# Research Papers — 2026-10 · bundle page 6 of 156 > Full-text markdown of papers #251–300 of 7799 published in 2026-10, newest first. AI-readable mirrors of the full 7799-paper 2026-10 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-10/page/6 [← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/5.md) [Next bundle →](https://aipod.dev/papers/archive/2026-10/page/7.md) ### Selection Is Not Learning: Choosing Model Futures Authors: Haoyi Yang, Kshitij Mishra, Nils Lukas, Salem Lahlou Published: 2026-10-15 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.22029v1 Self-evolving language models train on tasks they generate, so every round must decide which updated model the next round inherits. Current loops such as R-Zero, Agent0, and Skill Self-Play make this decision through the task, admitting tasks that look learnable and inheriting every resulting update. We identify a selection-to-update gap in these loops, where a task that looks learnable need not produce a useful update. Most updates on verified tasks fail to make correct answers more likely, yet these loops inherit them all. HeirCommit freezes the updated model and uses fresh evaluation questions to decide whether to inherit it or retain the current model. EvoHeir separates parameter inheritance from experience inheritance, using current and archived experience to repair failures and preserve successes before final confirmation. Our analysis characterizes selection-induced optimism and decomposes regret into selection, absorption, and confirmation errors. HeirCommit cuts the harmful share of accepted updates by 68-83% relative to random inheritance at the same acceptance rate. Across four backbones, EvoHeir achieves the best mathematics and general-reasoning averages, raising the math average by 5.3% relative to the strongest baseline, Agent0. The framework lets useful experience persist even when the model that produced it is not inherited. --- ### Augment the Cache, Replace the Activation: A 2x2 Study of Internal-State Sharing Authors: Adam Deng, Shicheng Liu, Yada Zhu Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.22026v1 Multi-agent systems built from frozen language models communicate through text, which discards the internal state the sender already computed and makes the receiver pay to rebuild it. Work on internal channels avoids that cost, framing the problem as transfer: agent A sends information to agent B. Every agent pair that must communicate needs its own transfer bridge, and every agent added to a system needs a new bridge fitted against each previous agent. We reframe this problem as sharing: an agent publishes its internal state once into a pool, and every peer reads that pool through a single mechanism fitted to no sender in particular, so adding an agent adds a publisher and no new bridge. Whereas the carrier of the internal state and the fusion mechanism are usually chosen independently in the literature, our main finding is that these two choices are coupled; the deciding factor is where the carrier enters the receiver's computation. Consumability is a property of the receiving computation rather than of the representation. We find that the key-value (KV) cache must be augmented rather than swapped in: the receiver cross-attends over a peer's cache while keeping its own, and with only a small query-side adapter trained, matches the accuracy it gets from reading the source document. Activations invert this. They are best written straight into the receiver, with no bridge, no adapter and no training, which beats exposing them alongside the receiver's own, though that advantage rests on knowing where in the receiver the state belongs. The rule is augment the KV cache, replace the activation, established under one protocol with a shared control ladder. One adapter, trained once, also serves publishers it never trained on, and a pool of several at once, but only those whose internal state lies close to the receiver's own. --- ### Not All Unlabeled Data Are Equal: Unlabeled Data Selection for Semi-Supervised QSAR Modeling with Large Chemical Repositories Authors: Oliver Majer, Saso Dzeroski, Jurica Levatic Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.22025v1 Semi-supervised learning (SSL) is particularly attractive in domains where labels are expensive to obtain, while unlabeled examples are readily available. Quantitative Structure-Activity Relationship (QSAR) modeling, which predicts molecular biological activity from chemical structure, is one such domain. Obtaining activity measurements requires costly wet-lab experiments, resulting in small labeled datasets, while chemical repositories such as ChEMBL contain millions of unlabeled molecules. Despite this apparent suitability, the potential of large chemical databases to improve QSAR models via SSL remains largely unexplored. Furthermore, it is unclear how unlabeled data should be selected in realistic SSL settings with genuine unlabeled data. We conduct a systematic study of unlabeled-data selection using 10 QSAR datasets extracted from ChEMBL and a pool of more than 400,000 unlabeled molecules. To this end, we evaluate three established SSL algorithms (self-training, co-training, and label propagation) and formulate four unlabeled-data selection strategies based on random, nearest-neighbor, furthest-neighbor, and mixed sampling. Our results show that unlabeled data from large chemical repositories can improve QSAR models, with at least one SSL configuration outperforming the supervised baseline on each of the 10 datasets. Moreover, the most reliable improvements are generally achieved by selecting a relatively small number of chemically distant molecules. This suggests that the composition of the unlabeled set may be more important than its size, challenging the common premise that SSL necessitates an abundance of unlabeled data. --- ### DemoStream: Batched Task-and-Motion-Plan Optimization for Automated Learning from Demonstration Authors: Caelan Garrett, Neel Jawale, Fabio Ramos Published: 2026-10-15 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.22020v1 Learning from demonstration is an effective strategy for endowing robots with manipulation capabilities but requires training on datasets comprised of massive amounts of demonstrations. Existing demonstration collection approaches either require constant human involvement, for example, via teleoperation or are specialized for a single task. Task and Motion Planning (TAMP) systems are able to autonomously solve and therefore generate demonstrations for general-purpose tasks; however, learning from their demonstrations has proven difficult due to multi-modality in their solutions. In this work, we build on recent work leveraging GPU acceleration for TAMP and propose DemoStream, a TAMP system that optimizes demonstration quality in order to improve behavior cloning success rates. First, we introduce batched TAMP sampling operations along with a new TAMP algorithm that correspondingly performs batched constraint satisfaction in order to explore many solutions in parallel. We conduct simulation experiments ablating different action spaces, cost functions, and batch sizes to analyze the effect of these choices on downstream policy success rates. Finally, we show that we can leverage the safety guarantees of TAMP to directly collect real-world demonstrations at an average rate of 66 demos per hour. See our website for policy rollout videos: demostream.github.io. --- ### Priming Base Models for Agentic Post-Training: An Open Midtraining Recipe Authors: Soumye Singhal, Joseph Jennings, Ka Shun Shum, Alexander Bukharin, Khushi Bhardwaj, Zirui Liu, Venkat Srinivasan, Jonathan Li, Yash Jain, Kezhi Kong, Markus Kliegl, Olivier Delalleau et al. Published: 2026-10-15 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.22019v1 Frontier models, general-purpose and domain-specific alike, now include a midtraining stage between pretraining and post-training, but little has been published about how this stage is built. We present an open midtraining recipe that primes base models for agentic post-training, and we show its effectiveness on two Nemotron models of different sizes. Pretraining typically runs at short context and sees almost no agent interactions, while SFT data is mostly synthetic, and very long SFT stages can hurt the model's later RL trainability. Our midtraining stage sits between the two. It trains at 256K context for hundreds of billions of tokens on a diverse mixture of natural software-engineering data like packed repositories, pull requests, commit histories, and build logs together with a small share of agentic traces. On Nemotron 3.5 Lightning, 500B midtraining tokens improve SWE-bench Verified from 60.7 to 64.5, SWE-bench Multilingual from 55.8 to 62.8, and SWE-bench Pro from 32.0 to 35.5 after identical full SFT. On the larger Nemotron 3.5 Super, 280B midtraining tokens improve DeepSWE, a harder benchmark of long-horizon tasks, from 11.7 to 15.7 and SWE-bench Verified from 72.9 to 74.3 after identical full SFT. Midtraining also improves robustness across agent harnesses on both models. We report the ablations behind the recipe and an evaluation protocol built on short fixed SFT runs, since base-model proxy metrics did not correlate reliably with post-training results. --- ### When Models Misread Ink: Auditing VLMs for Arabic Calligraphy Recognition Authors: Niaz Mohaiman Abtahi, Evan Maksud, Reyana Shahid Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.22018v1 Arabic calligraphy is a high-stakes cultural visual domain for Muslim communities, yet VLM evaluations rarely test its fine-grained script traditions. We audit 12 VLMs on a cleaned 1,195-image six-way benchmark using numeric, visual-cue, and Diwani-first prompts. In this closed-set setting, the best numeric-prompt accuracy is 60.9% for closed/API models and 47.2% for open-weight models. The main failure is default-label collapse: every numeric run is dominated by a single label taking 33-63% of predictions, against an 18.8% majority-class baseline. For 8 of 12 models the default is Thuluth; GPT-4o labels 52.9% of all images as Thuluth although it is only 18.1% of the set. Prompting helps unevenly: visual cues move accuracy by more than 10 points for some models but hurt others, and foregrounding Diwani often relocates the default rather than removing it. Arabic calligraphy recognition is therefore a compact, culturally grounded stress test for VLM robustness and for measuring which visual traditions become legible to automated systems. --- ### CARA: Causally Guided Retrieval Agent for Question Answering Authors: Weicheng Liu, Xiangcheng Wu, Depeng Xu, Haopeng Zhang, Xi Niu Published: 2026-10-15 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2610.22017v1 Retrieval-augmented generation over text-rich knowledge graphs requires identifying evidence that contributes to answering a question, rather than merely matching its textual content. Existing agentic retrievers adaptively balance global search and multi-hop exploration but primarily rely on textual similarity to guide retrieval. We propose CARA, a training-free, CAusally guided Retrieval Agent that explicitly evaluates the causal relevance of retrieved evidence through two complementary criteria: consequentiality, measuring whether adding evidence improves answerability, and essentiality, assessing whether removing evidence weakens answerability. CARA integrates these assessments with path consistency to guide graph exploration and employs reciprocal rank fusion to aggregate evidence from multiple agents. We evaluate CARA on three STaRK datasets and augment STaRK with ground-truth answers to enable downstream question answering evaluation. Experimental results demonstrate that CARA consistently outperforms existing retrievers. CARA also improves answer quality and faithfulness while examining fewer graph nodes and requiring fewer retrieval operations. These findings demonstrate that causal relevance assessment enables more focused and effective agentic graph retrieval, improving both retrieval performance and downstream question answering. --- ### Close Enough is Nowhere Near: Fidelity-Consumability Decoupling in KV Caches Authors: Adam Deng, Shicheng Liu, Yada Zhu Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.22016v1 Multi-agent LLM systems communicate in text, which is lossy and wasteful: the sender throws away the reading it already did. A growing literature instead transmits the KV cache, the natural carrier of that reading. We study the most appealing form of the idea: both models stay frozen and the sender's cache is substituted for the reader's with nothing trained. It does not work, and the obstruction belongs to the reader. A frozen transformer reads only caches close to the one it would have written itself, a region we call its tolerable set. We show that geometric similarity to that cache does not predict whether the reader can use it, and the failure is not gradual, so a more faithful substitute is not the remedy it appears to be. We explain this phenomenon with competitive attention. Experimentally, we test eight diverse substitution methods against two measured anchors: an uninformative cache baseline and the reader's own accuracy from having read the passage. All sit at the baseline. That turns a long tail of separate negative results into one outcome with a stated condition for overturning it. A different design does work: if the reader keeps its own cache and attends over the sender's as extra context, adapting only its queries, it beats its own baseline, and also clears the baseline with eight senders pooled where undirected attention over the same union falls below it. We offer this as a characterization on one probe, not a leaderboard claim. Our central contribution is a redirection: pursuing faithful substitute caches is the wrong target; adding a cache to what the reader already has is the promising one. --- ### GraphBEAT: Graph-based Bucketized Event Aggregation Over Time for Early Risk Stratification of Gastrointestinal Cancers Authors: Leo Y. Li-Han, Chris Varghese, Elizabeth B. Habermann, Mark J. Truty, William R. G. Perry, Cornelius A. Thiels, Hojjat Salehinejad Published: 2026-10-15 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2610.22014v1 Longitudinal electronic health records (EHRs) contain complementary diagnostic and laboratory information about patient health. However, irregularity, sparsity, heterogeneity, and incompleteness in documentation complicate EHR modeling. To address the problem, we introduce GraphBEAT (Graph-based Bucketized Event Aggregation over Time), a temporal graph representation learning framework that groups clinical events into time buckets and transforms observed features into patient-specific, temporally directed graphs to enable early risk stratification for gastrointestinal cancers. Each node in an EHR graph encodes feature identity, observed value, and temporal position of clinical events. A context-aware, time-decay message-passing mechanism models transitions between successive buckets, followed by multi-head attention pooling to generate patient-level representations. GraphBEAT was trained and evaluated on multi-site longitudinal diagnosis codes and laboratory tests to predict pancreatic, liver, and intestinal cancer at 12-, 24-, and 36-month diagnostic lead times. In all experimental setups, GraphBEAT consistently outperformed baseline graph neural networks (GraphSAGE, GAT, and GCN) and sequence modeling approaches (Transformer and long short-term memory (LSTM)). GraphBEAT also required over 250x fewer floating-point operations per patient and over 20x fewer parameters than the Transformer and LSTM baselines. These results support GraphBEAT as an accurate and computationally efficient approach for cancer risk stratification from sparse, heterogeneous longitudinal health records. --- ### Visual Sanad: Auditing Vision-Language Models for False Acceptance of Qur'anic Citations Authors: Niaz Mohaiman Abtahi, Reyana Shahid, Evan Maksud Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.22009v1 Vision-language models (VLMs) are increasingly used to interpret religious, educational, and culturally sensitive visual content, yet their reliability on citation verification remains underexplored. We introduce Visual Sanad, a controlled benchmark for testing whether VLMs falsely accept incorrect Qur'anic verse attributions when the same quote-reference pair is rendered under different visual framings. The benchmark contains 2,400 rendered images built from 100 English Qur'an verses, four factual states per verse, and six visual conditions ranging from plain text to decorative and authority-framed layouts. All quote-reference pairs were source-checked and human-audited before rendering. Across 24 full model-prompt runs, we observe a sharp capability split. Proprietary API-served VLMs range from near-perfect citation verification to moderate false acceptance, whereas recent open-weight VLMs often show weak discrimination and default to VALID on most images. De-biased prompting can reduce false acceptance, but often by shifting the decision threshold toward INVALID rather than by producing strong citation discrimination. Visual Sanad therefore exposes a culturally grounded multimodal factuality failure: fluent VLMs may treat displayed religious citations as self-authenticating unless they have strong citation-verification ability. --- ### Stateful Testing for Deep Learning Compilers Authors: Jiatao Ge, Ruixiang Qian, Yibo Hu, Yanzhou Mu, Yinglong Zou, Jiawei Liu, Zhenyu Chen, Chunrong Fang Published: 2026-10-15 Categories: cs.SE, cs.LG arXiv: https://arxiv.org/abs/2610.22003v1 Modern deep learning compilers compile models lazily and cache artifacts across invocations. Retained compiler state can make a later invocation incorrect even when it is correct in isolation. An analysis of 26 publicly reported issues in PyTorch, JAX, and TensorFlow shows that detecting such bugs requires sequences that first establish compiler state and then expose a failure. We present Rondo, a stateful testing method that represents each test case as a model paired with an invocation sequence. Rondo mutates invocation sequences and uses a Markov chain Monte Carlo sampler guided by compilation behavior to favor rarely observed behaviors. Its three-way differential oracle compares execution without compilation, compiled execution with compiler state reset, and compiled execution with state retained. It confirms stateful bugs by replaying the same invocation with state retained and reset while preserving inputs and program state, then shortens sequences while preserving the original bug. We evaluate Rondo on four deep learning frameworks using 24 common model architectures. Rondo detects 117 bugs, including 35 stateful bugs. We submit 39 bug reports, including 18 for stateful bugs; developers confirm 35 reports, including 16 for stateful bugs, and fix 18 bugs, including 11 stateful bugs. Rondo achieves the highest behavior coverage and compiler Python source-line coverage among the compared methods, with 99.26% alarm precision. It shortens 44 of 117 invocation sequences, reducing the lengths of these 44 sequences by 83.34% on average. --- ### Interfaces and Modules: A Scalable Representation for Weight-Space Learning across Architectures Authors: Michał Kozyra, Michael Bronstein Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21995v1 Weight-space learning across architectures requires representations that respect parameter symmetries and scale to large networks. Existing equivariant tensor encoders require architecture-specific symmetry specifications, while neuron graphs make global attention costly. We introduce a typed graph organized around shared coordinate-permutation actions. Its nodes, called interfaces, represent coordinate spaces whose bound tensor axes must be permuted jointly; operation occurrences, called modules, form typed hyperedges. A compiler derives this graph from supported PyTorch models, with exact recovery of the permutation factors defined by its constraints. The Universal Weight Encoder (UWE) combines local equivariant updates to parameter features with global attention over invariant interface summaries. The small number of interfaces makes global attention practical, and a single encoder processes different supported architectures. UWE is competitive with architecture-specific state-of-the-art methods on established benchmarks. On a new zoo of 1,200 transformers spanning 100 architectures, it outperforms the evaluated baselines at ranking networks within unseen architectures, including architectures larger than those used for training, while using substantially less memory. --- ### Data-Efficient ConvNeXt Pretraining for High-Resolution Sentinel-1 GRD SAR Representation Learning Authors: Yuki Natsume, Sang-Ho Yun, Chai Kiat Yeo Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.21993v1 Labeled Sentinel-1 Ground Range Detected (S1GRD) datasets remain scarce, limiting supervised deep learning for land-surface remote sensing, while self-supervised pretraining for this domain has been dominated by Vision Transformer (ViT)-based masked image modeling requiring substantial data and compute. We propose SparK-SAR, a sparse convolutional masked pretraining framework built on SparK and ConvNeXt, designed to exploit convolutional inductive biases for S1GRD imagery under constrained compute. SparK-SAR makes three core technical contributions: (i) ConvNeXt-RS, a domain-adapted encoder that halves the stem stride (4x to 2x) to produce 14x14 feature maps at stride 16x, providing 4x more spatial tokens at identical parameter count; (ii) a lightweight ConvNeXt-512 reconstruction decoder with PixelShuffle upsampling that reduces pretraining VRAM by 7.5-8.5% per GPU relative to a UNet-128 baseline while matching its accuracy; and (iii) an empirical characterization of FP16 mixed-precision pretraining instability in the standard SparK UNet-768 decoder on S1GRD data, complemented by a finding that finetuning with layer-wise learning rate decay (LLRD) improves downstream performance of pretrained encoders. On four Copernicus-Bench S1GRD tasks, our finetuned 27.8M-parameter ConvNeXt-RS encoder improves over a non-pretrained baseline by +4.7, +5.0, +5.8, and +1.4 pp on EuroSAT-S1, BigEarthNet-S1, DFC2020-S1, and Flood-S1 respectively, and outperforms the Copernicus-FM ViT (139M parameters, frozen) on three of four tasks at roughly 5x fewer parameters. A pretraining data-scaling study reveals task-dependent data efficiency: classification saturates with as little as 10-25% of the SSL4EO-S12 v1.1 corpus, while dense prediction continues to improve significantly up to 25-50%. The study also includes a CKA representation analysis and a geographic de-collision experiment. --- ### VesperMUSE: Multi-Modal Fusion Vision Transformer for Image Semantic Segmentation Under Diverse Light Conditions Authors: Raul Balmez Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.21990v1 Semantic segmentation is critical for the reliable navigation and scene understanding of autonomous systems, yet its performance degrades severely in low-light environments due to noise, artifacts, colour distortion, and the loss of structural details. To address this challenge, this work introduces VesperMUSE, a unified multi-modal framework designed to achieve robust image semantic segmentation across diverse illumination conditions. The architecture addresses the light deficiency problem on multiple levels: proposing a novel physics-informed synthetic low-light adaptation algorithm, approaching Low-Light Image Enhancement (LLIE) through a thermal-guided paradigm, and finally aiming to recover the intrinsic scene details for semantic segmentation through a multi-modal fusion approach. Following enhancement, the framework dynamically integrates the primary RGB features with auxiliary thermal representations and surface normal maps highlighting geometric discontinuities and material boundaries. At the core of this integration is a highly scalable, modality-agnostic self-attention mechanism that extracts both global context and fine local features from an arbitrary number of auxiliary inputs. By combining and projecting these hierarchical multi-modal representations into a unified component, the attention module allows the network to selectively attend to relevant cross-modal cues without requiring any architectural modifications when different sensor configurations are used. Extensive evaluations demonstrate that VesperMUSE successfully recovers missing visual cues, providing a flexible, state-of-the-art (SOTA) solution for machine perception in extreme and unstructured real-world environments. --- ### GRID-Planner: Graph-Structured Relational Reward Shaping for Diffusion Planning Authors: Zhao Yang, Zezhong Qian, Yinan Shi, Longjun Liu, Yawei Jueluo Published: 2026-10-15 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2610.21984v1 Autonomous driving planning requires selecting trajectories that remain safe, rule-compliant, and comfortable under complex multi-agent interactions, yet robust closed-loop decision making in planning-critical scenes remains challenging. Existing diffusion- and RL-based planners often rely on generic rewards and use relational structure mainly for representation, limiting how planning-critical rollouts shape policy improvement. We present GRID-Planner, a graph--reward integrated diffusion planning framework that aligns generative planning with structured relational signals. Specifically, GRID-Planner combines graph-aware importance weighting, a time-indexed relational reward surrogate, and an advantage-weighted diffusion objective with on-policy self-distillation, forming a rollout-return-advantage-update loop without explicit likelihoods. Experiments on nuPlan and interPlan show that, compared with the diffusion-only baseline, GRID-Planner reduces collision rate, lane deviation, and jerk by 35.8\%, 60.8\%, and 27.4\%, respectively, and improves the planning score by 3.4\%. Project website: https://github.com/yangzhaojason/GRID-Planner. --- ### OntoAct: Hypothesis-Based Neuro-Symbolic Agents through Ontological Abstraction Authors: Donghyeok Lee, Bong Gyun Kang, Junyong Ahn, Sungroh Yoon Published: 2026-10-15 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.21983v1 Learning from experience requires recovering reusable knowledge and determining when it applies to a new situation. We introduce OntoAct, a hypothesis-based neuro-symbolic framework built on ontological abstraction. OntoAct turns individual experiences into axioms over entity types, relations, and conditions, making their underlying patterns reusable across situations. A large language model (LLM) uses this world knowledge to propose hypotheses and actions, while a symbolic verifier assesses their supporting conditions against current observations. Verification identifies missing evidence, and new observations and intervention outcomes update hypothesis assessments and guide subsequent actions. We also introduce Tacit-FLE, a benchmark built on the Factorio Learning Environment (FLE) for diagnosing and repairing hidden faults in existing factories using inherited operational records. Its synthetic maintenance corpus distributes evidence across records, and an executed repair may leave the fault unresolved, requiring agents to reassess their diagnoses through interaction. Across Tacit-FLE, Mars, and ALFWorld, OntoAct achieves leading task performance with efficient interaction and inference. --- ### Deep Learning of Topological Equivalence Classes Authors: Linden Disney-Hogg, Max Zuev Published: 2026-10-15 Categories: math.AG, cs.LG arXiv: https://arxiv.org/abs/2610.21981v1 We demonstrate an application of deep learning to the problem of identifying whether two generating vectors correspond to topologically equivalent finite group actions on Riemann surfaces, showing how architectures developed for semantic understanding have use for pure mathematical problems. In doing so we discuss how desired algebraic structure influences model design, and explore which structural properties of topological equivalence classes the model has successfully learnt. --- ### P2TPO: Structure-Aligned Page-to-Token Credit Assignment for Full-Page Document Parsing Authors: Yong Zu, Ninghua Yang, Zezhong Tan, Xinhong Ma, Ziqiang Dong, Yang Yang Published: 2026-10-15 Categories: cs.LG, cs.CL, cs.CV arXiv: https://arxiv.org/abs/2610.21979v1 End-to-end full-page document parsing must preserve the content and structure of heterogeneous elements, yet otherwise accurate outputs can contain local errors and omissions. A single page-level advantage does not explicitly identify which elements differ in quality across sampled outputs or where their errors occur. We introduce Page-to-Token Policy Optimization (P2TPO), a structure-aligned credit-assignment framework that operates on full-page policy trajectories. Reference-element alignment first associates block-level quality contrasts with generated spans. Unified Anchor Pricing (UAP) incorporates matched and missing element outcomes and provides feedback relative to a fixed quality ceiling when sampled outputs receive identical but imperfect scores for a reference element. Within matched blocks, Span Credit Redistribution (SCR) redistributes existing block credit according to aligned correct and erroneous spans, using boundary proxies for within-block omissions while preserving the total block-derived advantage across its tokens. Under a fixed-weight additive page utility, we show that element-indexed contrasts retain local quality differences that page-level aggregation can cancel out. In the main experimental comparison, P2TPO outperforms all evaluated policy optimization baselines in seed-averaged scores on each of the six document parsing benchmarks. Additional experiments demonstrate gains across model scales and initializations. --- ### Learn Only Where You Can't Compute: Structured Corrections to Gaussian Process Posteriors Authors: Tobias Weber, Carla Sagebiel, Philipp Hennig Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21977v1 Physics-informed Gaussian processes (GPs) provide a principled framework for scientific inference by expressing uncertainty over empirical measurements, boundary conditions, and governing equations. However, they struggle to scale to larger datasets and nonlinear information. Conversely, diffusion models excel at learning complex distributions over functions from massive datasets, but struggle in scientific contexts where training data are scarce and physical or observational information must be incorporated. We propose a structured posterior-correction framework that combines a tractable Gaussian reference with data-driven corrections via generative learning. After a mean correction and covariance calibration, a diffusion model learns a targeted correction in a low-dimensional subspace selected to capture discrepancies from the reference, while the complement retains its Gaussian law. Across three scientific inference tasks, these corrections recover hidden modes, improve sample structure, and reduce prediction error relative to the Gaussian reference. Comparisons with full-field generative models reveal task-dependent benefits and tradeoffs between predictive accuracy and computational cost. --- ### Time to Fake It: Continual Machine Learning Intensifies Unfairness in Deepfake Detection Authors: Kajaree Das, Oliver Böse, Subarnaduti Paul, Gabriel Storch, Martin Mundt Published: 2026-10-15 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.21974v1 Deepfake detectors safeguard against misinformation, rendering their failures as much a societal as a technical concern. As generators constantly evolve, detectors trained on static datasets miss newly arising threats, while progressive training interferes catastrophically. Yet, mitigating forgetfulness in terms of accuracy, the prevalent focus of continual learning (CL), is insufficient to reflect real-world impact. We therefore scrutinize the effects of CL methods with respect to evolving fairness. To this end, we analyze four dimensions of fairness across diverse demographics for detectors trained progressively on a chronological sequence of 11 real-world deepfake generation methods. Tracking fairness metrics over time, we expose non-trivial interplays between performance and fairness that depend on both the choice of CL algorithm and demographic intersections. We find accuracy degradation to typically exacerbate unfairness and, conversely, fairness improvements to remain scarce over time. Dissecting the CL method families (rehearsal, prompts, regularization), we observe that inequitable forgetting is not explained by underrepresentation alone, but likely entangled with representation learning. We thus issue a call to action for CL methods to take social outcomes into account when balancing the stability-plasticity tradeoff. --- ### Uncertainty-Aware Federated Learning for Joint Communication and Computation in XR Systems Authors: Robert-Jeron Reifert, Hayssam Dahrouj, Alaa Alameer Ahmad, Haris Gacanin, Aydin Sezgin Published: 2026-10-15 Categories: cs.IT, cs.LG arXiv: https://arxiv.org/abs/2610.21969v1 The convergence of communication and computation, empowered by artificial intelligence (AI) enhanced capabilities, emerges as a key enabler for realizing the promising functionalities of the sixth-generation of communication systems (6G). To this end, this paper investigates an AI-native, robust network, where multiple base stations (BSs) simultaneously serve several extended reality (XR) devices. The paper then presents an innovative distributed deep learning-based approach to simultaneously manage the transmit and computing powers, and computation allocation, amidst uncertainties in both channel and computing states information. Under such an interference network scenario, the paper proposes a robust approach for minimizing the worst-case delay across the served XR devices, subject to computation and power constraints. The paper leverages a deep neural network (DNN) that can be trained either centrally or in a federated manner to map estimated channels and computation requirements to optimized resource allocations per BS. During training, uncertainty samples, reflecting both channel and computing estimation errors, are injected after the DNN output to account for such joint uncertainties. Our results validate the enhanced robust delay performance and scalability of joint uncertainty injection as compared to uncertainty-agnostic methods and state-of-the-art benchmarks, with federated learning-based solutions achieving near-centralized performance while enhancing efficiency, privacy, and adaptability. --- ### Why Shallow Neural Networks Learn Sinusoidal Components First: An Orthogonal Polynomial Dynamics Framework Authors: Yanyang Xiao Published: 2026-10-15 Categories: cs.LG, cs.AI, math.SP, stat.ML arXiv: https://arxiv.org/abs/2610.21964v1 Neural networks often learn low-frequency components before high-frequency ones --- a phenomenon called spectral bias or termed F-principle. Many numerical studies have demonstrated this phenomenon, and theoretical studies have sought to explain it. However, several goals remain insufficiently addressed: (1) explaining spectral bias asymptotically, including its origin and why the preferred components are generally sinusoidal; (2) extending the explanation to general activation functions, such as sigmoid and ReLU; (3) making weak assumptions about the data distribution; (4) avoiding aliasing effects, such as those arising in finite-domain Fourier analysis; and (5) developing a theoretical framework applicable to arbitrary networks. We address these goals theoretically by analyzing the eigenvalue decay and eigenvector shapes of the dynamics matrix H, which arises when we expand the network output as a general Fourier series (GFS) in an orthogonal polynomial basis and express the gradient-descent dynamics of the GFS coefficients as ordinary differential equations. The graded structure of H enables an intuitive explanation of spectral bias, while orthogonal polynomial theory helps us analyze how learning bias depends on the activation function and, potentially, the network architecture. --- ### ChronoSheaf: Joint Spatio-Temporal Spectral Filtering on Cellular Sheaves Authors: Abeer Mostafa, Zahra Ahmadi Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21958v1 Sheaf neural networks generalize graph neural networks by assigning each edge a learned linear restriction map that specifies how the local representations of adjacent nodes relate to one another, thereby mitigating oversmoothing. Existing sheaf architectures, however, define the sheaf Laplacian as an instantaneous spatial operator and therefore cannot explicitly model how signals evolve over time on a graph. We first introduce a spectral sheaf Laplacian, defined jointly over space and time, such that a single operator regulates signal propagation across both the spatial graph and the temporal domain. Directly extending this operator's receptive field in space or time would require increasingly complex sheaf structures. Instead, we propose a spectral filter that expands the effective receptive field along both dimensions without enlarging the underlying sheaf. Building on this operator, we introduce ChronoSheaf, a spatio-temporal sheaf network with two key components. First, Transitional Restriction Maps (TRM) decompose each node's restriction map into a spatial term and two temporal frames that independently control its relation to neighboring nodes at the preceding and following timestep, enabling asymmetric temporal influence. Second, the Dual Spectral Filter (DSF) applies a single bounded-degree polynomial filter with an adaptively estimated spectral radius to the joint spatio-temporal sheaf operator. We validate ChronoSheaf on a diverse set of spatio-temporal forecasting benchmarks, where it outperforms state-of-the-art baselines. The code is available at: https://anonymous.4open.science/r/ChronoSheaf-1D7D/. --- ### SelectPFN: Amortized Bayesian Variable Selection for Sparse Nonlinear Regression with Prior-Data Fitted Networks Authors: Bin Luo Published: 2026-10-15 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2610.21950v1 Tabular foundation models based on prior-data fitted networks (PFNs) predict on a new data set in a single forward pass, but they cannot report which inputs their predictions depend on. We propose SelectPFN, a PFN that performs Bayesian variable selection and prediction jointly for sparse nonlinear regression. In one forward pass it returns posterior inclusion probabilities for every variable, and its predictions are computed from sampled subsets only, so excluded variables provably cannot affect them; we show this is exactly what the Bayesian model prescribes. The network is pretrained once on synthetic tasks from a meta-prior built around a group spike-and-slab sparse-input neural network. Column-permutation equivariance combined with random interleaving of original and knockoff columns yields knockoff statistics with exact finite-sample false discovery rate control, even when the meta-prior is wrong. We develop theory for amortized selection: a Bayes-regret bound for the amortized median probability model, a characterization showing that the information needed to detect a variable grows like $(2/λ)\log(1/ω)$ in the prior weight $ω$ of structures that can use it, where $λ\le1$ is an efficiency, and a negative result showing that average-case training does not transfer to frequentist guarantees at a fixed truth. On out-of-prior nonlinear designs with p up to 500, SelectPFN improves the area under the precision-recall curve by 0.23-0.28 over the lasso, BART, DART, LassoNet and a group-penalized sparse neural network, at a false discovery proportion of about 0.07 and 0.04 seconds per data set. The advantage persists on real covariate matrices, and, as the theory predicts, detection of interaction-only variables rises monotonically with their prior weight. --- ### Following Delegation Traces to Study Human Oversight in Interaction with Multimodal AI Authors: Agnese Chiatti, Matteo Matteucci, Viola Schiaffonati Published: 2026-10-15 Categories: cs.HC, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2610.21948v1 As multimodal AI systems are becoming our everyday thought partners, the shift towards studying how users transfer control to these systems becomes especially urgent. In this paper, we report an overview of findings from an online study with n=56 AI-literate participants and 392 Video Question Answering trials. After inspecting each multimodal AI response, participants rated task-specific trust and selected one of three intended policies for an analogous future task: reject AI assistance, use it with verification, or delegate fully to the AI. Within participants, post-interaction trust strongly predicted the user's delegation choices (OR=14.97, 95% CI (7.14,31.37)). Trust and delegation nevertheless remained distinct. Verification was the dominant policy and often co-occurred with high trust. Delegation also increased with task order after adjustment for momentary and participant-level trust (OR=1.22, 95% CI (1.02,1.45)), indicating an experience-related component that self-reported trust ratings alone could not capture. These findings motivate a shift from intended delegation to a non-intrusive observation of how delegation is enacted in collaborative human-AI tasks. We therefore conclude by proposing enacted delegation traces as a methodological framework for future studies of multimodal and agentic AI use. --- ### Learning Speech Attribute Representations via Multi-Teacher Distillation Authors: Maryem Bouziane, Salima Mdhaffar, Yannick Estève Published: 2026-10-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.21941v1 Recent post-training approaches have shown that self-supervised speech encoders can be adapted to learn a single utterance-level speech attribute through knowledge distillation. In this work, we investigate whether a single speech encoder can simultaneously preserve multiple complementary utterance-level speech representations. We propose a multi-attribute multi-teacher distillation framework that learns semantic, speaker, language, and emotion representations from four frozen teacher models using a shared SSL backbone and attribute-specific branches. Experiments on multilingual semantic retrieval, speaker verification, language identification, and speech emotion recognition show that the proposed model achieves performance close to dedicated single-task students while requiring only a single shared encoder. Furthermore, the learned layer aggregation strategies reveal that different speech attributes naturally emerge at different depths of the SSL encoder, providing new insights into the organization of speech representations. --- ### Agentic-Priors: Using Large Language Models to Address the Cold Start Problem in Preference Elicitation Authors: Bruno Arcanjo, Zhaoxing Li, Sebastian Stein Published: 2026-10-15 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2610.21938v1 An agent that learns a user's preferences from their choices faces a cold start problem: before any choices are observed, it must rely solely on a prior. Informative priors are usually fitted to historical choice data, which new tasks and deployments lack. To address this, we present Agentic-Priors, which lets agentic systems use the implicit knowledge of large language models (LLMs) instead. Specifically, an LLM states a new user's trade-offs in interpretable terms, such as willingness to pay or value of time, which we map to a prior over multinomial logit coefficients. This prior is then refined by a Bayesian learner as choices are observed. Across seven choice datasets, Agentic-Priors improves cold start accuracy over an uninformed prior by 5-40 percentage points. It also outperforms AutoElicit, a state-of-the-art LLM prior-elicitation method, on three datasets and loses on one. It is up to 18 points more accurate than in-context learning, and never significantly worse. Against hierarchical Bayes models fitted to surveys of increasing size, our approach is worth roughly 6-40 surveyed users, in terms of log-loss, so its value is concentrated in a deployment's first users. --- ### Flow-CLIP: Image-Conditioned Residual Scoring for CompositionalImage-Text Matching Authors: Minghong Wei, Enguang Zhu, Pu Cao, Lu Yang, Qing Song Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.21937v1 Vision--language pre-training has enabled substantial progress in image--text retrieval and zero-shot recognition by learning transferable cross-modal representations. However, sentence-level alignment remains insufficient for compositional matching, where captions can describe the same objects and attributes while differing in how these elements are bound. To address this limitation, we propose Flow-CLIP, which refines global image-text alignment through image-conditioned changes in token--patch evidence. Global Compatibility Calibration (GCC) learns a sentence-level similarity metric, while Image-Conditioned Evidence Refinement (ICER) incorporates visual patch context into token queries and uses the resulting change in local matching evidence to adjust the global score. We further introduce Cross-Image Margin Ranking (CIMR), which compares the same positive--negative caption pair across matched and mismatched images, encouraging a larger ranking margin when the image supports the correct description. Experiments on compositional matching and attribute--object binding benchmarks show that Flow-CLIP improves binding accuracy over pretrained CLIP and learned global alignment through local evidence refinement, while cross-image ranking supervision mitigates reliance on language priors. Code will be released at link. --- ### The Lagrangian Structure of Cost-Penalized Active Feature Acquisition Authors: Joseph Bingham Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21919v1 Active Feature Acquisition (AFA) sequentially selects features to maximize predictive accuracy under an acquisition budget. A natural heuristic greedily maximizes Expected Information Gain minus a cost penalty, EIG(j|x_S)-lam*c_j (EIG-Cost), but its theoretical guarantees are unclear. We provide an a-posteriori certificate decomposing the gap to the optimal hard-budget policy into a budget-slack term and a greedy-suboptimality term Del_lam. The certificate is computable when the penalized optimum is tractable; on small heterogeneous-cost instances, we verify it exactly, finding Del_lam near zero across 4,000 solved cases. On the large clinical task, Del_lam is not computable, so the decomposition is diagnostic rather than a numerical guarantee. Our analysis identifies EIG-Cost as one-step greedy optimization of the Lagrangian relaxation of expected-cost AFA, with lam as its dual multiplier. For equal costs, we prove Del_lam <= exp(-gam*I(Y;X_(d))), where gam is the submodularity ratio, yielding a structural guarantee in the near-submodular regime. Supporting results establish an adaptive chain-rule identity showing that cumulative stepwise EIG optimizes transcript information, and prove that A -> I(Y;X_A) need not be submodular, even for jointly Gaussian variables. This rules out a direct Nemhauser-Wolsey-Fisher bound; naive Bayes restores submodularity. We also bound the randomization gap between hard-budget and expected-cost optima, analyze sensitivity to dual-multiplier error under local smoothness, and derive a conditional CVAE generator-shift bound given a bound on forward predictive KL. We treat this KL as a diagnostic quantity, not as a transform of training loss. Experiments cover synthetic Gaussian settings, AFA benchmarks, and clinical differential diagnosis on MIMIC-IV. On MIMIC-IV, EIG-Cost achieves macro-F1 0.180 +/- 0.012 at a $30 budget, using 6.6% of full-panel cost. --- ### BridgeAfford: Bridging Query-Target Gap for Generalizable 3D Affordance Grounding Authors: Yijie Su, Yakun Zhu, Yi Bin, Pengpeng Zeng, Liang Chen, Guoqing Wang, Jingkuan Song, Heng Tao Shen Published: 2026-10-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.21918v1 Language-guided 3D affordance grounding aims to localize functional regions on 3D objects according to interaction intents. However, existing methods exhibit poor generalization to unseen objects. One challenge is the query-target gap: a semantic discrepancy between query features encoding interaction intents and target-region features, leading to inaccurate grounding. End-to-end training favors query representations tailored to seen objects, limiting their compatibility with shifted target-region features on unseen objects. To address this issue, we propose BridgeAfford, a framework that reduces the query-target semantic gap from two complementary perspectives. First, Interaction Region Alignment (IRA) promotes cross-object alignment among geometrically similar interaction regions during training. This enables structures on unseen objects that resemble previously observed ones to produce representations similar to those learned during training, facilitating target-region recognition. Second, Visual Evidence Aggregation (VEA) leverages a 2D affordance model to refine the attention distribution and guide the query toward the correct region, yielding more target-aware queries. Extensive experiments on the LASO and PIAD datasets demonstrate that BridgeAfford outperforms existing methods, particularly on unseen objects, highlighting its robust generalization ability. --- ### CodeHealth Predicts the Token Cost of Agentic Code Review: A Pre-Registered Study of 465 Agent-Authored Pull Requests Authors: Lo Gullstrand Heander, Markus Borg, Emma Söderberg, Christofer Rydenfält Published: 2026-10-15 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2610.21914v1 Context: Organizations increasingly delegate code review to multi-agent large language model (LLM) pipelines billed per token. Low code quality raises human maintenance effort, but its effect on agent review cost has not been measured. Objective: We test whether baseline CodeHealth predicts the token cost of agentic review and fix beyond pull request (PR) footprint. Method: In a pre-registered observational study, we sampled 465 agent-authored PRs from 246 repositories in AIDev, jointly stratified by CodeHealth and footprint under repository caps. A frozen five-agent review-and-fix pipeline processed every PR with three different LLMs. Per LLM, we fitted a mixed model of log tokens on CodeHealth and log PR footprint, with a repository random intercept. Results: For all three LLMs the CodeHealth coefficient was negative, with 95% CI excluding zero; each CodeHealth point was associated with 3.5% to 6.1% lower token cost. CodeHealth was unrelated to PR footprint, so PR size does not explain the association. Two of three registered smells predicted fix cost; batch fixing left median CodeHealth unchanged. Exploratory analyses located the association in the review stage, where agents iterated more on unhealthy code without writing more comments. Adjusting for the size of the scored files absorbed the association, which we interpret as mediation by file size. Conclusions: A quality metric calibrated against human maintainability also predicts the inference cost of multi-agent review and fix. The effect is modest per point and not carried by PR size, so the refactoring case extends to agent-reviewed code when it changes structure and file size, not only the score. Data Availability: https://doi.org/10.5281/zenodo.22127920 --- ### Learning Rewards in Strategic Games Authors: Antoine Bergerault, Cyrille Kone, Giorgia Ramponi Published: 2026-10-15 Categories: cs.GT, cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2610.21912v1 Understanding the objectives that drive agents' behavior is crucial in strategic environments, where multiple agents interact and adapt to one another. This challenge arises in domains such as cybersecurity, autonomous driving, and online markets, where agents' underlying incentives are typically unknown and must be inferred from their observed behavior. Inverse reinforcement learning (IRL) provides a natural framework for this problem, but extending IRL to strategic multi-agent settings introduces a fundamental challenge: a reward that explains observed behavior does not necessarily induce optimal strategies when used for decision-making. In this work, we study this problem through the lens of \emph{strategic identifiability}, and show that learning from a fixed policy pair can be insufficient to strategically identify rewards, independent of the number of expert samples $K$. Motivated by this, we introduce an active IRL algorithm for two-player repeated Stackelberg games, in which the leader adaptively selects policies to elicit informative responses about the follower's reward. Combining maximum-likelihood estimation with experimental design, we derive finite-sample guarantees on the Stackelberg sub-optimality induced by the learned reward, matching up to polylogarithmic factors a new minimax learning rate of $K^{-1/2}$. Finally, we evaluate our approach in a realistic anti-poaching setting, demonstrating the benefits of active exploration for reward recovery and downstream strategic performance. --- ### CoWorld Reasoner: Agentic Multimodal Reasoning with the Evolving Video World Model-Driven Harness Authors: Dongjie Cheng, Ruifeng Yuan, Ran Wei, Ning Song, Yongqi Li, Wenjie Li Published: 2026-10-15 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.21910v1 As Vision-Language Models (VLMs) advance, they are increasingly expected to perform spatiotemporal reasoning over motion, actions, and view changes beyond what the input images show. Video World Models (VWMs) provide a natural unified engine for visual imagination by generating task-relevant visual content as additional evidence. However, existing approaches typically invoke a VWM through fixed pipelines centered on a predefined form of visual imagination, such as novel-view synthesis. To support flexible visual imagination for diverse reasoning tasks, we introduce CoWorld Reasoner, an agentic framework that uses a VWM as a unified visual reasoning engine. Its Video World Model-Driven Harness exposes multiple visual operations through tools and uses skills to guide the VLM in deciding whether to imagine, selecting an operation, constructing a generation request, and interpreting the resulting visual evidence. We further introduce VLM-VWM Harness Co-Evolution, which jointly revises tools, skills, and execution code based on execution feedback while keeping both model weights fixed. Across seven benchmarks spanning spatial intelligence, temporal and causal reasoning, and perceptual reasoning, CoWorld Reasoner outperforms direct reasoning with the same VLM. The strongest variant improves on all seven benchmarks, including relative accuracy gains of 11.8% on MMSI and 13.0% on ERQA, and outperforms all five evolving baselines using the same backbone across five benchmarks. --- ### The Tilted Sampling Problem: Optimism, Posterior Sampling, and Logarithmic Regret Authors: Mohammad Pedramfar, Siamak Ravanbakhsh Published: 2026-10-15 Categories: cs.LG, math.OC, stat.ML arXiv: https://arxiv.org/abs/2610.21908v1 Many modern learning systems are used not to select a single best action, but to sample diverse high-quality outputs. Examples include entropy-regularized reinforcement learning, RLHF, and reward-guided generative modeling. Motivated by these settings, we study tilted sampling: an online learning problem in which the goal is to sample from the Gibbs distribution induced by an unknown reward function. We define sampling regret as the temperature-normalized cumulative KL divergence between the learner's distribution and the Gibbs target, corresponding to the suboptimality gap in entropy-regularized reward maximization. Our framework generalizes recent sampling-based and KL-regularized formulations in bandits and reinforcement learning. Our main contribution is a general regret framework that captures the different uncertainty structure of tilted sampling and classical regret minimization. While standard regret scales as $\widetilde{O}(\sqrt{T})$ due to first-order uncertainty, sampling regret is governed by squared confidence widths, yielding logarithmic-in-$T$ regret in standard settings at fixed temperature. We develop sampling analogues of UCB and Thompson sampling (UCS and TS$^2$) and provide unified regret decompositions that explain and extend recent logarithmic guarantees, including the first Bayesian sampling-regret decomposition based on new bounds for posterior-random Gibbs distributions. Our results show that learning to sample from a soft-optimal distribution is statistically distinct from learning to identify a hard optimum. --- ### Seeing is Believing: Benchmarking and Mitigating the Interference of Language Priors on Visual Perception in MLLMs Authors: Li Ma, Yan Zhuang, Jiawen Deng, Fuji Ren Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.21902v1 Language priors in MLLMs can override visual evidence, impairing visual perception. Existing benchmarks assume shared priors across models and rely solely on accuracy. However, cross-model prior agreement falls below 80%, and over 15% of items can be answered correctly without any image, causing accuracy to overestimate prior resistance. To address these limitations, we introduce ViSiP-Bench, a benchmark grounded in rare-but-real scenarios across six prior categories, with a model-aware dynamic prior induction protocol and three metrics that more precisely and continuously measure prior resistance than accuracy alone. Evaluating 24 models reveals that prior adherence is pervasive, does not improve with scale, and larger models tend to either strongly override their priors or firmly adhere to them. We therefore propose MAPS, which identifies each model's strongest priors and fine-tunes on factually grounded knowledge to soften excessive prior reliance without disrupting general capabilities. --- ### In-Context Learning Before Specialisation: Theory of Learning in a Solvable Topic Model Authors: Francesco Cagnetta, Casper L. Christensen Published: 2026-10-15 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2610.21901v1 Sequence models can generalise in fundamentally distinct ways: extrapolating structures seen in training, or using their prompt to adapt to novel tasks. The prevailing strategy determines how the model behaves beyond its training distribution, controlling potential capabilities and risks. Recent work has characterised many different strategies theoretically, and empirically described the transition between them as the quality and quantity of pretraining data vary. However, there is no learning theory for this competition, hence no fundamental understanding of \textit{why} one solution might be favoured over another. Here we provide such a theory in a solvable topic model---documents drawn from a finite mixture of latent topics, mirroring the heterogeneity of modern pretraining data. The dataset comes with two distinct solutions: a specialised predictor, that stores the finite topic mixture, and an in-context one, optimal for topics never seen during training. Our theory predicts their emergence at distinct sample complexities: the in-context predictor is available early, while the specialised one emerges only after the training tokens outnumber the parameters of the data-generating process. Trained transformers reproduce both scales and the separating crossover in remarkable agreement with the theory, with no fitted parameters. --- ### TEM: Temporal State Tracking Memory for Large Language Model Agents Authors: Pengfeng Song, Yunkai Zhuang, Liye Shi Published: 2026-10-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.21900v1 An agent's memory must remain useful as information changes across conversations. A new fact may replace an earlier value without making that value irrelevant to a later question. We present Temporal State Tracking Memory (TEM), a training-free framework for tracking these changes. TEM stores facts as versioned records linked to their sources. It retrieves relevant versions, places them in source order and labels historical records before passing the evidence to the answer model. Event memory and original dialogue turns provide details missing from structured facts. We evaluated TEM on 420 questions drawn from 600 sessions involving 60 personas. Across four open-weight models, key-entity hit rates reached 75.71-97.62% overall and 96.67-100.00% on the historical-residence subset. Adding historical labels to otherwise identical, source-ordered facts increased overall hit rates by 17.14-55.95 percentage points. The evidence was compact: 94-118 tokens per answer, or 48.75-56.73% less than session-aware retrieval-augmented generation (RAG). A further evaluation on 1,540 LoCoMo questions tested hybrid memory under a fixed evidence budget. TEM exceeded session-aware RAG with both backbones; conservative score differences were 7.92 and 10.58 percentage points. The results support preserving fact versions and exposing their temporal status to the generator when answering questions about repeatedly updated information. --- ### Self-Calibrated Subspace Robust PCA for Parameter-Free Removal of Mixed Noise and Line Stripes in Hyperspectral Images Authors: Wenzhe Wang Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21896v1 Hyperspectral images (HSIs) are often corrupted by Gaussian noise, impulses and stripes whose intensities vary across bands. The representative coefficient correlated total variation (RCCTV) model restores HSIs efficiently, but it lacks Gaussian and stripe terms, and its noise-aware extensions require the noise level and a stripe budget to be specified in advance. This article proposes a self-calibrated RCCTV (SC-RCCTV) model, in which the noise parameters are estimated adaptively during the iterations. A band-wise dense term and line-constant column and row stripe terms are introduced into RCCTV. The noise level of each band is initialized by a stripe-invariant median absolute deviation estimator and refined online from the residual, which turns the separation of sparse and dense noise into a band-adaptive Huber loss, and the stripe thresholds follow from a significance test whose false-alarm rate can be predicted analytically. No parameter is tuned for a specific image, and every subproblem of the alternating direction method of multipliers (ADMM) solver is solved in closed form. Under line stripes, SC-RCCTV outperforms the parameter-free FastHyMix by 8.17 dB on native-size cubes, and, without any noise information, it exceeds S3TTV, which is given the true noise parameters, by 1.52 dB while running more than 1,900 times faster. On average, it stays within 0.01 dB of the same model given the true noise levels, and its online calibration improves on a two-stage estimate-then-denoise scheme. Experiments on real Indian Pines and Urban data further delineate its scope for stripes that do not span entire lines. --- ### SurgStreamer: proactive, request-driven monitoring of surgical video streams with structured state and a reusable reasoning layer Authors: Bowen Xiang, Jon S. Heiselman, Michael I. Miga Published: 2026-10-15 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.21895v1 Surgical teams could benefit from assistants that watch the video for requested events, such as "tell me when the clip applier appears during Calot triangle dissection", and answer questions about what has already happened. We formalise this task as request-driven event detection with exact semantics and causal alert decisions, and present SurgStreamer. It separates procedure-specific perception, which maps the 1-FPS stream to a structured state over tools, actions and phases, from procedure-agnostic reasoning, which evaluates a formal monitoring condition over this state and an event memory, either with exact rules or with a small controller that addresses entities through pointers rather than names. We evaluate this monitoring core with gold conditions. On a pre-registered, sealed test set of 12 laparoscopic cholecystectomies, SurgStreamer reached a macro event-F1 of 0.598 over full procedures. On identical 75-s windows with only the window available to every method, it scored 0.677. The tested configurations of the streaming vision-language model OneStreamer-4B, including ones fine-tuned on SurgStreamer's state or on video with an adapted vision encoder, scored at most 0.540 (Holm-adjusted p <= 0.012). At three external centres, re-training only perception raised performance on held-out procedures from 0.239 to 0.470. On a new procedure, pituitary surgery, controller weights trained on cholecystectomy matched an in-domain controller on oracle state (0.947 vs 0.938). On predicted state they missed the pre-registered non-inferiority margin (-0.024 (-0.040, -0.008)) unless the alert thresholds were re-calibrated. The results support separating perception from reasoning and point to perception under distribution shift as the main limitation. --- ### Graph Dirichlet Energy for Fitting-Free Bivariate Causal Inference Authors: Povilas Daniušis Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21892v1 We propose the Graph Dirichlet Energy Causal Inference (GDECI) criterion for distinguishing cause from effect in bivariate settings. For a candidate direction $X\to Y$, we construct a kernel similarity graph from $X$ and treat $Y$ as a graph signal. The score is the Dirichlet energy of this signal, induced by the unnormalized graph Laplacian. The reverse score swaps the variables, and the lower-energy direction is inferred as causal. The method requires neither regression fitting nor independence testing. After the empirical probability integral transform, its decisions are invariant under strictly monotone transformations of either variable, and the graph operations reduce to convolutions, allowing exact evaluation in $O(n\log n)$ time and $O(n)$ memory. At the population level, GDECI compares degree-weighted Bayes squared prediction errors plus a non-negative smoothness term of the regression function. As the bandwidth tends to zero, the weights become uniform and the smoothness term vanishes, reducing the criterion to a RECI-type comparison of expected conditional variances in copula space. Empirically, we compare GDECI with QCCD, GRCI, CAM, IGCI (two variants), RESIT, LOCI, QPE-k, and RECI (two variants), in accuracy on 1899 cause--effect pairs and in inference speed. GDECI ranked fifth of eleven methods in macro-averaged accuracy on 12 synthetic suites ($0.776$). On the real-world Tuebingen pairs, it ranked third in unweighted accuracy ($0.684$) and fifth in weighted accuracy ($0.678$). It was also the fastest of the compared methods up to $n\approx10^{4}$. The accompanying code repository: https://github.com/povidanius/gdeci. --- ### Anvil: Scaling Embodied Spatial Intelligence through Knowledge Recirculation Authors: Weilin Ruan, Bohan Zhou, Jinghao Li, Yangkai Wei, Wenqian Li, Peiran Dong, Yingjie Xu, Leyi Wu, Haojian Huang, Binghui Xie, Han Yang, Kaiwen Zhou et al. Published: 2026-10-15 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.21882v1 Embodied spatial intelligence requires more than visual recognition: models must localize entities, reason about their relations, and express this knowledge in action-relevant forms. Yet joint training can conflate the knowledge needed at different stages because supervision is fragmented across incompatible geometric and semantic outputs. We introduce Anvil, a unified framework for data generation, training, and benchmarking for embodied spatial intelligence. (1) Anvil-Gen is a heterogeneous data flywheel that combines form-specific generation, cross-stage consistency verification, and failure-driven recollection across 11 Ground--Relation--Act forms. (2) Anvil-Ben provides scene-aligned supervision for learning diverse embodied spatial capabilities and scene-linked assessments that distinguish stage competence from error propagation. (3) Anvil-MOPD is a 9B student that consolidates heterogeneous spatial knowledge through distillation from specialist teachers. Semantic-first-fault routing targets teacher supervision to the student's earliest decisive failure, while stage normalization controls the loss scale across native outputs of different lengths. We construct 5.8M source-grounded records spanning 160K images, reserving 580K spatial records for held-out evaluation and using the remaining 5.22M records for training. On average across seven public benchmarks, Anvil-MOPD outperforms Qwen3.5-9B and RynnBrain-8B by 9.29 and 11.03 percentage points, respectively, while reaching 80.67% on VABench-Point. Controlled studies support targeted coverage, selective transfer, and feedback-guided repair as complementary framework components. --- ### Learning Lean Feedback for Efficient Proof-Candidate Evaluation on GPUs Authors: Lazar Milikic, Etienne Bamas, Emmanuel Abbe, Viktor Kunčak Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21881v1 Training language models for formal theorem proving requires large numbers of verifier calls. Slow checks and timeouts often leave GPUs waiting and withhold rewards from potentially correct proofs. We train neural evaluators to predict Lean 4 verdicts and compiler feedback on GPUs using 4.84M labelled programs. We compare two validity classifiers with a feedback model that annotates proof attempts with predicted goal states and error messages. A two-stage evaluator that screens candidates with a classifier and scores its positive predictions with the feedback model reaches 87.8% accuracy in predicting whether Lean accepts a candidate proof, evaluated on a 68,932-program benchmark collected from public provers. The combined evaluator processes 3.5 candidates per second on one H100, 4.2x the throughput of our 64-worker Lean baseline with environment reuse. Finally, we combine our evaluator with Lean verdicts to obtain a hybrid reward system that shortens GRPO training steps by approximately 31%. After one GRPO epoch with hybrid rewards, the prover achieves nearly the same miniF2F pass@1 as with Lean-only rewards (62.2% versus 62.4%) and a pass@32 of 72.4% versus 71.3%. --- ### Provably Efficient Risk-Sensitive Offline Reinforcement Learning via Optimized Certainty Equivalents Authors: Hanzhen Wang, Boxiang Lyu, Zhongxiang Dai, Siyang Gao, Dake Zhang, Shuang Qiu Published: 2026-10-15 Categories: cs.LG, cs.AI, math.OC, stat.ML arXiv: https://arxiv.org/abs/2610.21870v1 Risk-sensitive reinforcement learning (RL) addresses decision-making under uncertainty and adverse outcomes. Learning risk-sensitive policies from fixed datasets is important when additional interaction with the environment is costly. Existing work on risk-sensitive RL focuses on online settings or specific risk measures, leaving a general theoretical understanding of its offline setting underdeveloped. In this work, we investigate efficient offline risk-sensitive RL using the optimized certainty equivalent (OCE), a general class of risk measures. We consider two formulations: dynamic-OCE RL, which applies OCE recursively at each step, and static-OCE RL, which applies a single OCE to the cumulative rewards. Under the tabular setting, we develop pessimistic value-iteration algorithms for both formulations and establish tight sublinear suboptimality bounds, explicitly capturing how risk sensitivity, encoded by utility functions, affects learning difficulty. Moreover, we establish minimax suboptimality lower bounds to characterize the problem's fundamental limits. To handle large state spaces, we further develop pessimistic least-squares value-iteration algorithms that incorporate the general OCE risk criterion into linear function approximation. We establish utility-dependent sublinear guarantees with explicit dependence on the feature dimension $d$. To our knowledge, we provide the first offline algorithms along with the first upper and lower suboptimality bounds for both dynamic- and static-OCE RL. --- ### Lethe: Link Inference Attacks For Evaluation of Edge Unlearning Methods Authors: Andrea D'Angelo, Francesco Gullo, Davide Mottin, Sayan Ranu, Giovanni Stilo Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21869v1 Graph Unlearning (GU) aims to remove the influence of specific training data from a Graph Neural Network without retraining anew. While most methods address removal of nodes and features, Edge Unlearning (EU), aiming at forgetting edges, has recently attracted dedicated attention as a standalone problem due to the hardness of completely removing the impact of an edge in the model. Despite this, no dedicated benchmark for EU exists besides isolated evaluations on feature-rich datasets such as Cora and Citeseer where GNNs exploit node features rather than graph structure. As such, we show two systematic limitations in existing evaluations: dedicated EU methods can be slower than retraining anew, and accuracy metrics are uninformative due to the negligible impact of edge removal. To address this gap, we introduce Lethe, the first comprehensive benchmark for Edge Unlearning, featuring 15 methods and 8 datasets. Our benchmark emphasizes large graphs, introduces evaluation tasks of varying difficulty, and proposes empirical measures based on link inference attacks rather than proxy accuracy metrics. We provide a comprehensive empirical analysis, demonstrate the shortcomings of current evaluations, and release Lethe as a reproducible, extensible benchmark. --- ### From Sketch to Scene: Relightable Garment Reconstruction from Marked Sketches Authors: Ahmet Neçirvan Doğan, Eren Acar Başaran, Metin Sezgin, Zahra Montazeri Published: 2026-10-15 Categories: cs.GR, cs.CV, cs.HC arXiv: https://arxiv.org/abs/2610.21865v1 Sketching is one of the most natural ways for people to express visual ideas. Unlike the text prompts that most generative systems rely on today, a sketch captures spatial structure and intent directly, including where a designer wants highlights and shadows to appear. Yet current sketch-to-3D pipelines have no input for that intent: they bake the drawn shading into the texture as dark paint, and the light cannot be moved. We propose a system that transforms a rough garment sketch, marked with sparse annotations (a star for a highlight and a moon for a shadow), and a short text prompt describing the texture, into a textured 3D mesh with an explicit directional light, as a scene that can be relit and edited after generation. The two symbols are the most frequent proposals in an elicitation study with art students. From the marked sketch the system predicts the scene's directional light and, separately, a displacement field that bends the surface so that the moon-marked regions fall into shadow under that light, rather than being painted dark in the texture. --- ### Dynamic ordering in sequence generation in Transformers: experiments on low-resource Tagging scenarios Authors: Giacomo Bais, Tejaswini Deoskar Published: 2026-10-15 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.21862v1 This paper introduces TagInsert, a Transformer-based model that challenges the common left-to-right (L2R) generation paradigm in Transformers, for sequence labelling tasks where dependencies may not naturally align with a L2R order. TagInsert lifts the L2R constraint by removing the decoder-side attention mask and enabling the model to decide a preferred order of tag generation at inference time. We experiment on the common Part-of-Speech tagging and the more complex Combinatory Categorial Grammar (CCG) supertagging tasks on multiple languages and low-resource data settings. We show that, although encoder-only models like BERT remain strong performers for tagging tasks, TagInsert frequently outperforms a L2R encoder-decoder model, achieves results comparable to encoder-only models under data-scarce conditions, and shows better results on the long-tail of rare categories. Our analyses show that TagInsert alleviates error propagation typical of autoregressive models by learning to defer difficult predictions to later stages in the decoding process. --- ### Seeing Both the Forest and the Trees: Reusing Holistic 3D Priors for Part-Decomposed Generation Authors: Yiqun Zhao, Binbin Huang, Haobin Duan, Zibo Zhao, Shenghua Gao Published: 2026-10-15 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2610.21850v1 Generating 3D assets with explicit parts requires two properties at once: geometric detail inside each part and structural coherence across the whole object. Recent image-conditioned 3D generation models adopt a structured 3D representation that anchors tokens to explicit spatial positions, capturing both properties, but their output remains a single fused mesh. We ask how a part-decomposed generation model can inherit both properties of these holistic priors, and answer with STRUCTPARTS, a framework whose core technical contribution is a dual-frame coordinate interleaving mechanism: each shape latent is interpreted under both a part-local and an object-global reference frame, and processed through a global-local inter-leaved transformer that simultaneously reuses the prior's geometric detail at the part scale and its structural coherence at the whole-object scale. As its structural input, STRUCTPARTS uses a segmented mesh scaffold, a coarse mesh paired with a face-level segmentation mask kept entirely within the prior's native 3D representation. With only lightweight finetuning, both intra-part detail and inter-part coherence emerge from the same pretrained weights. On PartObjaverse-Tiny, STRUCTPARTS matches or surpasses prior part-decomposed models at a fraction of their training cost, while naturally supporting part-mesh refinement and part-level editing. --- ### Temporal Periodic Image Registration with Implicit Neural Representations Authors: Mathias Lowes, Kristine Aavild Sørensen, Maxime Sermesant, Rasmus R. Paulsen Published: 2026-10-15 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2610.21845v1 Implicit Neural Representations (INRs) have recently gained popularity for their ability to model functions using simple and lightweight networks. These methods have also proven effective in deformable image registration. In this work, we extend INRs for deformable image registration to the domain of temporal image registration, focusing on periodic temporal image sequences. Our approach, Temporal-IDIR, optimizes a single INR to model deformations across all frames in a temporal image sequence simultaneously, allowing for self-regularization through its own deformation predictions. To achieve this, we introduce a temporal consistency loss that penalizes discrepancies between direct source-to-target transformations and those traversing intermediate frames. We evaluate our framework on the DIR-LAB dataset, using the target registration error (TRE) between annotated and moved landmarks as the metric. Here, we achieve a TRE of 1.03 mm, outperforming other INR-based registration methods. Additionally, our framework supports smooth interpolation between time frames by estimating deformations between the given input frames. Code is publicly available at https://github.com/MMLowes/Temporal_INR. --- ### Physics-Aware Distillation of a Grid Foundation Model for Real-Time Edge Inference Authors: Vlad Alexandru Sănduleac, Shahariya Rasheed Rayamarakkarveettil Puthanpurayil, Noman Khan, Remus Teodorescu Published: 2026-10-15 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.21833v1 Grid foundation models (GridFMs) are graph transformers pretrained on a masked power-flow objective, and evaluate power flow orders of magnitude faster than an iterative solver. Their size, however, targets data-centre inference, whereas low-inertia operation increasingly asks for inference at the edge, next to the measurement, at batch size one and within a single synchrophasor (PMU) reporting frame. We distil a 20.1M-parameter, 12-layer GridFM teacher into a 0.62M-parameter, 6-layer student, a $32.5\times$ parameter reduction at half the message-passing depth, using an objective that combines the masked power-flow loss with output, physics-residual, latent-state and attention transfer. The physics-residual term, which matches the teacher's power-residual profile, has not previously been applied to typed power-grid graphs. We evaluate on a renewable variant of the IEEE 118-bus benchmark grid in which 26 of the conventional (synchronous) generators are replaced by inverter-based renewables with battery flexibility and standardised reactive-power limits (IEEE 1547), giving $97\,529$ physically valid (AC-OPF-feasible) operating scenarios, and we ablate all eight combinations of the transfer terms, reporting accuracy against ground truth and fidelity to the teacher separately. Distillation lowers student error on all four predicted quantities, by up to $48\%$ on active-power dispatch, and every distilled student is more accurate than its teacher on reactive-power dispatch. The gain is largely carried by the output matching term. The physics-residual term, in the layer-aggregated form used here, is consistently harmful, which we attribute to the spatial averaging it applies to a bus-level residual field. The resulting student is $32.5\times$ smaller in parameter count and has half the message-passing depth, making it a candidate for subsequent edge-device benchmarking. --- ### Local Epochs as Optimal Stopping: Theory and Evidence in Federated Learning Authors: Ke Xiao, Qiyuan Wang, Christos Anagnostopoulos Published: 2026-10-15 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2610.21832v1 How long should each client train before communicating? Federated learning usually answers with a fixed, hand-tuned number of local epochs, yet an extra epoch helps only while the loss it removes outweighs the client drift it adds. We cast local training as finite-horizon optimal stopping: a client trains while the next epoch's expected loss reduction exceeds its priced squared drift. We prove that this one-stage look-ahead threshold is optimal whenever stopping decisions are nested, derive closed forms for geometric and power-law returns, and show that an estimated rule loses utility only through wrong-signed decisions. Epoch-level evidence supports the model's structure (geometric diminishing returns and additive drift) and reveals a simple regularity: the stage of training, not the client, sets the budget. The resulting rule, FedStop, is never more than one point below tuned FedAvg in mean accuracy. Where its budgets vary, it raises worst-decile accuracy by 4-8 points over tuned FedAvg and an adaptive-interval baseline, and every one of these gains remains significant after a Holm correction. Against the client-side stopping rules ALT and FLASH, it has the highest CIFAR-10 worst decile on every seed and is level elsewhere. It stays stable where per-client early stopping collapses, and each stopping decision takes constant time. ---