, we show that Latent Flow-Matching (LFM) models are robust to different types of perturbations, including data reduction and model capacity shrinkage. We characterize this stability by these models' tendency to generate similar outputs under identical noise seeds. We provide a perspective relating this phenomenon to flow matching theory, which indicates that this stability is inherent to the FM objective. We further exploit this stability to derive practical algorithms for more efficient training and inference. Concretely, first, we show that by training LFM models on significantly reduced datasets, performance is preserved, and in compute-constrained regimes, the model converges faster while maintaining quality. This yields multiple advantages, including savings in the training time due to faster convergence, and alleviating annotation effort when training conditional models. Second, LFM stability under architectural shrinkage gives rise to a two-model coarse-to-fine approach, one using a light-weight architecture for the first phase of the FM trajectory, and one with higher capacity for the second, thereby reducing the inference cost substantially. To determine which samples are informative, we introduce three sample-scoring criteria and evaluate them under standard metrics for generative models. Our results are thoroughly evaluated on multiple datasets, demonstrating the practical advantage of this stability, including data savings and a more than two-fold inference speedup while generating comparable outputs. --- ### Delivering Science as a Service: Sci-Orchestra's Cloud-Native Approach to HPC Authors: Harinarayan Krishnan, Shubhabrata Mukerjee, Jeffrey Donatelli, Daniela Ushizima Published: 2026-05-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.08396v1 The increasing complexity of modern computational environments often burdens researchers with infrastructure management, authentication protocols, and container deployments. We present Sci-Orchestra, a layered orchestration framework designed to fully automate experimental workflows, allowing scientists to prioritize scientific discovery over backend operations. By abstracting execution through an API-driven interface, the system assumes responsibility for secure authentication, resource management, and scalable deployment across diverse high-performance computing environments using Kubernetes architectures. A key innovation of Sci-Orchestra is its autonomous marketplace, which serves as a catalyst for cross-institutional collaboration. Through an intuitive user interface, researchers can rapidly deploy and share specialized services via simple selections, eliminating the need for complex installations and technical setups. This modular infrastructure is specifically designed to facilitate industry partnerships as it provides a secure execution environment and allows external collaborators to test and validate proprietary tools without the need for source-code exchange. This ``black-box'' interoperability protects intellectual property while enabling seamless integration into broader scientific pipelines, ultimately accelerating the transition from laboratory prototypes to industrial-scale applications. --- ### Geometry-Aware Discretization Error of Diffusion Models Authors: Samuel Hurault, Thomas Moreau, Gabriel Peyré Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08392v1 Practical diffusion sampling is a numerical approximation problem: under a fixed inference budget, one must simulate a reverse-time ODE or SDE using only a limited number of denoising steps, so discretization error is often the dominant source of error. Existing non-asymptotic analyses provide convergence guarantees, but are typically too loose and too insensitive to diffusion parameters to guide practical design: broad families of schedules receive the same rates, which depend on coarse worst-case quantities such as the dimension or the drift Lipschitz constant. We take a less ambitious but more informative route. In the exact-score setting, we derive first-order asymptotic expansions of the Euler-Maruyama weak and Fréchet discretization errors. These formulas hold for general smooth reverse diffusions and become fully explicit under Gaussian data. They show how discretization error adapts to the geometry of the data through the covariance spectrum, and how this geometry interacts with key diffusion parameters, including the diffusion schedules and the diffusion-term coefficient. This yields tractable objectives for geometry-aware parameter optimization. Finally, we show that the qualitative predictions of the Gaussian formulas remain robust across diffusion sampling problems with different geometries, including image generation on different datasets and image posterior sampling. --- ### SACHI: Structured Agent Coordination via Holistic Information Integration in Multi-Agent Reinforcement Learning Authors: Nikunj Gupta, James Zachary Hare, Jesse Milzman, Rajgopal Kannan, Viktor Prasanna Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08391v2 Cooperative multi-agent reinforcement learning agents that act on partial local observations face a fundamental information bottleneck: the knowledge needed to select jointly optimal actions is scattered across the team, yet each agent must commit to a decision without access to its teammates' observations, intentions, or chosen actions. Existing methods either ignore this bottleneck, compress it into a scalar mixing signal, or route around it with learned communication channels. Framing action coordination as a problem of structured information integration among agents, we propose \textit{structured agent coordination via holistic information integration}, or SACHI, in which graph transformer convolutions over an inter-agent coordination graph enrich each agent's representation with receiver-sensitive, content-dependent signals from teammates prior to action selection. We evaluate SACHI across five cooperative tasks spanning spatial, communicative, and adversarial coordination challenges against twelve baselines. SACHI consistently matches or outperforms the best baseline on every task, and rigorous aggregate statistical analyses, including normalized metrics with bootstrap confidence intervals, Friedman ranking, and performance profiling, confirm that this advantage is statistically significant, robust across environments, and not attributable to increased model capacity. Parameter-matched ablations further trace the source of the gains to a single architectural property: the degree of content-dependence in the message-passing operator. --- ### The Power of Second Order Methods for Sequence Preconditioning Authors: Annie Marsden, Elad Hazan Published: 2026-05-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.08390v2 Sequence prediction methods for linear dynamical systems with long memory, i.e. marginally stable systems, typically achieve regret that grows linearly with the hidden dimension of the underlying generative model. While many methods have been developed to address this regime with varying success, we show that simply using the second-order Vovk-Azoury-Warmuth (VAW) algorithm to learn a short autoregressive-with-inputs (ARX) model achieves astoundingly strong results: for bounded sequential data from a marginally-stable linear dynamical system with spectra in the complex disk except for angular wedge of width $δ$ around the negative real axis, this algorithm achieves dimension-free regret $O\left( δ^{-4} \log^2 T \right)$. These bounds are state-of-the-art to our knowledge. The key components for our result come from 1) using the theory of ``Universal Sequence Preconditioning'' (USP) \cite{marsdenuniversal} to prove the existence of an optimal setting of autoregressive coefficients, 2) the application of VAW which takes better advantage of the memory compression provided by USP, and 3) the analysis of Faber polynomials on circular sectors to extend these results to systems with complex spectra. --- ### Decoupling Endpoint and Semantic Transition Learning for Zero-Shot Composed Image Retrieval Authors: Mingyu Liu, Sihan Huang, Yijia Fan, Yinlin Yan, Quan Zhang, Jian-Fang Hu, Jianhuang Lai Published: 2026-05-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.08389v2 Zero-shot composed image retrieval (ZS-CIR) retrieves a target image from a reference image and a text modification without human-annotated CIR triplets. Projection-based ZS-CIR methods are attractive because they do not rely on LLMs at inference and remain lightweight, but they often underperform LLM-based approaches on complex semantic modifications. This gap reflects a semantic transition bottleneck in projection-based ZS-CIR: endpoint-level matching can let the edit text act as a target-side attribute cue rather than grounding it as a source-conditioned semantic transition. We further show that adding semantic transition supervision to the same text adapter creates an endpoint--transition conflict between endpoint alignment and semantic transition alignment. To address this conflict, DeCIR decouples endpoint and transition learning. It constructs paired forward/reverse edit tuples from image-caption pairs, trains separate low-rank text adapter branches for endpoint alignment and semantic transition alignment, and merges them with Low-Rank Directional Merge (LRDM) into one deployable adapter. Extensive experiments on CIRR, CIRCO, FashionIQ, and GeneCIS demonstrate that DeCIR consistently improves projection-based ZS-CIR without increasing inference complexity. --- ### PLACO: A Multi-Stage Framework for Cost-Effective Performance in Human-AI Teams Authors: Pranavkumar Mallela, Vinay Kumar, Shashi Shekhar Jha, Shweta Jain Published: 2026-05-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.08388v1 Human-AI teams play a pivotal role in improving overall system performance when neither the human nor the model can achieve such performance on their own. With the advent of powerful and accessible Generative AI models, several mundane tasks have morphed into Human-AI team tasks. From writing essays to developing advanced algorithms, humans have found that using AI assistance has led to an accelerated work pace like never before. In classification tasks, where the final output is a single hard label, it is crucial to address the combination of human and model output. Prior work elegantly solves this problem using Bayes rule, using the assumption that human and model output are conditionally independent given the ground truth. Specifically, it discusses a combination method to combine a single deterministic labeler (the human) and a probabilistic labeler (the classifier model) using the model's instance-level and the human's class-level calibrated probabilities. --- ### SkillLens: Adaptive Multi-Granularity Skill Reuse for Cost-Efficient LLM Agents Authors: Yongliang Miao, Ziyang Yu, Liang Zhao, Bowen Zhu, Hasibul Haque Published: 2026-05-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.08386v1 Skill libraries have become a practical way for LLM agents to reuse procedural experience across tasks. However, existing systems typically treat skills as flat, single-resolution prompt blocks. This creates a tension between relevance and cost: injecting coarse skills can introduce irrelevant or misleading context, while rewriting entire skills is expensive and often unnecessary. We propose SkillLens, a hierarchical skill-evolution framework that organizes skills into a four-layer graph of policies, strategies, procedures, and primitives, and retrieves them at mixed granularity. Given a task, SkillLens first retrieves semantically relevant skill seeds, expands them through degree-corrected random walk over the skill graph, and then uses a verifier to decide whether each visited unit should be accepted, decomposed, rewritten, or skipped. This enables the agent to reuse compatible subskills directly while adapting only locally mismatched components. To improve the system over time, SkillLens further refines multi-granularity skills and verifier in order to improve its routing decisions. We provide theoretical analysis showing that mixed-granularity adaptation incurs sublinear cost under sparse mismatch assumptions and that the evolutionary update rule monotonically improves the validation objective until a local optimum. Across MuLocbench and ALFWorld, SkillLens consistently improves over strong skill-based baselines, achieving up to a 6.31 percentage-point Acc@1 gain for bug localization and raising agent success rate from 45.00% to 51.31%. ---