# Research Papers — 2026-07 · bundle page 51 of 222 > Full-text markdown of papers #2501–2550 of 11075 published in 2026-07, newest first. AI-readable mirrors of the full 11075-paper 2026-07 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-07/page/51 [← Previous bundle](https://aipod.dev/papers/archive/2026-07/page/50.md) [Next bundle →](https://aipod.dev/papers/archive/2026-07/page/52.md) ### Sparse Gaussian-Mixture-Model Q-Functions via Hadamard Overparametrization for Online Reinforcement Learning Authors: Minh Vu, Konstantinos Slavakis Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23474v1 This paper develops an online, off-policy policy-iteration framework for reinforcement learning (RL), based on sparse Gaussian-mixture-model Q-functions (S-GMM-QFs). The framework reconciles streaming, non-stationary data with the Riemannian structure of the parameter space while handling distributional mismatch through experience replay. S-GMM-QFs are introduced via Hadamard overparametrization, enabling interpretable sparsification through smooth regularization that facilitates Riemannian-based optimization. Overparametrization allows the framework to adaptively identify meaningful components from a large initial pool, yielding sparse models where interpretability emerges naturally from geometry: each component's parameters (means and covariances) explicitly encode its geometric role in the ambient state-action space. These geometric roles are learned through online gradient descent on a smooth objective over a (Cartesian-product) Riemannian manifold. Numerical tests demonstrate that S-GMM-QFs match or exceed deep RL methods while using substantially fewer parameters and achieving faster improvement per observed transition. Notably, parameter efficiency and interpretability combine to maintain strong generalization in low-parameter regimes where sparsified deep RL approaches degrade. --- ### VIPER: Visual In-Context Physics Reasoning for Physically Plausible Video Generation Authors: Tianxiao Chen, Hanmo Chen, Huajin Chen, Bo Li, Qi Ye, Peng-Tao Jiang Published: 2026-07-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.23472v1 Modern video generation models can synthesize visually compelling and temporally coherent clips, yet controlling their physical behavior remains difficult with standard text and image conditions. The core challenge is a conditioning bottleneck: material response, contact interaction, deformation, and motion trajectory are continuous and relational physical cues that are hard to specify exhaustively in language but can be demonstrated naturally by video. We propose VIPER, a Visual In-Context Physics Reasoning framework for reference-guided image-to-video generation. Given a target image, a brief target prompt, and a reference video, VIPER treats the reference as a dense visual demonstration of the desired physical process rather than an appearance template. It uses a Multimodal Large Language Model (MLLM) to extract reference-derived physical cues and guide a pretrained image-to-video generator through a hierarchical training strategy, enabling physical behavior transfer while preserving the visual prior of the base generator. To support this setting, we construct VIPER-19K, a curated dataset with material, trajectory, and physical-impact annotations, together with filtered reference-target pairs. Experiments on an unseen validation set show that VIPER achieves stronger reference-video physical similarity and higher human preference than representative video generation and video-as-prompt baselines, while maintaining competitive general video quality. Qualitative results further demonstrate that VIPER can transfer reference-derived physical behavior to new target scenes without requiring carefully engineered prompts. --- ### When Every Simulation Counts: Value-Based Reinforcement Learning for Accelerated Photonics Inverse Design Authors: Longying Wen, Feiyang Wu, Jinglin Yu, Chongxian Yuan, Renjie Li, Zhaoyu Zhang Published: 2026-07-26 Categories: physics.optics, cs.AI, cs.LG, physics.app-ph arXiv: https://arxiv.org/abs/2607.23469v1 Photonic-crystal surface-emitting lasers (PCSELs) can combine high-power operation with narrow-divergence surface emission, but optimizing coupled parameters requires costly full-wave simulations. Deep Q-network (DQN) optimization can reuse simulated transitions to guide edits, yet which value-learning mechanisms remain reliable under tight simulation budgets is unknown. We address this gap by comparing baseline DQN and six value-based variants for a seven-variable PCSEL design under a shared objective, simulator, 83-call budget, and four matched initializations. Beyond endpoints, we analyze sample efficiency, policy behavior, and physical response to separate learning gains from favorable starts or exploratory jumps. Dueling DQN is the only variant to improve all four seeds. Relative to the first evaluated designs, its selected structures increase the mean quality factor () from to (), reduce wavelength error by 64%, and increase upward power by 47%; compared with baseline DQN, they achieve a higher mean under the same budget. Other variants yield no consistent improvement; Double DQN reproduces baseline trajectories, while Rainbow-lite shows high upside but strong seed dependence. These results identify Dueling DQN as the most reliable configuration tested for simulation-budget-limited PCSEL inverse design and provide a reproducible framework for attributing algorithmic gains in scientific optimization. The source code is publicly available at https://github.com/Longying-Wen/PCSEL-RL. --- ### Robust 6-DoF Object Pose Tracking with Built-In Recovery under Occlusions and Rapid Object Motions Authors: Balázs Opra, Léo Ghafari, Thomas Stewart, Cyrill Stachniss Published: 2026-07-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.23468v1 Real-time 6-DoF object pose tracking is essential for many robotics applications, and several approaches exist. Yet even today's approaches remain unreliable under temporary full occlusions and rapid object motions. Once tracking is lost, most methods struggle to detect the failure and recover automatically, often requiring manual re-initialization. In this paper, we address the problem of robust model-based 6-DoF tracking of unseen objects from RGB-D data, especially in scenarios with occlusion and fast motion. We propose a novel method that combines efficient learning-based keypoint matching with optimization-based alignment and introduces a novel failure detection and recovery module. Our system monitors pose reliability, detects tracking divergence or occlusions, and performs a global re-detection and pose estimation step that robustly verifies recovery candidates before resuming tracking. Our evaluation on standard tracking benchmarks and on a new dataset of occluded and fast-moving scenes shows that our method matches state-of-the-art accuracy on easy tracking sequences, maintains high tracking speed at 57.6 frames per second, and provides the most robust tracking performance under challenging conditions. Thus, we believe that our approach is a relevant step forward in robust 6-DoF object tracking from RGB-D data. --- ### Learning to Optimize: Joint Routing and Flow Allocation on Sparse Non-Euclidean Networks Authors: Haomiao Sun, Fang He, Congyuan Ji, Xindi Tang Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23467v1 We study an integrated pickup-and-delivery problem on sparse, non-Euclidean networks that jointly optimizes cyclic routing, cargo flow allocation, and cross-cycle service. The tight coupling of these operational constraints creates a complex discrete-continuous decision space with highly restricted feasible regions. To overcome these computational challenges, we propose Double-Channel Graph Attention (DCGA), an end-to-end reinforcement learning framework. DCGA isolates network reachability and demand-service logic into separate graph channels and constructs valid routes using a simulator-coupled, constraint-informed decoder. Experiments on LinerLib benchmarks demonstrate that DCGA achieves seconds-level inference and delivers state-of-the-art solution quality on instances beyond a specific scale, with its advantage over existing baselines widening significantly as problem size increases. Supported by extensive stability and ablation analyses, our results demonstrate that this structure-aware learning approach provides an effective, low-latency engine for realistic routing-and-flow optimization. --- ### Extending Fourier Neural Operators for Modeling Parameterized and Coupled PDEs Authors: Cheng Jing, Uvini Balasuriya Mudiyanselage, Abhishek Verma, Kallol Bera, Shahid Rauf, Kookjin Lee Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23466v1 Parameterized and coupled partial differential equations (PDEs) are central to modeling phenomena in science and engineering, yet neural operator methods that address both aspects remain limited. We extend Fourier neural operators (FNOs) with minimal architectural modifications along two directions. For parameterized dynamics, we propose a hypernetwork-based modulation that conditions the operator on physical parameters. For coupled systems, we conduct a systematic exploration of architectural choices, examining how operator components can be adapted to balance shared structure with cross-variable interactions while retaining the efficiency of standard FNOs. Evaluations on benchmark PDEs, including the one-dimensional capacitively coupled plasma equations and the Gray-Scott system, show that our methods achieve up to 55-72% lower errors than strong baselines, demonstrating the effectiveness of principled modulation and systematic design exploration. --- ### Direction-adaptive Mamba: Spatial-Frequency Dual-Domain Collaborative Learning for PolSAR Image Classification Authors: Junfei Shi, Yu Cheng, Haojia Zhang, Wenqiang Hua, Junhuai Li, Maoguo Gong Published: 2026-07-26 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2607.23464v1 Deep learning dominates polarimetric synthetic aperture radar (PolSAR) image classification, with Mamba architectures serving as favorable backbones due to linear complexity and strong global modeling capacity. However, existing PolSAR Mamba methods have two critical flaws: pure spatial processing discards fine-grained edges and textures, and fixed scanning patterns fail to model direction-variant anisotropic scattering and weak boundaries essential for PolSAR physical analysis. This work proposes DA-Mamba, a direction-adaptive Mamba framework with dual-domain collaborative learning for PolSAR classification. Equipped with an edge-aligned direction-adaptive scanning scheme, DA-Mamba captures long-range spatial dependencies and accurate boundary details. It adopts the Non-Subsampled Contourlet Transform (NSCT) to separate PolSAR data into low-frequency global components and multi-directional high-frequency subbands, extracting anisotropic structural features from high-frequency information while preserving global context via low-frequency branches. A dual-domain collaborative learning module further integrates spatial scattering and frequency-domain representations to strengthen feature discriminability. Evaluated on three real-world PolSAR datasets, DA-Mamba surpasses state-of-the-art methods, verifying the efficacy of the proposed adaptive scanning and dual-domain fusion designs. Code will be publicly available. --- ### Two Regimes of Chain-of-Thought Unfaithfulness: Metric-Based Detection Fails Where Models Are Wrong Authors: Suramya R. Angdembay, Dikshant Aryal, Nick Rahimi Published: 2026-07-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.23458v2 Chain-of-thought (CoT) explanations support oversight only if they are faithful: the stated reasoning must actually produce the answer. Auditing black-box (behavioral) detection of unfaithful CoT against FaithCoT-Bench's human annotations, we find answer correctness structures the problem at every level. Answer incorrectness alone (an oracle diagnostic, not a deployable detector) outperforms every purpose-built signal (AUROC 0.696), because 69% of annotated unfaithfulness occurs on incorrect answers. Stratifying by correctness splits detection into two regimes: on correct answers, behavioral signals moderately separate faithful from post-hoc reasoning (0.63-0.67); on incorrect answers, where most unfaithfulness lives, no tested signal is detectably above chance (replicated on all four models for benchmark-wide signals). The standard step-removal metric anti-correlates with human labels; this inversion reproduces on the benchmark's released scores and on hint-dependent counterfactually labeled traces. Linear probes decode the behaviorally blind regime in Llama-3.1-8B and the correct-answer regime in Qwen-2.5-7B, with no shared, positively aligned direction detected across regimes; instructed answer-first traces (7 models) transfer to neither annotated regime, while hint-induced unverbalized answer flips do, in model- and source-dependent settings. We also independently verify and resolve a documentation-data mismatch in the benchmark's label semantics. --- ### Two Regimes of Chain-of-Thought Unfaithfulness: Behavioral Detection Fails Where Models Are Wrong Authors: Suramya R. Angdembay, Dikshant Aryal, Nick Rahimi Published: 2026-07-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.23458v1 Chain-of-thought (CoT) explanations support oversight only if they are faithful: the stated reasoning must actually produce the answer. Auditing black-box (behavioral) detection of unfaithful CoT against FaithCoT-Bench's human annotations, we find answer correctness structures the problem at every level. Answer incorrectness alone (an oracle diagnostic, not a deployable detector) outperforms every purpose-built signal (AUROC 0.696), because 69% of annotated unfaithfulness occurs on incorrect answers. Stratifying by correctness splits detection into two regimes: on correct answers, behavioral signals moderately separate faithful from post-hoc reasoning (0.63-0.67); on incorrect answers, where most unfaithfulness lives, no tested signal is detectably above chance (replicated on all four models for benchmark-wide signals). The standard step-removal metric anti-correlates with human labels; this inversion reproduces on the benchmark's released scores and on hint-dependent counterfactually labeled traces. Linear probes decode the behaviorally blind regime in Llama-3.1-8B and the correct-answer regime in Qwen-2.5-7B, with no shared, positively aligned direction detected across regimes; instructed answer-first traces (7 models) transfer to neither annotated regime, while hint-induced unverbalized answer flips do, in model- and source-dependent settings. We also independently verify and resolve a documentation-data mismatch in the benchmark's label semantics. --- ### Generalization bounds and sample complexity for remaining useful life prediction from complete degradation trajectories Authors: Huy Hoang Le, Kim-Anh Nguyen Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23454v1 Data-driven remaining useful life (RUL) prediction requires complete degradation trajectories for training, yet such run-to-failure data are scarce and expensive. Practitioners currently lack principled guidance on how many failure examples suffice for a given model and accuracy target. This paper develops a sample complexity framework for RUL prediction comprising seven main results organised around three themes. First, we establish fundamental learning rates: a distribution-free generalization bound shows that the uniform deviation of the mean squared error decreases as $O(B^{2}\sqrt{p/n})$, where $p$ is the model complexity and $n$ the number of trajectories, and a minimax lower bound proves that the $Θ(p/n)$ rate is unimprovable.} \rev{Second, we quantify how domain knowledge accelerates learning: incorporating degradation physics reduces data requirements by up to two orders of magnitude for deep networks, a Bernstein-type analysis achieves the minimax-optimal $O(p/n)$ rate under high signal-to-noise conditions, and closed-form penalties reveal when an incorrectly assumed physics model hurts rather than helps. Third, we characterise the impact of data quality: fleet variability induces an irreducible bias$-$variance tradeoff, while right-censored observations suffer an efficiency loss that depends critically on the degradation class.} Closed-form expressions are provided for exponential, power-law, and stretched-exponential degradation. \rev{Cross-domain validation against published turbofan, battery, and bearing benchmarks confirms the theoretical predictions within a factor of 2$-$3 on average. The results yield practical guidelines for planning data collection, selecting model complexity, and evaluating physics model assumptions in prognostics applications. --- ### Multi-Modal Object Re-Identification with Prompt-S6 and Semantic-Aware Knowledge Guidance Authors: Weixiang Zhou, Jiabei Zuo, Yuhao Wang, Cong Wang, Huchuan Lu, Zhixun Su Published: 2026-07-26 Categories: cs.CV arXiv: https://arxiv.org/abs/2607.23451v1 Multi-modal object Re-Identification (ReID) aims to retrieve specific objects by integrating complementary information from multiple modalities. However, existing multi-modal ReID methods do not effectively address background interference suppression or achieve tri-modal alignment, instead focusing on pairwise feature fusion. Moreover, many current aggregation approaches suffer from high computational complexity. To address these limitations, we propose PRISM, a novel multi-modal ReID framework built upon Prompt-S6 (PS6) and semantic-aware knowledge guidance. PS6 maintains the linear complexity and strong sequence modeling capability of Mamba while enabling efficient cross-modal interaction. Leveraging these advantages, we design two key components: Semantic-Driven Token Pruning (SDTP) and Progressive Fusion Network (PFN). Parsing semantic priors from the segmentation foundation models, the SDTP then leverages these priors and applies dynamic token pruning to suppress background noise and refine feature representations. The PFN progressively aggregates multi-modal features to achieve tri-modal alignment and fully exploit modality complementarity. With the proposed modules, PRISM generates more robust multi-modal representations under complex scenarios. Extensive experiments on four multi-modal object ReID benchmarks demonstrate the effectiveness and efficiency of our approach. The source code is available at https://github.com/zw-absin/PRISM. --- ### Local Regularization Does Not Characterize Multiclass PAC Learnability Authors: Eric Hou Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23449v1 Local regularization assigns each hypothesis a test-point-dependent score and predicts with a minimum-score hypothesis consistent with the sample. Asilis et al. asked whether this principle characterizes multiclass PAC learnability. We give a negative answer. There is a countable class of Daniely--Shalev-Shwartz dimension at most two with realizable PAC sample complexity \( O\!\left(\frac{1}{\varepsilon}\log\frac{1}δ\right), \) that no local regularizer learns. Hypotheses are edges of complete graphs and instances are tournaments. At a test tournament, the scores fix an edge ranking while the training sample independently removes competitors. Cyclic triangles force enough inversions that surviving competitors produce constant population error at arbitrarily large sample sizes. --- ### Constraint-Bound Agnostic Bayesian Optimization: One Model for All Thresholds Authors: Jin Wang, Xi Lin, Handing Wang Published: 2026-07-26 Categories: cs.NE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2607.23448v1 Expensive constrained optimization problems in real-world industry design often involve constraint thresholds that are difficult to determine in advance. Engineers may need to adjust constraint thresholds to explore different feasibility-performance trade-offs, requiring solutions under a wide range of threshold settings. However, existing constrained Bayesian optimization methods treat each threshold configuration independently, leading to repeated optimization and failing to exploit the shared relationship among continuously varying thresholds. To address this challenge, we propose constraint-bound agnostic Bayesian optimization (CBA-BO), a learning-based framework that learns a parametric constraint model mapping thresholds to optimal solutions. Once learned, CBA-BO directly predicts solutions for arbitrary unseen threshold configurations without additional optimization, with a one-step Bayesian optimization refinement further improving solution quality. Experiments on benchmark and engineering problems demonstrate that CBA-BO learns a transferable threshold-solution mapping, enabling efficient prediction and optimization for arbitrary threshold queries. An intent-guided constraint-bound recommendation mechanism is further developed to improve objective performance while satisfying user-specified constraint preferences. --- ### PerturbPFN: Probing the Limits of Synthetic Priors in Drug Perturbation Modelling Authors: Yuche Gao, José Miguel Hernández-Lobato, Siyuan Guo Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23447v1 Predicting cellular responses to unseen chemical perturbations is challenging due to unknown targets and mechanisms, high-dimensional expression responses, and limited experimental coverage of the large small-molecule design space. We propose PerturbPFN, a PFN-style amortized model for unknown-target perturbation prediction under a hierarchical synthetic structural prior. Instead of directly regressing high-dimensional expression responses, PerturbPFN infers a latent system graph, sparse atomic intervention targets, and intervention strengths, then propagates their effects through an SCM decoder. The model is trained entirely on prior-predictive synthetic episodes generated from biologically motivated graph and expression simulators, enabling structured in-context learning without test-time gradient updates. We evaluate PerturbPFN on both real single-cell perturbation data and synthetic benchmarks, covering effect prediction, target identification, and regulatory structure discovery. Our results show that PerturbPFN offers a complementary trade-off to specialized baselines, achieving competitive perturbation prediction with low inference cost while exposing interpretable intermediate estimates of targets, strengths, and system structure. --- ### Do Small Models Use the Law You Give Them? Context-Injected Fine-Tuning for Legal QA in Bangladesh Authors: Moniruzzaman Mahadi, Abrar Mohammed Tanzim Alam, Sayma Siddika Monalisa, Mir Mohammad Asif Abdullah, Swakkhar Shatabda, Md Adnan Arefeen Published: 2026-07-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.23446v1 A small language model can receive the governing statutory provision and still answer incorrectly. We test whether fine-tuning on examples containing relevant law improves later use of retrieved law. We curate 2{,}165 bilingual QA records from six Bangladeshi acts and three schedules, then fine-tune Qwen3.5 at 0.8B, 2B, and 4B. Evaluation uses the 2022 and 2023 Bangladesh Bar Council exams in Bangla and machine-translated English, with no retrieval, BM25, or FAISS, scored by strict consistency over three seeded runs. At 0.8B, fine-tuning raises the 2022 English FAISS score from 2 to 34 of 100. Gains at 0.8B and 2B survive paired testing, but the 4B model has no detectable net gain: Bangla improves while several English conditions regress. Fine-tuning also reduces answers that drift from Bangla into mostly English from 44.0--53.2\% to 0.2--0.7\%, with adjusted $p<.001$ at every scale. Retrieval quality is therefore not the only bottleneck. Small bilingual legal models also differ in how they use supplied law and whether they answer in the requested language. The dataset is publicly available at https://huggingface.co/datasets/momahadi/bangladesh-legal-qa-dataset. --- ### Omni-Prune: Query-Aware Unified Token Pruning for Efficient Omnimodal Large Language Models Authors: Yiming Zhong, Chang Nie, Caifeng Shan Published: 2026-07-26 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2607.23445v1 Omnimodal large language models (OmniLLMs) are rapidly extending multimodal reasoning to cover synchronized audio and video. However, the resulting audio-video token sequences are long, leading to high prefill latency and GPU memory usage at inference time. Existing token pruning methods, designed mainly for vision-only inputs, miss both the cross-modal links between audio and video and the user query that decides which content matters. To bridge this gap, we present Omni-Prune, a training-free, query-aware audio-visual token pruning framework that jointly removes redundancy from both modalities while keeping task-relevant cross-modal evidence. Specifically, Omni-Prune first splits the token sequence into adaptive time windows placed at audio saliency peaks, then scores audio and video tokens on a single scale that combines encoder attention with text-query relevance, and pairs related audio-video tokens so that they are kept together. Within each window, a final K-medoids step then selects a few representative tokens, adding diverse cues that score-based selection alone would miss. Extensive experiments demonstrate that Omni-Prune outperforms established baseline methods, delivering up to 3.25x prefill speedup and 1.3x memory reduction while retaining over 99% of full-model performance. --- ### Do LLM Debates Repeat Arguments Differently Across Languages? Authors: Huiqian Lai Published: 2026-07-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.23442v2 LLM debate is usually evaluated by final answers, yet transcripts reveal whether later turns develop new arguments or return to earlier claims in new wording. We study this process with \textit{prior-argument similarity}, which compares extracted argument units with earlier units in the same debate. In controlled eight-turn debates over 71 motions, six languages, and four model agents, Chinese is the only tested language with a consistently positive gap relative to English across three multilingual embedding models. The gap persists across agents, turn positions, regression adjustment, metric variants, extraction-length controls, a second-extractor subset, and cross-encoder tail rescoring. Manual calibration shows weak item-level alignment but a high-similarity tail enriched for substantive repetition. A diversity-aware prompt lowers \textit{prior-argument similarity} across languages, yet does not significantly narrow the Chinese--English gap. Multilingual debate evaluation should therefore measure argumentative development over time and report both average and gap terms. --- ### Reasoning or Memorization: Can LLMs Understand and Generate Chinese Xiehouyu Riddles? Authors: Hai Hu, Siyuan Song, Chongtian Shao, Kejia Zhang, Tianjian Zhu, Xiaojing Zhao Published: 2026-07-26 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.23440v1 In this paper, we push the boundary of LLM reasoning by testing them in a Chinese language game, xiehouyu, with novel xiehouyu created by linguists that had not existed before to avoid data contamination. We use multiple-choice questions (MCQ), free-form explanation generation, and new xiehouyu creation to evaluate LLMs' ability to understand and create xiehouyu. In MCQ, we use the delta of accuracy ($Δ_{acc}$) between existing but low-frequency xiehouyu and novel ones as an index for memorization. $Δ_{acc}$ for native speakers is very low, suggesting similar processing mechanisms. However, we found that frontier Chinese models have on average a $Δ_{acc}$ of 23.6\%, while English-centric models tested have a mean $Δ_{acc}$ of 5.1\%, suggesting that frontier Chinese models are likely trained with much larger Chinese data, thus memorizing more low-frequency xiehouyu. For novel xiehouyu, Gemini 3.1 Pro demonstrated remarkable ability with acc 92.6, which is 24\% higher than human accuracy. In xiehouyu creation, those created by LLMs receive much worse ratings than those by humans. These results suggest that claims about the reasoning abilities of LLMs may need careful re-examination considering the data contamination issue, and that LLMs' creativity in language-related tasks may still be behind human experts, at least in Chinese xiehouyu. --- ### A Characterization of the Orthocomplement of the Tangent Space of Semiparametric Markov Models Authors: Trung Phung, Ilya Shpitser Published: 2026-07-26 Categories: stat.ME, cs.AI arXiv: https://arxiv.org/abs/2607.23439v1 Graphical models are ubiquitous in social and empirical science as they are intuitive and easy to use. These models belong to the broader class of Markov models, defined using solely conditional independence (CI) restrictions. In order to estimate finite-dimensional target parameters in such models efficiently, semi-parametric theory provides a principled framework for constructing regular and asymptotically linear estimators via influence functions (IFs). These estimators are asymptotically normal and root-$n$ consistent. Characterizing the class of all influence functions for a target parameter is crucial for statistically efficient inference in these models. For models that are Markov relative to directed acyclic graphs (DAGs), the orthogonal complement of the tangent space is known, implying that for any target the class of all influence functions can be derived once an influence function is obtained. On the other hand, for Markov models not equivalent to a DAG model -- such as ordinary Markov models associated with undirected graphs, chain graphs, or acyclic directed mixed graphs -- the orthogonal complement has not been characterized, impeding semi-parametric inference in these models. We derive closed form expressions for the orthogonal complement of the tangent space for general Markov models and illustrate our results by characterizing the class of influence functions for the conditional mean parameter in several graphical models. --- ### Separating Capability from Permission: A Governance Framework for Agentic AI Autonomy Levels Authors: Haining Zheng, Qian Dong, Rodolfo K. Depena, Jonathan D. Bhatia, Feng Xiao, Peng Xu Published: 2026-07-26 Categories: cs.AI, cs.CY, cs.MA arXiv: https://arxiv.org/abs/2607.23438v1 As AI systems increasingly exhibit agentic behavior, discussions of autonomy often conflate what systems are technically capable of doing with what they should be permitted to do in practice. This paper introduces a governance framework that explicitly separates Allowed Autonomy Levels (AAL), which define the degree of autonomy an AI agent is authorized to exercise given risk, oversight, and accountability considerations, from Autonomous Capability Levels (ACL), which characterize an agent's inherent technical abilities. We present a structured set of autonomy levels spanning reactive execution, decision support, supervised action, goal-directed autonomy, and delegated operational authority, and describe how control, reversibility, and accountability change as autonomy increases. To operationalize this framework, we propose a risk-aware decision process for assigning allowed autonomy, analyze how risk and accountability evolve across autonomy levels, and demonstrate its application through a deployed enterprise data engineering agent, illustrating how a system assessed at a high capability level can be deliberately constrained to a lower allowed autonomy based on risk, reversibility, and organizational readiness. By distinguishing authorization from capability, this work provides practical guidance for the design, deployment, and governance of Agentic AI systems. --- ### Neural Representation of Minimal Surfaces Authors: Jiayin Sun, Albert Chern Published: 2026-07-26 Categories: cs.GR, cs.LG arXiv: https://arxiv.org/abs/2607.23437v1 We propose a neural representation for minimal surfaces. Unlike prior approaches based on discretization or Physics-Informed Neural Networks (PINNs), where meshes or neural fields are optimized to approximate the governing equations, our method builds on an exact representation, similar to the classical Weierstrass--Enneper parameterization, yielding minimal surfaces up to negligible quadrature error in evaluation. We formulate a training objective for the Plateau problem that optimizes over this representation. --- ### Two-Timescale Hierarchical Reinforcement Learning for Resilient Operations Authors: Young Hyun Cho, Franz Stoll, Will Wei Sun, Guang Lin, Stephan Biller Published: 2026-07-26 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2607.23434v1 Unexpected shocks recur in global operations, requiring decision rules that adapt as market and operating conditions change. Many operational systems also have hierarchical structures in which long-term and short-term decisions pursue a shared objective. We study how hierarchical reinforcement learning can strengthen resilience by adapting these interdependent rules jointly. We develop a two-timescale hierarchical reinforcement learning framework that adapts long-term and short-term policies at their respective time scales. Because the policies are interdependent, we synchronize their updates and prove, to our knowledge, the first convergence guarantees for coupled two-timescale learning. Over $T$ periods, our policies' average gap from an optimal policy pair is $O(T^{-1/2})$, improving to $O(\log T/T)$ when poor decisions produce clearer profit losses. In a used-car case study, inventory replenishment is the long-term decision and customer-arrival pricing the short-term decision. Relative to the strongest partially adaptive benchmark, the framework increases mean profit by $9.2\%$ under joint demand-supply shocks and by $11.8\%$ under a prolonged shock scenario, while maintaining a more stable profit trajectory over time. Short-term adaptation addresses routine seasonality and one-sided disruptions by responding immediately to changing conditions. Under joint demand-supply shocks, however, it is insufficient alone; long-term adaptation is also needed to create favorable conditions for short-term decisions. Joint adaptation thus yields higher and more stable profits through disruption and recovery. Because many organizations already use hierarchical planning, the framework strengthens operational resilience without altering existing decision structures. --- ### Short-Term Pain for Long-Term Gain: Adaptive Experiment with Post-Commitment Reward Shift Authors: Puping Jiang, Wei Tang Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23432v1 Decision-makers in learning environments face a dilemma when their short-term optimal actions may not favor their long-term benefits the most. To understand the fundamental tradeoff behind the dilemma, we study adaptive experimentation with post-commitment reward shifts. During an experiment phase, the decision-maker may adaptively test multiple options; during a subsequent commitment phase, the decision-maker must commit to a single option, whose reward may differ from its pre-commitment reward. We propose the Reserved Arm Eliminations for Commitment (RAEC) algorithm, which reserves a predetermined portion of the experiment phase to identify the best post-shift option while using the remaining rounds to minimize short-run regret. We establish regret upper bounds for RAEC across all parameter regimes and matching minimax lower bounds, providing a tight characterization of the cost of balancing short-term performance and long-term commitment. We also study two extensions. With prior structural knowledge linking pre- and post-shift rewards, we show that correctly identifying the ranking-changing component of the shift is more important than estimating its absolute magnitude. For settings with concave commitment rewards and portfolio choice, we develop the Reserved Online Stochastic Convex Optimization for Commitment (ROSCOC) algorithm, which directly converts its reserved exploration history into a commitment portfolio and achieves tight regret bound. Finally, we also conduct numerical experiments which confirm that our proposed algorithms achieve the desired regret predicted by our theory, and also outperform other baseline algorithms. --- ### TLA+-Bench: An Execution-Grounded Benchmark and Dataset for Natural-Language to TLA Specification Generation Authors: Arslan Bisharat, Eric Spencer, Brian Ortiz, Khushboo Bhadauria, Mujtaba Nazari, Beatriz Santos, Anisa Ramos, TaiNing Wang, George K. Thiruvathukal, Konstantin Läufer, Mohammed Abuhamad Published: 2026-07-26 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2607.23425v1 Large language models increasingly write TLA$^{+}$ formal specifications from natural-language descriptions, but progress is hard to measure: existing resources grade by resemblance to a reference or by whether the output parses, neither of which shows correctness. We present TLA$^{+}$-Bench, a dataset and benchmark that grades by execution. Every gold specification ships a configuration the TLA$^{+}$ model checker runs over the full reachable state space, deciding exactly whether the specification holds the properties that configuration names. The dataset holds 403 model-checked gold and 897 parse-only silver specifications from 13 public repositories, subsumes prior TLA$^{+}$ generation data, and carries four model-written descriptions in two styles from two providers, with difficulty and category labels. Our main finding is about measurement itself: an exact oracle gives not one correctness number but a range. Varying only the grading choices earlier benchmarks leave unstated, on one fixed set of model outputs, the correct rate moves sixfold, from 10.0\% to 1.7\%; adding the interface-supply choice, where the model is told the configuration's names, widens the range to elevenfold, from 18.7\% to 1.7\%. We call this range the correctness envelope and measure each of its bounds. The findings inside it are stable. Every model writes valid TLA$^{+}$ far more often than correct TLA$^{+}$: the strongest is correct 16\% of the time by default and 26\% when given the interface names, open models at most 1\%, and correctness falls sharply with difficulty. --- ### LA-RL: Label-Aware Self-Reflection for Reinforcement Learning in Information Extraction Authors: Xiao You, Tianwei Yan, Zixu Shan, Longyu Du, Shan Zhao Published: 2026-07-26 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.23420v1 Large language models show strong promise for information extraction (IE), but existing reflection-based correction methods are often misaligned with structured extraction outputs. Free-form self-reflection can flag an error, yet it rarely identifies whether the failure is a missing span, wrong label, boundary mismatch, invalid relation type, or reversed argument order. We introduce LA-RL (Label-Aware Reflective Reinforcement Learning), an outcome-supervised framework that guides IE self-correction with task-grounded diagnostic labels. A single backbone first predicts an extraction, diagnoses task-specific error labels, and then revises its output conditioned on the diagnosis. Training starts from diagnostic data labeled by an annotation model for cold-start supervised fine-tuning and proceeds through two GRPO stages that reward final extraction quality, format validity, and first-pass correctness, without a process reward model. Experiments on named entity recognition, relation extraction, and event extraction show consistent same-backbone gains over SFT, including 6.83 average F1 on SciER relation extraction, about 20 F1 on out-of-distribution relation extraction, and 14.80 trigger F1 plus 17.50 argument F1 on DuEE1.0. Ablations show that reflection structure is task-sensitive: stronger constraints benefit relation extraction, whereas named entity recognition needs less restrictive correction under domain shift. --- ### Harmonized Interpretable ECG Waveform Features for Robust Cross-Dataset Clinical Prediction Authors: Jie Lin, Weijie Sun, Sunil V. Kalmady, Anita Khalafbeigi, Abram Hindle, Padma Kaul, Russell Greiner Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23412v1 Electrocardiograms (ECGs) are widely used for cardiovascular risk prediction, yet models often fail to transfer across hospitals because of protocol, population, and measurement differences. We benchmark cross-dataset generalization on three tasks - heart failure classification, 30-day all-cause mortality, and 30-day mortality among sinus-rhythm ECGs - using two large cohorts (MIMIC-IV and the Alberta Cohort). To reduce vendor-specific measurement mismatch, we build a harmonized, interpretable feature representation computed directly from raw waveforms: FeatureDB morphology/heart-rate-variability summaries plus compact time-frequency descriptors (autoregressive and wavelet features). We train XGBoost models on this unified feature space and evaluate with patient-disjoint internal and bidirectional external testing. We pre-specify two hypotheses: (H1) external AUROC retains at least 90% of source-site internal AUROC under transfer, and (H2) internal AUROC of the harmonized feature set stays within 10% of dataset-native machine-measurement models. Across tasks, internal AUROC is 0.79-0.82 and cross-dataset AUROC is 0.74-0.78, with larger and direction-dependent AUPRC shifts under transfer. As an exploratory benchmark, an end-to-end ConvNeXt model trained directly on raw ECG waveforms with age and sex achieves higher internal AUROC, while the harmonized representation remains competitive in relative cross-dataset transfer stability. These findings show that a consistent waveform-derived feature interface preserves performance, supports realistic external validation, and provides a transparent alternative for cross-site clinical prediction. --- ### NeurGO: Learning to Generate Elite Candidates for Meta-Black-Box Expensive Optimization Authors: Jintao He, Huixiang Zhen, Wenyin Gong Published: 2026-07-26 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.23408v1 Expensive black-box optimization is ubiquitous in science and engineering, where function evaluations are costly and the evaluation budget is limited. Traditional evolutionary algorithms and Meta-BlackBox Optimization (MetaBBO) approaches typically consume most evaluations on candidate selection, often wasting precious budget on inferior solutions. Although surrogate-assisted evolution and Bayesian optimization aim to reduce evaluations through surrogate models, constructing an accurate global model from limited data remains challenging, and model bias can easily trap the search in local optima. To overcome these limitations, we propose NeurGO, a generative MetaBBO framework that directly synthesizes elite candidates from historical population states. Specifically, we employ an attention-based encoder to capture the population-level search trend and condition a decoder on this representation to generate high-quality candidates, avoiding the expensive evaluation of large offspring pools. We then design a quality-diversity loss to maintain solution quality and population diversity throughout the search. Through extensive benchmarking on CEC 2008 and the COCO BBOB test suites, our method achieves better optimization performance under the same evaluation budget and exhibits faster convergence. --- ### Blood Pressure Estimation from PPG: A Comparative Study of Direct and ECG-Mediated Deep Learning Pipelines Authors: Bo Wu, Haoling Wang, Zhuodiao Kuang, Kateryna Shapovalenko Published: 2026-07-26 Categories: cs.LG, cs.AI, cs.ET, stat.AP arXiv: https://arxiv.org/abs/2607.23406v1 Continuous cuffless blood pressure (BP) monitoring is essential for connected health systems and wearable devices, enabling early detection, longitudinal tracking, and personalized management of cardiovascular disease. Many prior approaches attempt to estimate BP indirectly by reconstructing electrocardiography (ECG) from photoplethysmography (PPG), assuming ECG provides a stronger physiological link to BP. However, ECG sensing is less accessible in wearable settings and may introduce unnecessary complexity. In this work, we first perform a large-scale physiological correlation analysis on the MIMIC-III waveform database, revealing that PPG exhibits substantially stronger coupling with arterial blood pressure (ABP) ($|r|=0.247$, $p<0.001$) than ECG does ($r=0.018$, $p=0.187$), challenging the assumption that ECG provides a superior intermediate representation. Motivated by this insight, we conduct a systematic comparison between direct PPG-to-BP prediction and ECG-mediated pipelines using multiple state-of-the-art deep learning models. Across 1.74M segments from 3,127 patients, direct PPG-to-BP prediction achieves British Hypertension Society Grade A performance ($\mathrm{MAE}_{\mathrm{SBP}} = 4.82 mmHg$, $\mathrm{MAE}_{\mathrm{DBP}} = 4.31 mmHg$), outperforming all ECG-mediated approaches, which achieve only Grade B accuracy. Our findings suggest that accurate continuous BP monitoring can be achieved directly from wearable PPG signals, enabling simpler, more efficient pipelines for real-world connected health systems. --- ### Transfer Learning Architectures for Scalable Multi-Fidelity Bayesian Optimization Authors: Jaewook Lee, Ethan Errington, Christian D. Lorenz, Miao Guo Published: 2026-07-26 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.23404v1 Self-driving laboratories increasingly rely on multi-fidelity Bayesian optimization (MFBO) to balance cheap, approximate evaluations against scarce, expensive ones, with a predictive surrogate at its core. Gaussian processes (GPs) are the default choice, but they scale poorly as data accumulate and assume a smooth landscape that molecular and materials search spaces routinely violate. Transfer learning offers an alternative suited to this regime: it learns a representation from abundant cheap data and adapts it to sparse expensive data. Despite its use in property prediction, transfer learning has not been tested as the engine of a closed-loop optimization. Here we benchmark eleven transfer-learning surrogates against four GP methods under an identical selection rule, fidelity budget, and model size, across nine tasks spanning synthetic functions to real chemistry and materials problems. GPs win on smooth, low-dimensional functions but perform worst on molecular and materials problems, where transfer-learning surrogates reach substantially better solutions using far less computation. Because acquisition policy is held fixed across surrogates, this advantage is attributable to the surrogate itself. Uncertainty-driven exploration is not reliably beneficial, and calibration does not predict optimization performance, so greedy exploitation of the transfer-learned mean is the more robust default. Transfer learning is therefore the surrogate of choice for molecular and materials MFBO. --- ### A Statistical Difference between Single-Layer Learning and Hierarchical Learning in Wide Neural Networks Authors: Sumio Watanabe Published: 2026-07-26 Categories: cs.LG, math.ST, stat.ML arXiv: https://arxiv.org/abs/2607.23397v1 Hierarchical neural networks are widely used in artificial intelligence, yet their mathematical properties remain incompletely understood. In the infinite-width limit, two different theoretical frameworks have been proposed. One reduces deep learning to kernel regression with a fixed kernel by assuming that the parameters remain close to their initialization, whereas the other allows the parameters to move away from their initialization, requiring the kernel itself to be optimized. In this paper, we study a three-layer neural network with a finite but large number of hidden units. We show that training the input-to-hidden weights yields a smaller generalization error than keeping them fixed. Furthermore, the latter setting exhibits singularities in the parameter space, whereas the former does not. These findings indicate that singularities play an essential role even in wide neural networks. --- ### When the Preconditioning Exponent Turns Negative: Learning-Rate Coupling and Cross-Environment Generalization Authors: Gongyue Zhang, Honghai Liu Published: 2026-07-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30271v1 Adaptive optimizers are commonly parameterized by a fixed power of the second-moment estimate. Existing partially adaptive methods study exponents between momentum-like updates and the standard Adam square root, while the interaction between this exponent and the global learning rate is less understood. We perform a controlled cross-environment study using a paired four-environment classification problem with stable sparse features, environment-dependent spurious sparse features, dense features, and high-dimensional noise. Across \NumRuns{} source-training runs covering 21 preconditioning exponents $p\in(-0.5,0.5)$ and five learning rates $η\in(10^{-4},10^{-2})$, we find that the exponent maximizing cross-environment accuracy decreases almost linearly with $\log_{10}η$. The fitted slopes range from $-0.270$ to $-0.300$, with $R^2$ between $0.972$ and $0.996$. At $η=10^{-2}$, source-validation selection still prefers positive exponents in all four environments, whereas cross-environment and worst-environment criteria prefer negative exponents. Checkpoint decomposition shows that lower $p$ reduces the learned spurious-to-stable and noise-to-stable weight ratios; under reversed correlation, it also reduces the magnitude of the harmful spurious margin. Negative $p$ is therefore not a universally optimal setting. It is a high-step-size allocation regime produced by the joint action of learning rate and preconditioning. The study also exposes a model-selection conflict: source-domain validation systematically selects a different preconditioning regime from the one that maximizes robustness to environmental change. The results are a single-seed, finite-budget mechanism study rather than a broad benchmark claim. --- ### Towards Secure Cloud-Native Computing: Unveiling Kubernetes Misconfigurations with Large Language Models Authors: Mostafa Anouar Ghorab, Mohamed Aymen Saied Published: 2026-07-25 Categories: cs.CL, cs.SE arXiv: https://arxiv.org/abs/2609.20834v1 In the rapidly evolving landscape of cloud-native computing, Organizations are increasingly adopting infrastructure models that emphasize scalability, flexibility, and efficiency. Kubernetes has become the de facto standard for orchestrating containerized applications in these environments. However, the inherent complexity of cloud-native ecosystems introduces significant challenges, particularly in the form of misconfigurations that can compromise both security and performance. This study explores the potential of Large Language Models (LLMs) in identifying Kubernetes misconfigurations. We introduce a comprehensive taxonomy of common misconfiguration types, offering a structured framework to better understand and categorize these issues. Additionally, we conduct an empirical evaluation of state-of-the-art detection tools to benchmark their effectiveness. Furthermore, we analyze the Kubernetes objects most prone to misconfiguration and evaluate the severity of the identified issues. By leveraging advanced machine learning techniques, including LLMs, we provide novel insights into enhancing misconfiguration detection methodologies. --- ### Scene-Conditioned Relation Routing for urban cellular activity forecasting Authors: Qingzhong Li, Jingye Lin, Hui Ma, Yajun Zhang, Xinjun Pei, Ming Yan, Fei Xing Published: 2026-07-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.20209v1 Urban cellular activity forecasting requires jointly modeling heterogeneous spatiotemporal signals, including SMS usage, mobile network traffic, and call activity. Existing methods often separate temporal modeling, spatial relation learning, and multi-signal prediction, relying on fixed graph structures or static multi-task learning schemes, which limits their adaptability to changing urban scenes. We propose SCRR-Net, a scene-conditioned spatial relation routing framework in which urban contextual information jointly controls spatial dependency selection and cross-task knowledge transfer. SCRR-Net includes a context encoder, a spatial graph expert routing module, a temporal Transformer encoder, and a task knowledge routing module. Experiments on the Milano and Trento datasets demonstrate that SCRR-Net consistently outperforms competing methods on SMS, network traffic, and call activity forecasting, while providing interpretable routing behaviors. --- ### Foundations of Stochastic Lexical Calculus: Semantic Descent and Random Dynamics on Probability Simplices Authors: Matthew F Dixon Published: 2026-07-25 Categories: cs.CL, math.CT, math.PR, stat.ML arXiv: https://arxiv.org/abs/2609.20207v1 Large language models produce prompt-dependent probabilities over words, whereas scientific systems require uncertainty over meaningful states that can be updated as evidence arrives. We develop an observable framework for determining when language-derived probabilities support such a sequential state representation. Theoretically, we define typed measurable transformations of contextual language, construct a minimal closed representation, and give necessary and sufficient conditions for semantic updates to exist uniquely. We bound irreducible nonclosure and accumulated error, and under average contraction prove existence, uniqueness and stability of an external random recursion on a probability simplex. These results define a stochastic lexical calculus without attributing an internal calculus to the language model. Empirically, frozen experiments test the observable implications. Raw prompt-conditioned probabilities fail the prespecified invariance gate; after prompt-specific calibration, a common three-state representation passes the stability gates and covers 28 of 30 untouched eight-step paths, or 0.933 at nominal level 0.90. Accordingly, language probabilities support a stochastic state only conditionally on verified closure, stability and coverage within a declared operating domain. --- ### JointMatch: A Unified Heterogeneous Graph Neural Solver for Large-Scale Ride-Sharing Matching Authors: Kun Zhao, Xu Chen Published: 2026-07-25 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.20200v1 Ride-sharing platforms must continuously decide which open requests to bundle into shared trips and which idle vehicles should serve them. The dominant academic approach decomposes this into two sequential matching problems -- request pairing first, then vehicle assignment -- and applies a separate solver to each. This decomposition is convenient computationally but loses revenue and scales poorly because the first stage commits to ride bundles before the available vehicles are known. We propose JointMatch, a learning-based framework that handles request pairing and vehicle assignment together on a single graph. The graph is sparsified by spatial proximity so that its size grows linearly rather than quadratically with the number of vehicles and requests, and a graph neural network scores all candidate decisions in one forward pass. On the New York City Yellow Taxi data, the framework already exceeds both the classical Blossom heuristic and a faithfully-trained two-stage GNN baseline -- often by a wide margin -- and at city scale (fleet 10000) it runs more than $20\times$ faster per dispatch epoch than either. A supervised training stage closes most of the remaining revenue gap, and a policy-gradient fine-tune aligns the trained model with realised revenue. --- ### Towards a Unified Modality-Agnostic Multimodal Framework for Cognitive Workload Assessment Authors: Stefanos Gkikas, Christian Arzate Cruz, Calvin Joseph, Giorgos Giannakakis, Raul Fernandez Rojas Published: 2026-07-25 Categories: cs.LG, cs.HC arXiv: https://arxiv.org/abs/2609.20199v1 Cognitive workload reflects the mental effort required during task performance and is central to the design of adaptive human-machine systems. The use of biosignals to measure cognitive workload has been extensively researched and documented; however, studies examining the effects of combining heterogeneous biosignal modalities for this purpose remain limited. To provide insight into this area, we developed a unified, modality-agnostic, hierarchical Transformer-based architecture to process heterogeneous biosignal modalities within a single model. We use this framework in a pilot study evaluating all $31$ possible combinations of five modalities: Electrocardiogram (ECG), Electrodermal Activity (EDA), Respiration (RESP), Peripheral Oxygen Saturation (SpO$_2$), and Electroencephalogram (EEG), under leave-one-subject-out validation across three cognitively distinct tasks: abstract reasoning (IQ), arithmetic problem solving (MATH), and a game task (GAME). In this pilot setting, the results suggest that: (i) EEG is the strongest single modality, ranking highest in IQ, GAME, and the pooled ALL setting, where samples from all three tasks are combined; (ii) adding more modalities does not consistently improve performance; (iii) the full five-modality combination achieves the highest \textit{Average} score of $73.02%$ on IQ and $68.08%$ when the \textit{Average} scores are averaged over the four evaluation settings: IQ, MATH, GAME, and ALL; and (iv) the proposed method reduces model size by approximately $50%$ compared with late-fusion alternatives while maintaining a lower inference time. --- ### Adaptive Interpolatory Curve Subdivision with Learned Local Angles Authors: Hassan Ugail, Newton Howard Published: 2026-07-25 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.17566v1 Curve subdivision is pivotal in computer graphics for generating smooth geometric objects from control polygons. Interpolatory subdivision is especially attractive because the refined curve is guaranteed to pass through the designer's control points. Classical four-point and six-point schemes preserve this property, but their behaviour is governed by a single global tension parameter, limiting their ability to adapt across flat regions, sharp turns and varying local geometries. We introduce an adaptive local-angle formulation that keeps the interpolatory structure intact while learning how each new vertex should be inserted. A compact edge-wise predictor assigns one insertion angle per edge, while the original vertices are copied exactly at every refinement level. Interpolation is therefore a structural property of the operator and does not depend on the trained weights. The same predictor is used with geometry-specific geodesic primitives on the Euclidean plane, the two-sphere and the Poincaré disk. Under a matched-density evaluation protocol, the method reduces nearest-neighbour error by factors of five to seventeen over the best validation-tuned fixed-tension baseline, and by about 1.8 over centripetal Catmull-Rom in the Euclidean case. It also substantially reduces bending energy and tangent roughness, while remaining competitive with separately trained per-geometry models. --- ### A Heisenberg Lift Descriptor for Order Sensitive Online Handwriting Recognition Authors: Hassan Ugail, Newton Howard Published: 2026-07-25 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.17565v1 Online handwriting recognition systems typically represent pen trajectories through fixed-length Euclidean shape descriptors that capture the spatial outline of each stroke, but are insensitive to the order in which that outline is produced. Two strokes that trace the same region of the plane in opposite directions are indistinguishable to any such order-blind representation, yet their traversal directions may carry decisive class information in characters where loop orientation and stroke sequencing matter. This paper introduces a Heisenberg-lift framework that addresses this gap through a compact, interpretable, order-sensitive augmentation for online pen-trajectory features. The simplest instance is the terminal signed area, a single parameter-free scalar appended to an existing Euclidean descriptor at negligible computational cost. Evaluated on two standard online handwriting benchmarks, this one-scalar addition, consistently raises classifier accuracy over the Euclidean baseline. On the hardest character pair in our study, the letters o and y, the signed area alone achieves perfect separation while the Euclidean baseline falls short. The advantage grows further under additive coordinate noise, a practically relevant degradation in pen-trajectory data. A richer fifteen-dimensional extension, derived from a noncommutative Heisenberg-group subdivision scheme, provides additional gains in noisy and loop-structured conditions. Dimension-matched statistical controls confirm that all improvements reflect geometric information rather than feature-count inflation. The resulting descriptor is lightweight, closed-form, and directly interpretable, making it a practical augmentation for online handwriting and related document-trajectory classification pipelines in which the direction of stroke execution carries discriminative information. --- ### CTAN: Cycle-Temporal Attention Network for Embodied Audio-Visual Navigation Authors: Teng Liu, Yinfeng Yu Published: 2026-07-25 Categories: cs.MM, cs.AI, cs.SD, eess.SP arXiv: https://arxiv.org/abs/2609.17420v1 Audio-visual embodied navigation equips robots with the capability to infer the locations of sound sources by integrating visual inputs and acoustic information (e.g., depth observations and binaural audio cues). The core challenge lies in establishing effective semantic interactions across heterogeneous modalities (which exhibit distinct feature distributions). Existing feature fusion strategies, however, often rely on simple multimodal aggregation and therefore fail to capture the underlying geometric and semantic relationships, leading to information degradation in complex environments. To overcome these limitations, this work presents the Cycle-Temporal Attention Network (CTAN), a framework designed for active semantic-enhanced fusion (rather than straightforward multimodal combination). Specifically, the proposed Audio-Visual Reconstruction Cross-Attention (AVRCA) module employs a bidirectional cycle-consistency constraint (between visual and acoustic representations) to reinforce the spatial semantic attributes of both modalities, thereby facilitating more robust cross-modal interaction. Additionally, we design a Temporal Cross-Modal Memory (TCMM) mechanism to dynamically integrate real-time enhanced multimodal features with historical context, reducing performance drops caused by auditory dead zones. Experimental results obtained on the Replica and Matterport3D benchmarks indicate that the proposed approach achieves superior performance over previous audio-visual navigation methods in terms of success rate (SR), success weighted by path length (SPL), and scene navigation accuracy (SNA). --- ### A derivative-fidelity failure mode in physics-informed neural networks: strengthened benchmark evidence from function-value training Authors: Koji Koyamada Published: 2026-07-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.13171v1 Physics-informed neural networks (PINNs) use automatic differentiation to impose differential-equation residuals, but good agreement in function values does not necessarily imply accurate derivatives. This paper formulates derivative fidelity as a failure mode of PINNs and tests it with one-dimensional benchmarks. Multilayer perceptrons are trained only on function values for sin(x) and exp(x), while second derivatives obtained by automatic differentiation are evaluated separately. The hypothesis is strengthened by additional tests over training-point density, activation functions, endpoint-dense evaluation, and both L2 and maximum-error diagnostics. The results show that visually accurate function approximation can coexist with substantially larger second-derivative errors, especially near high-curvature boundary regions. The experiment provides a diagnostic protocol for distinguishing value accuracy from physics-residual reliability. --- ### Harnessing agent memory to build lifelong AI partners for materials scientists Authors: Siyu Liu, Bo Hu, Beilin Ye, He Cao, David J. Srolovitz, Tongqi Wen Published: 2026-07-25 Categories: cs.AI, cond-mat.mtrl-sci, cs.CE, cs.CL, cs.MA arXiv: https://arxiv.org/abs/2608.11224v1 Materials research advances through accumulated experience - scripts that work, protocols that are trusted, warnings attached to failed calculations or experiments, and judgement that links a new question to an old result. This experience is essential for reproducibility and knowledge transfer, yet it is usually fragmented across notebooks, repositories, job logs and individual memory, and it is rarely portable across artificial-intelligence agents. Here we argue that a lifelong AI partner for materials science can be designed around persistent memory rather than around a particular agent implementation. We introduce a self-evolving memory framework that stores scientific experience as inspectable facts and executable skills, so that observations, failure boundaries, protocols and validation checks can be retrieved, revised and migrated across models. We evaluate the idea in three computational settings that expose different layers of materials-research competence. In 49 real-world materials-tool-use questions comprising 138 executable subtasks, memory nearly doubles GPT-5.2 task success without model-parameter updates. In elemental-solid equation-of-state calculations, memory converts a wavefunction-initialization failure into a pre-execution guardrail, improving outcomes from 22/1/4 to 25/2/0 Correct/Partial/Error and avoiding 92% of repeated errors. In 13 practical material simulation workflows, remembered skills and failure facts halve the aggregate trace burden (tokens) and reduce tool calls by over a factor of two by the third round, while preserving physically meaningful outputs in band-gap, phonon, vacancy and work-function analyses. These results show that agent memory can serve as a durable scientific asset; a portable, self-improving record of materials-research experience that outlives any single model or agent stack. --- ### Identifiability-Aware Source Apportionment in City-Scale Advection-Diffusion Systems Authors: Ankit Bhardwaj, Lakshminarayanan Subramanian Published: 2026-07-25 Categories: eess.SP, cs.CE, cs.LG arXiv: https://arxiv.org/abs/2608.00050v1 Source apportionment from sparse urban air-quality sensors is an inverse problem limited by sensor placement, wind-driven transport, background variation, and noise. Known or proxy emission inventories make attribution meaningful by restricting the unknown source field to a finite set of candidate groups, but do not guarantee those groups are distinguishable from the observations. We represent time-varying source activity with a low-dimensional nonnegative temporal basis and formulate inventory-based apportionment as a wind-conditioned lagged inverse problem in which each source--basis coefficient produces a sensor-time fingerprint. After projecting out a separate low-dimensional background space, the relevant object is the projected lagged response matrix $\widetilde H_Φ$: exact identifiability at the chosen basis resolution requires its full column rank, while noise-robust attribution is controlled by its singular values, coefficient visibility, background absorption, pairwise coherence, and ray distance. We propose an identifiability-aware apportionment (IASA) framework that estimates nonnegative source--basis coefficients, reconstructs activity trajectories, and reports uncertainty and conservative grouping recommendations for indistinguishable sources. We instantiate it on a New Delhi platform built from government PM$_{2.5}$ and wind records, regulatory sensor locations, and four proxy source groups, and define controlled and observed evaluations of recovery, ambiguity, wind diversity, background stress, transport error, inventory robustness, and residual adequacy. IASA reports the attribution resolution defensible under the declared inventories, transport, background, lag, and noise rather than the most detailed possible vector. --- ### Not All EEG Moments Are Equal: Position-Adaptive Time Scheduling for EEG Generation Authors: Boheng Liu, Ziyu Li, Chenghua Duan, Qing Li, Xia Wu Published: 2026-07-25 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.00048v1 Electroencephalography (EEG) generation is essential for alleviating data scarcity and enabling large scale neural modeling in brain computer interface applications. However, existing flow based approaches assume that every channel and every time segment within a sample shares a single global time progression, overlooking the fact that not all EEG moments are equal. To address this overlooked heterogeneity, we propose an adaptive EEG generation framework built on conditional flow matching. The framework introduces Position-Adaptive Time Scheduling, which tracks per position reconstruction error to modulate a position specific time progress within the flow matching trajectory. It further incorporates Factorized Spatio-Temporal Attention and a frequency aligned multi resolution spectral consistency loss to model inter channel dependencies induced by volume conduction and compensate for the power law spectral bias of EEG, thereby improving the quality of generated signals. Extensive experiments on three EEG datasets with distinct acquisition protocols and task semantics show that our framework consistently outperforms the strongest baseline, reducing TS-FID by up to 62.2\% and improving downstream classification accuracy gain by up to 6.77 percentage points. These results suggest that the proposed method represents a promising step toward scalable, high fidelity data augmentation for real world brain computer interface applications. --- ### Fracture Risk Prediction in Adults Over 50 Years Old Using DXA and EHR: Comparison of Traditional and Machine Learning Models in Two Large Cohorts Authors: Jiahe Qian, Hao Dai, Kunyu Yu, Hexin Dong, Xing He, Erik A. Imel, Jiang Bian, Yifan Peng, Yi Liu Published: 2026-07-25 Categories: stat.AP, cs.LG arXiv: https://arxiv.org/abs/2607.28671v1 Accurate fracture risk prediction is important for osteoporosis management, but commonly used clinical tools may not fully use information available in electronic health records (EHRs) and dual-energy X-ray absorptiometry (DXA) reports. We developed and externally validated time-to-event fracture prediction models among adults aged 50 years or older with clinically obtained DXA reports in 2 US health care systems. The development cohort was derived from NewYork-Presbyterian/Weill Cornell Medical Center and the external validation cohort from the Indiana Network for Patient Care. Predictors included demographics, lifestyle factors, prior fracture, comorbidities, medication exposures, osteoporosis treatment history, and DXA-derived T-scores extracted from radiology reports. The outcome was time from index DXA to first incident fragility fracture identified from structured diagnosis codes. We evaluated penalized Cox regression, random survival forest, gradient-boosting survival, and XGBoost survival models using 2 prespecified predictor settings and compared discrimination with clinically reported FRAX major osteoporotic fracture probabilities. The development cohort included 11,510 adults, of whom 858 sustained incident fragility fractures; the external validation cohort included 1,932 adults, of whom 180 sustained fractures. In internal validation, the expanded Cox model achieved a mean Harrell C-index of 0.779, compared with 0.653 for FRAX. In external validation, the corresponding Cox model achieved a Harrell C-index of 0.714, compared with 0.590 for FRAX; gradient-boosting survival had the highest external discrimination (0.725). EHR- and DXA-enhanced models showed better discrimination than clinically reported FRAX scores in this DXA-tested population, but calibration assessment, prospective evaluation, and implementation workflow assessment are needed before clinical use. --- ### Hierarchical Copula-Gumbel-Top-K Routing: Two-Sided Dependence Control for Frozen Mixture-of-Experts at Fixed Per-Token Routing Laws Authors: Richard Yi Da Xu Published: 2026-07-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.28670v3 A stochastic Gumbel-Top-K router defines, for every token of a mixture-of-experts (MoE) model, a routing law: a distribution over ordered expert lists and mixture weights. We ask which joint distributions over the routing choices of different tokens are reachable while every individual token's complete routing law is held exactly fixed. We give a two-sided construction, Hierarchical Copula-Gumbel-Top-K (CGA). Within a group of related tokens, an exchangeable Gaussian copula positively correlates the Gumbel perturbations at each expert coordinate, which can increase within-group expert-set coherence. Across disjoint pairs of groups, a tunable antithetic construction introduces a selectable amount of negative dependence. We prove that both operations leave each token's ordered Top-K sample, mixture weights, and inclusion probabilities identical in distribution to independent routing at a routing layer conditioned on its pre-routing logits; conditional expected expert traffic is preserved as a consequence. We characterize the resulting trade-off: positive within-group coupling can only inflate the variance of realized expert loads relative to independent routing, while nonnegative cross-group opposition can only reduce it relative to flat coupling at the same within-group strength. Coherence and load dispersion are thus controlled by two complementary dependence dials on the invariance constraint surface. Because the base model is untouched, the dials can be driven by a small controller over frozen features, trainable with a score-function estimator: the frozen network is evaluated only in the forward direction, and gradients are confined to the controller. An initial small-scale pilot validates the mechanism and the training route, but does not establish task-level fine-tuning gains. --- ### Hierarchical Copula-Gumbel-Top-\texorpdfstring{$K$}{K} Routing: Two-Sided Dependence Control for Frozen Mixture-of-Experts at Fixed Per-Token Routing Laws Authors: Richard Yi Da Xu Published: 2026-07-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.28670v1 A stochastic Gumbel-Top-$K$ router defines, for every token of a mixture-of-experts (MoE) model, a \emph{routing law}: a distribution over ordered expert lists and mixture weights. We ask which \emph{joint} distributions over the routing choices of different tokens are reachable while every individual token's complete routing law is held exactly fixed. We give a two-sided construction, \emph{Hierarchical Copula-Gumbel-Top-$K$} (\CGA{}). Within a group of related tokens, an exchangeable Gaussian copula positively correlates the Gumbel perturbations at each expert coordinate, which can increase within-group expert-set coherence. Across disjoint pairs of groups, a tunable antithetic construction introduces a selectable amount of negative dependence. We prove that both operations leave each token's ordered Top-$K$ sample, mixture weights, and inclusion probabilities identical in distribution to independent routing \emph{at a routing layer conditioned on its pre-routing logits}; conditional expected expert traffic is preserved as a consequence. We characterize the resulting trade-off: positive within-group coupling can only inflate the variance of realized expert loads relative to independent routing, while nonnegative cross-group opposition can only reduce it relative to flat coupling at the same within-group strength. Coherence and load dispersion are thus controlled by two complementary dependence dials on the invariance constraint surface. Because the base model is untouched, the dials can be driven by a small controller over frozen features, trainable with a score-function estimator: the frozen network is evaluated only in the forward direction, and gradients are confined to the controller. An initial small-scale pilot validates the mechanism and the training route, but does not establish task-level fine-tuning gains. --- ### LARA: Lightweight Adapters in the Residual Stream for Composable Adaptation and Alignment Authors: Pascal Ekin, Hyosun Choi, Wei Jie Published: 2026-07-25 Categories: cs.LG arXiv: https://arxiv.org/abs/2607.28669v1 We present LARA (Lightweight Additive Residual Adaptation), a method for efficient adaptation that operates in the residual stream of a frozen model rather than in its weights. Where LoRA adds an update of low rank to weight matrices, LARA reads the hidden state at a small set of layers and adds a correction of low rank back to the residual stream, leaving all base weights untouched. On a code fine-tuning task and on preference optimization (DPO), LARA matches LoRA at equal parameter counts. Because adaptation is a frozen base plus a residual, LARA exposes a scale γ, applied at inference, that interpolates smoothly between base and adapted behavior, a form of graded control that adaptation in weight space does not offer. Finally, because each behavior is a small residual module over a shared frozen base, many behaviors can be held resident at once and routed automatically per token. We place seven behaviors, six fine-tuned and one optimized for preference, on one frozen 1.5B model for roughly 33 MB of overhead, against one full model for each behavior. Because the base is untouched, behaviors are trained separately and selected per token rather than loaded on demand, which suits hosting many behaviors, and adding new ones, on a single model on a device. --- ### Guarantees on Dynamical System Distinguishability for LLM Token Generation Authors: Mohamed Akrout, Dan Wilson Published: 2026-07-25 Categories: cs.LG, cs.AI, math.DS arXiv: https://arxiv.org/abs/2607.28667v1 Recent work has shown that classifying large language models (LLMs)' responses can be distinguished by modeling token embeddings as trajectories of a black-box dynamical system (DS) and comparing prediction residuals of two DSs. Despite the empirical success of this dynamical approach, a theoretical understanding of why it works, how well it scales as a function of the token sequence, and when it transfers across embedding models remains lacking. We address these questions by formalizing the classification task as a binary hypothesis test between two stochastic linear DSs. We show that the total variation distance between the stationary marginal distributions of the two DSs can be arbitrarily small even when the dynamics differ substantially, which provides a fundamental accuracy floor for any classifier that ignores token dynamics. We then show that the misclassification probability of DS-based classification decays exponentially in the sequence length $L$, with the decay governed by a dynamical discriminability quantity $δ^2$ that captures the spectral distance between the two DSs. We also characterize cross-embedding generalization by introducing an approximate intertwining condition between embedding models and establishing a lower bound on the transferable discriminability in terms of the intertwining map's smallest singular value. Together, these results explain the empirical performance of DS-based classification and motivate further investigation into using DS theory to analyze AI systems, in contrast to the more common approach of using AI to model dynamical systems. --- ### The Checking Problem: What must be true before AI ships in a regulated firm Authors: Prerit Ahuja Published: 2026-07-25 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2607.28666v1 Enterprise AI programmes stall at a rate that is widely quoted and poorly explained. This paper measures the mechanism. Six document-heavy workflows of the kind performed daily in regulated financial services were run across four model families and three tool configurations, three times each, producing 5,093 scored output elements across 72 configurations. Each configuration was assessed twice: against a demonstration bar, being a single correct run on a single case, and against a production bar requiring sustained accuracy, reproducibility across repeats, verifiable attribution, and a confidence signal that carries information. 57 of 72 configurations cleared the demonstration bar and 32 cleared the production bar, a survival rate of 56.1%. The paper then computes the review burden each configuration imposes, estimated out of sample rather than with hindsight. A tool that states no confidence requires review of 100% of its output, because it offers a reviewer no basis for triage. Requiring the tool to cite its sources and state a confidence reduces that to 49% while holding the residual error tolerance in 17 of 20 configurations. Adding a self-verification pass costs 2.3 times the latency of the plain configuration, reaches 44%, and is the only configuration that fails to hold the error tolerance. The practical implication is that the value of an AI workflow is set less by how often it is right than by how much of it a human must still check, and that the second property is measurable and rarely measured. --- ### Sensitivity Analysis of GRU, LSTM and Transformer Encoder in Classification of Automated Driving Systems Authors: Bidhya Shrestha, Christos Papadopoulos Published: 2026-07-25 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2607.28665v1 Automated driving systems (ADSs) are becoming ubiquitous. Future Software Defined Vehicles (SDVs) may be able to run multiple ADSs, both native and aftermarket such as Comma.ai's Openpilot. Monitoring systems to independently verify which automated driving system is active are important for safety monitoring, regulatory compliance, insurance assessment, and anomaly detection. In this paper, we first evaluate the effectiveness of three sequence-based classification models: Gated Recurrent Units (GRU), Long Short-Term Memory (LSTM) networks, and a Transformer encoder model for identifying Level 2 automated driving systems using vehicle telematics data alone: Comma Openpilot, Tesla Autopilot, and Cadillac Super Cruise, along with manual driving. All three models achieve strong clean-data performance with macro F1-scores of 0.92 (GRU), 0.90 (LSTM), and 0.93 (Transformer encoder model) when trained on clean data; threat-matched training yields 0.904-0.916 macro F1 with only a modest clean-data penalty. Second, we introduce a modular robustness evaluation framework that simulates realistic telematics degradation through five corruption families at five severity levels (L1-L5). Continuous channels are perturbed using additive white Gaussian noise with cumulative drift, correlated cross-channel noise, and temporal jitter. Binary event signals are subjected to burst loss, delayed transitions, spurious toggles and cross-feature inconsistencies inspired by communication errors. Robustness is measured using macro-F1, which gives equal weight to each class and is suitable for imbalanced multiclass evaluation. Our evaluation reveals a sharp failure-mode split: event-level corruptions reduce macro-F1 only slightly (greater than equal to 0.87 at L5), while temporal jitter collapses macro-F1 to 0.44-0.50 across GRU, LSTM, and Transformer encoder model. ---