# Research Papers — 2026-10 · bundle page 41 of 156 > Full-text markdown of papers #2001–2050 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/41 [← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/40.md) [Next bundle →](https://aipod.dev/papers/archive/2026-10/page/42.md) ### BrowseSafe++: Benchmarking Prompt Injection Detection Across Web Contexts, Languages, and Browser Views Authors: Yuetian Chen, Kaiyuan Zhang, Yuntao Du, Thomas Li, Mark Tenenholtz, Kyle Polley, Jerry Ma, Denis Yarats, Ninghui Li Published: 2026-10-12 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2610.17457v1 Browser agents need to make decisions and execute actions on behalf of users while consuming untrusted webpages, creating opportunities for malicious instructions embedded in webpages to redirect legitimate tasks. We introduce BrowseSafe++, which extends BrowseSafe from 14,719 pages to 38,675 pairs of pages. Most of its prompt injections were observed in the wild on production websites. We keep each injection unchanged and rewrite the sensitive parts of the page around it to protect privacy. Each attacked page has a clean copy of the same page as its control. Both are provided as HTML, Markdown, an accessibility tree, and a screenshot. Every pair includes a user task and its expected answer. The benchmark includes twelve types of websites, from news articles to shared documents and forums, with 24 attack types appearing in 18 different locations on a page. We use these paired pages to train BrowseSafe-Guard, two 0.8B-parameter detectors for text and screenshots. Both achieve higher F1 scores than the detector based on frontier models, while keeping false-positive rates on benign pages and inference cost low. BrowseSafe-Adapt revises benchmark injections from browser-agent feedback, and attacks that succeed on only a few percent of tasks when replayed directly reach up to 95.41% success once adapted. --- ### Compositional RL: Evaluating and Training LLMs for Collaborative Reasoning Authors: Shangbin Feng, Yulia Tsvetkov, Jan Kautz, David Acuna Published: 2026-10-12 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.17456v1 Language models are often evaluated alone but increasingly deployed together, routed between, made to debate, ensembled, or collaboratively trained. For a model inside a system, what matters most is not how well it performs on its own, but how much the system gains from its presence. We call this a model's compositional strength and formalize it using the Shapley value of a cooperative game over coalitions of a model pool, and use a model's leave-one-out contribution to the full coalition as a tractable proxy, requiring n+1 rather than 2^n system evaluations. Evaluating individual vs. compositional strengths across five model pools, twelve collaboration methods, and twelve datasets, we find them largely uncorrelated: in 12 of 15 (pool, method-family) settings, individual strength does not translate into compositional advantages, and the coupling is weakest among general-purpose models that practitioners are most likely to deploy. Standalone benchmarks are therefore often poor proxies for models' value in collaborative systems. We then propose compositional reinforcement learning (C-RL), which trains a policy inside a live multi-model system and rewards the outcome produced by the system as a whole, rather than the policy's correctness in isolation. The leave-one-out baseline in Shapley values cancels under a group policy gradient, so this needs no additional rollout cost beyond RL on the collaborative system's outcome and never requires running a coalition without the policy. Across three collaboration protocols and eight tasks spanning reasoning, coding, and scientific discovery, C-RL produces the strongest collaborator in 21 of 24 settings, lifting collaborative performance by +4.0 to +5.4 pts on average (~3x the gain of baselines), up to +13.1 pts. --- ### Effect of Spherical Parameter Space on Number of Non-Forgetting Tasks and Capacity of Continual Learning Authors: Nazanin Mirhosseini Published: 2026-10-12 Categories: cs.LG, cs.IT, stat.ML arXiv: https://arxiv.org/abs/2610.17453v1 We propose a continuous learning formulation for $T$ non-forgetting tasks with the spherical parameter space $\mathbb{S}^{d-1}(\sqrt{d})$ in asymptotic regime $d\rightarrow \infty$. Each task $i$ has an optimum $\bmΘ_i$, minimizing its corresponding loss function $R_i(\bm{w})$ on sphere. We reduce each task $i$ to an acceptable loss region modeled by a geodesic ball $B_{\bmΘ_i}(r_i(d))$ on $\mathbb{S}^{d-1}(\sqrt{d})$ centered at $\bmΘ_i$ with radius $r_i(d)$. To model task dependencies (similarities), task optima $\bmΘ_i$ are distributed by a Neyman-Scott clustered point process $Φ$. Cluster centers $\{\bm{Y}_j\}_{j=1}^{N_p}$ form a homogeneous Poisson point process on sphere, while intra-cluster task optima follow a von Mises-Fisher distribution with mean $\bm{Y}_j$ and concentration $κ$. Each task $t\in \{1, ..., T\}$ learning algorithm is considered to be a channel $Q_{t}(\mathrm{d}\bm{w}_t | \bm{w}_{t-1}, \mathcal{K}_{t-1}(\bmθ_{1:t-1}), \bmθ_t)$, where $\bm{W}_t$ is the learned parameter of task $t$ and $\mathcal{K}_{t-1}(\bmΘ_{1:t-1})=\bigcap_{i=1}^{t-1}B_{\bmΘ_i}(r_i(d))$. Since in continual learning, the learner does not need to reach $\bmΘ_t$ exactly to successfully learn task $t$, to avoid forgetting as well as reliable learning, it only needs for the learned parameter vector $\bm{W}_{t}$ to enter $B_{\bmΘ_t}(r_t(d)) \cap \mathcal{K}_{t-1}(\bmΘ_{1:t-1})$. Using this model, we derive an upper bound on $T$ ensuring $\mathbb{P}(\mathcal{K}_{T}(\bmΘ_{1:T}) \neq \emptyset) \rightarrow 1$ as $d\rightarrow \infty$ and $κ/d \rightarrow α$, where $α>0$ is a constant. Finally, for $T$ tasks sampled uniformly from $Φ$, we derive an upper bound for total information theoretic capacity under the non-forgetting constraint for each task. --- ### On Sleep-Wake Disorder Identification from Polysomnographic Representations Authors: Alvise Dei Rossi, Luigi Fiorillo, Giuliana Monachino, Alessia Calzoni, Julia van der Meer, Markus H. Schmidt, Claudio L. A. Bassetti, Silvia Santini, Francesca Faraci Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17451v1 Self-supervised polysomnography foundation models (PSG FMs) aim to capture clinically useful information beyond sleep staging, yet are rarely compared with sleep-staging encoders pretrained at comparable scale and modality coverage. Using a common whole-night, missingness-aware state-space downstream modeling framework, we compared frozen representations from prominent large-scale, multi-cohort sleep-stage-supervised encoders and self-supervised PSG FMs for multilabel identification of six sleep-wake and circadian disorder groups in 7,906 PSGs from an unseen registry. Although PSG FMs outperformed an EEG-only, single-cohort supervised baseline, this ordering reversed when supervised pretraining approached comparable scale and modality coverage. Transfer learning from neural sleep scorers achieved higher overall performance while exhibiting greater downstream data efficiency. These findings underscore the need for scale- and modality-matched pretrained supervised baselines when assessing PSG FMs' downstream potential. --- ### Formal-Verification-Guided Bayesian Optimization for Non-Intrusive Last-Mile Coverage Closure Authors: Chengjia Liu, Priyadarsi Mishra, David Kebo Houngninou, Michael Quinn, Aakash Tyagi, Jiang Hu Published: 2026-10-12 Categories: cs.AR, cs.LG arXiv: https://arxiv.org/abs/2610.17449v1 Last-mile functional coverage closure for modern chip designs remains costly because a mature UVM environment exposes many controls while direct hits on the remaining bins are sparse. We present a non-intrusive workflow that closes pre-defined bins in an existing, human-authored functional coverage model; it does not synthesize verification objectives from the design specification. The workflow first uses abstract-syntax-tree (AST) parsing and cone-of-influence (COI) tracing to map each residual target to existing testbench knobs. Under a matched 20-simulation budget, Bayesian optimization (BO) is compared with continued constrained-random verification (CRV) and random search over the same AST/COI-reduced space. Formal reachability then removes only targets supported by complete unreachability proofs, after which retained targets are grouped by formal difficulty and processed by group-specific BO under a second matched 20-simulation budget. Cleanup and grouping never add numerator bins; only simulator coverage artifacts do. On OpenTitan AES, first-pass reduced-space BO reaches 241/271 bins; formal cleanup removes two denominator bins; the integrated post-cleanup stage - difficulty grouping, policy routing, and group-specific BO - adds 20 simulator-confirmed hits, yielding 261/271 raw coverage (96.31%) and 261/269 effective coverage (97.03%). On Entropy Source, first-pass reduced-space BO reaches 2117/2277 bins; formal cleanup removes ten denominator bins; the integrated post-cleanup stage - difficulty grouping, policy routing, and group-specific BO - adds 93 simulator-confirmed hits, yielding 2210/2277 raw coverage (97.06%) and 2210/2267 effective coverage (97.49%). These results characterize the integrated diagnosis-and-search workflow; the present evaluation does not isolate the causal contributions of difficulty classification, policy routing, and post-cleanup BO. --- ### TableVision: Recovering Verified Structure from Historical Tables Authors: Todd Y. Zhou, James M. Snyder Published: 2026-10-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.17448v1 Gaps, skips, and mistakes are the normal state of a historical register: rows go missing, tables cross page breaks, columns slip under wrong names, and column sums disagree with the corresponding row entries. Readers read digits correctly but lose the table structure. On seven clean panels of election returns, the best open-source reader reproduces 99.0\% of the digits but files only 83.7\% of the figures in the correct cells and gets 11 of 30 columns correct. TableVision treats table structure recovery as the problem. On three degraded 1888 pages, Amazon Textract files only 1 of 1494 figures correctly; TableVision correctly files 99.2\%, and all 51 columns its exact-total check releases are correct. A dynamic program cuts a volume into blocks by how many columns sum to their last line; witnesses read the cells; the gate classifies each column as exactly reconciling, flagged as faithful to a page whose arithmetic is wrong, or withheld. A theory of the sum check says which errors it cannot see and why its optimal block is about $θ^{-1/2}$ rows at defect rate $θ$. On 2201 columns of Wisconsin electoral data, the gate accepts 2042, flags 118, and withholds 41; failure rises with column length $R$ as $1-(1-θ)^R$ at 1.1 defects per 1000 cells in general elections; official returns confirm 933 of 934 testable accepted sums and all 35 testable flagged totals; and the deployment added 491\,359 gated precinct rows to a 9.3-million-row dataset. One implication is that the design of the verification block is critical. --- ### The belief-action gap in language model agents Authors: Ayse Sila Okcu, Weisheng Wang, Tongmeng Xie, Niall Dalton, Angelos Nalmpantis, Raghu Arghal, Mario Giulianelli Published: 2026-10-12 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.17444v1 Language model (LM) agents can fail for different reasons, requiring different explanations and interventions. We study how state beliefs, action values, and action preferences interact as LM agents reason and commit to decisions in text-rendered grid worlds. We estimate task-relevant beliefs using report-based and representation-based probes, and action values using inverse reinforcement learning. We find that task-relevant state and value information is often represented more reliably than it is reflected in behaviour. Tracking action preferences throughout reasoning further shows that decisions sharpen around a commitment point and reported belief uncertainty decreases near commitment, although individual belief changes do not consistently explain shifts in action preference. Finally, forced actions follow reasoning prefixes from other decisions when transplanted, whereas the tested reasoning-window activation edits provide limited control over the action. Together, these results suggest that some failures arise not from missing task-relevant information, but from how available information is translated into action. Our measurements distinguish information availability from belief-action consistency without by themselves identifying the cause of each failure. Our code is available at: https://anonymous.4open.science/r/belief-action-gap-F745/ --- ### Know-Thy-Not: Training-Free Targeted Negation in Dual-Encoder Vision-Language Models Authors: DoYoung Kim, Changkyu Choi, Adín Ramírez Rivera, YunSeok Choi, Jee-Hyong Lee Published: 2026-10-12 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.17440v1 Dual-encoder vision-language models systematically fail at negation, scoring ''a cat without a dog'' nearly identically to ''a cat with a dog.'' We show that this failure is inherent to the single-embedding framework. Under contrastive training, single-embedding methods can at best ignore the negated concept (dilution) but never penalize its presence (rejection). We propose Know-Thy-Not (KTN), a training-free method that introduces an image-side rejection dimension to penalize the negated concept directly in the visual representation. We also introduce COCO-Confusion, the first benchmark requiring genuine rejection to succeed. Across 9 backbones, KTN improves COCO-Confusion mAP by up to +25.6 over vanilla VLMs and +9.6 over the strongest existing method. --- ### Large Language Models Extend Known Research But Do Not Create New Research Directions Authors: Zhihua Liu, Kurt Butler, Damian Machlanski, Danilo Bueno, Sotirios A. Tsaftaris Published: 2026-10-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.17439v1 The success of autonomous research systems relies on the quality of the scientific hypotheses they generate. Recent studies report that human reviewers often rate LLM-generated ideas as more novel than those from human experts, raising hope that AI can accelerate scientific discovery. Yet it is unclear whether these models can do more than extend existing knowledge. Can LLMs drive genuine innovation by revealing new research directions, or are they limited to repeating and recombining what science already knows? This unresolved question lies at the heart of discussions about AI and the future of science. To address this, we distinguish two types of novelty: distance, which measures how far a hypothesis lies from existing work, and direction, which captures whether the hypothesis pursues new lines of inquiry beyond those already present. These two measures diverge sharply for LLM-generated work. We represent scientific papers in a shared conceptual space using a pretrained scientific-paper embedding model. Within this space, we embed 26.9 million human-authored papers across 1,325 research topics in six natural-science fields and 524,308 LLM-generated hypotheses from eight models. Distance is measured using existing metrics. To quantify direction, we introduce the orthogonal novelty score (ONS). LLMs consistently generate ideas more distant from existing work than human-authored papers, up to 16\% further when the LLM is given the papers farthest from the centre of the topic. However, LLM hypotheses have a 15--23\% lower median ONS than published papers from the same topic. Thus, LLMs move further from what is already known, but remain within established directions. This pattern persists across changes in various configurations. As a result, LLMs excel at extending the frontier but the creation of new directions is still led by human scientists. --- ### SchemaWeaver: Accelerating On-Device Mobile Agentic AI via Schema-Guided Parallel LLM Decoding Authors: Boyuan Yang, Wei Gao Published: 2026-10-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.17436v1 Large language models (LLMs) on mobile devices can act as agents that call external software tools to complete complex tasks, but the major challenge of supporting on-device mobile agentic AI with LLMs is the long latency of LLM tool calling on mobile devices. In particular, the decode phase is the main contributor of such long latency, but is hard to speed up on mobile processors as it is bound by memory bandwidth rather than compute power. To address this limitation, in this paper we present SchemaWeaver, a new on-device mobile agentic AI technique that enforces schema-guided parallel decoding in the decode phase of LLM tool calling, to reduce its latency by producing multiple tokens in each decode pass. SchemaWeaver can be generically applied to the LLM inference engine on mobile devices without involving any training efforts or changes on LLMs, and our experiment results on major mobile SoCs and LLM families show that SchemaWeaver can reduce the LLM tool calling latency by up to 3.9 times, compared to the competitive baselines. --- ### World-T3: Test-Time Training as Persistent Memory for Real-Time Video World Models Authors: Junyi Wu, Zekun Li, Chenhongyi Yang, Chen Guo, Lingchen Yang, Yongan Zhang, Xinyi Zhang, Hongbing Hu, Bartek Rymkowski, Srinath Sridhar, Yan Yan, Xijie Huang Published: 2026-10-12 Categories: cs.CV, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2610.17434v1 Recent advances in video world models have enabled plausible interactive environments, yet long-horizon generation places a fundamental demand on visual memory: the previously observed world must remain consistent as the model continues to generate. Existing autoregressive approaches often implicitly rely on the generative model itself to retain such information, leading to spatial memory decay beyond the active context. We introduce World-T3, a camera-controlled video world model with a test-time training memory that consolidates past observations into dedicated model parameters. To robustly update this memory, we propose a recursive least-squares rule that incrementally absorbs novel visual features via online pose-keyed memory updates. Our TTT memory reads out previously observed scene features from the fast-weight state by camera pose, enabling consistent generation with constant storage and per-step computation regardless of rollout length. To facilitate real-time interactive generation with this robust memory, we further distill the autoregressive model into a few-step generator via dual-teacher distribution matching. On a long-horizon revisit benchmark, World-T3 raises revisit PSNR from 13.52 to 18.31 dB and reduces absolute rotation error from 3.344 to 1.075 relative to the strongest baseline on both metrics, while sustaining 20.6 FPS at 1280 x 704 on a single NVIDIA B200 GPU. --- ### Monarch-Muon: Structure-Aware Muon for Fast and Memory-Efficient Pretraining Authors: Alexander Miasnikov, Ruslan Kabirov, Gleb Molodtsov, Dmitrii Vlashchuk, Vladimir Aletov, Nail Bashirov, Daniil Tikhonov, Maxim Krutikov, Aleksandr Shestakov, Aleksandr Beznosikov Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17429v1 Training large language models is constrained by both wall-clock time and memory footprint, and recent matrix-aware optimizers sharpen this tradeoff: Muon can improve optimization but its dense polar update remains expensive. We propose Monarch-Muon, a structured matrix-aware optimizer for weights parameterized by Monarch factors. The resulting structure reduces weights, gradients, and optimizer state, and exposes small independent update subproblems in place of a single dense orthogonalization; we formalize this through a blockwise trust-region interpretation. In quality runs, Monarch-Muon improves over Monarch-AdamW at fixed structure: 3.06 vs. 3.09 in the dense 257M model and 2.68 vs. 2.77 in the 1B MoE model; the two-block MoE variant also matches dense AdamW while using 15% fewer active and 25.4% fewer total parameters. In 6.89B-parameter systems measurements, Monarch-Muon reduces peak memory from 68.9 to 52.9 GB, Muon optimizer state from 14.60 to 6.77 GB, and the Muon optimizer step from 1288 to 304 ms. --- ### Reinforcement Learning In Partially Verifiable Environments for Colon Cancer Treatment Planning Authors: Li-Ching Chen, Andrés Franco Valiente, Yuta Ishiyama, Zachary R. Chalmers, William Chen, Kelsey Natsuhara, Taema Brinjikji, Madhumita Sushil, Travis Zack, Julian C. Hong Published: 2026-10-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.17426v1 Clinical decision-making is a sequential process of information collection, clinical inference, and forecasting under partial observability, often without a uniquely correct treatment decision. Large language models (LLMs) have demonstrated strong medical question-answering capabilities but extending these advances to clinical reasoning remains challenging because many clinical tasks are only partially verifiable. Recent clinical reinforcement learning (RL) work has therefore largely focused on diagnostic endpoints, where correctness can be more directly verified. Oncology treatment decision making is a particularly challenging instance of this problem that extends beyond medical question-answering in which multiple decision trajectories can be clinically defensible for the same partially observed patient state. We formulate oncology treatment planning as an interactive RL problem under partial observability and construct patient-specific, process-aware rewards from decision processes observed in real world clinical records. The agent must learn to sequentially acquire information from a structured patient representation before committing to a first-line treatment recommendation, and is evaluated on information gathering, treatment selection, and reasoning alignment. Using GRPO and GDPO to optimize Llama-3.1-8B-Instruct within this environment, RL substantially improves the base policy beyond frontier model capabilities. The GDPO-trained policy increases Mean@3 scalar clinical reward by $79.60\%$ over base Llama and exceeds GPT-5.2 by $18.82\%$ under the same interactive environment, while reducing query count during training. Overall, our results show that patient-specific, partially verifiable rewards can provide an effective learning signal for interactive clinical RL when both the patient state and the underlying clinical decision process are only partially observed. --- ### When Sparsity Breaks Connectivity: Functional Reachability and Guided Topology Dynamics in Extreme Sparse Training Authors: Mohammad Pasande, Mateusz Jaworski, Jeroen Rook, Elena Mocanu Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17425v1 Dynamic Sparse Training (DST) periodically prunes and regrows connections, inducing a discrete dynamical mechanism over network topologies. At extreme sparsity ($\ge 99\%$), this dynamic tends to destroy input-output information pathways, and performance with standard magnitude/random updates degrades sharply. We formalize forward, backward, and functional reachability on layered sparse graphs, derive closed-form expectations for reachable neurons and edges, and show that Monte Carlo realizations match the theory. Guided by this reachability-based structure, we bias pruning toward non-functional edges and regrowth toward functionally consistent candidates. We introduce this reachability-driven modification of topology evolution as Reachability-guided in DST, referred to as guided DST. On CIFAR-10/100 and Tiny ImageNet, guided DST recovers large accuracy gaps versus unguided DST (up to $+14.4$ points at $99.9\%$ sparsity on CIFAR-100), and on MLP-Mixer guided DST remains trainable at $99.99\%$ sparsity while unguided random regrowth degrades sharply. These results cast extreme-sparse training as a connectivity-preserving dynamical process. Our implementation is publicly available at https://github.com/mzjaworski/ea_dst. --- ### On Making $\mathrm{SE}(2)$-Invariant Networks Optimal Authors: Tomas Karella, Emily Shinkle, Alice Elisabeth Anastasia Allen, Pieter J. Swart, Nicholas Lubbers, Roxana Bujack Published: 2026-10-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.17424v1 Respecting a-priori known symmetries of underlying data is a key principle in designing efficient neural architectures. Convolutional neural networks are a canonical example, encoding translation equivariance. Extending this inductive bias to richer symmetry groups such as $\mathrm{SE}(2)$ in the most effective way possible is an ongoing architectural and computational challenge. Building on classical invariant theory, we derive invariant feature maps that are provably optimal: minimal and information-preserving with respect to the symmetry group. Our construction is designed to integrate seamlessly into standard architectures, requiring no modification to nonlinearities or normalization layers. As a concrete implementation, we introduce a plug-and-play module for convolutional neural networks that enforces $\mathrm{SE}(2)$ invariance, capturing both translations and rotations within standard architectures. The resulting models are both efficient and expressive. Experiments show competitive or improved accuracy compared to state-of-the-art equivariant methods at significantly lower computational cost. Finally, we provide a theoretical and empirical analysis of the relationship between equivariance and invariance, offering insight into their roles in controlling robustness and flexibility. --- ### ClearNav: Perception-Aligned Quadrotor Flight Using Differential-Physics Authors: Aditya Azad, Arthur Huang, Guanrui Li Published: 2026-10-12 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.17419v1 Autonomous quadrotor flight in cluttered environment is an key capability for tasks like inspection and search. Navigation policies trained in simulation and transferred to hardware can fly at high speed in open space, but they can still collide as obstacle clearance shrinks. We attribute most of these collisions to a single mechanism, namely motion in directions the onboard camera does not observe. Heading that is not aligned with the direction of travel and lateral excursion during obstacle avoidance both carry the vehicle toward obstacles outside its field of view. We introduce ClearNav, a navigation policy trained with two reward terms that constrain motion to the observed direction, a forward-looking heading target and a penalty on lateral velocity. A third term penalizes speed near obstacles so that an obstacle entering the field of view remains avoidable. In simulation, across 18 held-out scenarios with 2304 paired flights per policy unseen in training, ClearNav improves success by 6.8 points, cuts the collision rate by 65% relative to the state of the arts, and flies at maximum 5m/s. For the proximity term alone under matched commanded speeds, the collision reduction still holds at 70%. The effect is significant on static scenes and consistent in direction on moving obstacles, though not significant there at the suite level, showing that safer flight in clutter is possible even on the limited sensors and compute of small drones. We also validate our navigation policy in a real-world cluttered environment, showing that it enables the quadrotor to navigate through the cluttered environment successfully and safely. --- ### Why Agentic Workflows Fail: A Taxonomy Derived from Literature and Empirical Data Authors: Xin Li, Rami Bahsoon, Tao Chen Published: 2026-10-12 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2610.17418v1 Agentic workflows combine large language models with planning, memory, tools, execution environments, and iterative feedback. Their failures are difficult to diagnose because the same user-visible symptom may originate from model behavior, workflow orchestration, external tools, infrastructure, or their interaction. This paper presents a source-aware taxonomy of agentic workflow failures and applies it to evidence from academic literature and GitHub. The taxonomy combines source attribution with three empirical dimensions: a concrete failure-type dictionary, workflow-stage localization, and agent-capability localization. Its central structure is a two-axis localization matrix: workflow stages form the horizontal process axis and agent capabilities form the vertical functional axis. This design moves beyond taxonomies that mainly classify final failure manifestations by localizing intermediate process states inside agent operation. Our empirical comparison shows that literature-derived failures emphasize planning, reasoning, learning, perception, and knowledge-related mechanisms, whereas GitHub-derived failures concentrate on execution-stage, actuation-related, tool, API, and infrastructure failures. The dominant hotspot is iterative execution and actuation. These findings indicate that agentic workflow reliability should be treated as a systems problem rather than only a model-output problem. The taxonomy provides a basis for structured evaluation, root cause analysis, failure injection, and engineering triage. --- ### Bottlenecked semantic compression by representation reconstruction Authors: Xiaoya Tang, Shilong Liu, Daniel Smolyak, Xiwen Li, Quynh Nguyen, Tolga Tasdizen Published: 2026-10-12 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.17412v1 Vision Transformers incur high computational costs when processing long visual token sequences. Existing reduction methods typically rank tokens independently and discard those deemed unimportant or redundant. We propose a reconstruction-guided framework for visual token condensation (ReConV), which summarizes the complete visual sequence with a compact Summary Bank (SMB) and retains tokens whose information is not adequately captured by the SMB, as indicated by their reconstruction residuals. At each compression layer, a lightweight position-conditioned decoder reconstructs the current tokens, leaving shared information compressed in the SMB while explicitly preserving poorly reconstructed details. Experiments on image classification and multimodal understanding demonstrate favorable performance-efficiency trade-offs: ReConV surpasses the uncompressed baseline with 48\% fewer GMACs on ImageNet and enables reusable image-level representations for efficient multi-turn inference. Code is available at TokenDrop. --- ### Agency as a New Organizing Abstraction for Computer Science Authors: Ian Foster Published: 2026-10-12 Categories: cs.AI, cs.DC, cs.HC, cs.MA arXiv: https://arxiv.org/abs/2610.17410v1 Advances in large-scale machine learning have enabled agentic AI systems: persistent, autonomous computational entities that pursue goals, initiate actions, and interact with open environments. These systems are often framed as applications of foundation models. We argue that this framing is incomplete: agency is also a candidate new organizing abstraction for computer science. What is new is not the concept of an agent, nor runtime policy generation per se, but its general-purpose, open-ended form, in which systems can synthesize courses of action from goals, observations, and memory, over tools and environments for which no complete task-specific model was specified in advance. Thus agency becomes a systems-level abstraction as well as an AI modeling concept. Where the process abstraction organizes computation around executing programs, agency organizes it around pursuing goals under bounded resources: a process executes a program; an agent pursues a goal. From this perspective, the recurring failures of deployed agentic systems (runaway resource consumption, behavior redirected by adversarial content, opaque continuous operation) are symptoms of structural abstraction gaps, not isolated engineering defects. We examine these gaps across six dimensions (execution, interfaces, correctness, oversight, coordination, and persistence) and argue that closing them requires coordinated advances in systems, --- ### Position: Beyond Training Time, Test-Time Coordination is Essential for Cooperative MARL Authors: Dongsu Lee, Sooraj Sathish, Joonkyung Kim, Woojun Kim, Amy Zhang Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17409v1 We argue that cooperative multi-agent reinforcement learning (MARL) should treat test-time coordination as part of the problem definition rather than a deployment artifact. Our focus is on long-horizon, centrally monitored cooperative systems such as warehouse robotics, automated storage-and-retrieval systems, and cloud-resource scheduling, where teams share a reward, operate under shifting objectives, and already run with global telemetry at deployment. The dominant paradigm in this regime, centralized training with decentralized execution (CTDE), uses global information only during training and provides no coordination channel at execution. This is sufficient when agents are weakly coupled or the environment is stationary, but it fails when the team must realign after conditions change. The limitation is structural rather than algorithmic, and more training cannot add a channel that does not exist at execution time. We call this the recoverability gap. To address it, we name the underexplored design space of intermittent test-time coordination as centralized training with mixed execution (CTME), in which a coordinator observes global state and emits high-level signals while agents act locally. CTME is not a particular learning algorithm, but a cooperative MARL problem setting that makes test-time information structure explicit. It specifies team-level information, high-level signals, and update frequency after training, with CTDE recovered as a limit case. We do not argue that CTDE is obsolete. We argue that removing the test-time channel is a modeling choice, not a necessity, and the wrong default when global telemetry is available and teams must recover from drift. --- ### Cross-View Time-Frequency Diffusion for Extreme-Aware Time-Series Generation Authors: Lin Jiang, Rongchao Xu, Guang Wang Published: 2026-10-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.17407v2 Generating realistic time-series requires faithfully reproducing rare but high-impact extreme events. Existing approaches often rely on specialized representations or extreme-focused training objectives, which may restrict the diversity of captured extremes or compromise overall distributional fidelity. We propose Cross-View Time-Frequency Diffusion (CrossTF), a framework that exploits discrepancies between time- and frequency-view denoisers for extreme-aware time-series generation. Our key empirical finding is extreme-sensitive cross-view asymmetry: moving along the direction connecting the two views' clean-sample estimates substantially changes extreme fidelity while having limited effects on overall generation quality. We further identify a late-stage cross-view window in which this discrepancy stabilizes and cross-view adjustments have a particularly strong effect on extreme fidelity. These findings motivate a staged optimization strategy in CrossTF: after training and freezing both denoisers, it only needs to learn step-specific fusion parameters within this window using an Extreme Value Theory-based objective under an energy-distance constraint. This formulation supports interpolation between and extrapolation beyond the view-specific estimates while limiting degradation in overall generation quality. Extensive experiments on seven real-world datasets against 11 baselines show that CrossTF improves extreme fidelity by 26.1% and overall performance by 7.4% relative to the strongest baselines. These results highlight cross-view discrepancy as an effective basis for low-dimensional optimization in extreme-aware time-series generation. --- ### Event-Centric Earthquake Monitoring with a Multi-Station Transformer Authors: Haoyu Wang, Weiqiang Zhu Published: 2026-10-12 Categories: physics.geo-ph, cs.AI arXiv: https://arxiv.org/abs/2610.17404v1 Earthquake catalogs underpin studies such as fault structure, earthquake processes and seismic hazard, and their value depends on being both complete and reliable. Machine learning has improved individual tasks of earthquake monitoring, but processing stations independently discards the consistency of arrivals across the network, and running tasks in sequence fragments the information describing each earthquake. We present EQNet2, a geometry-aware multi-station Transformer that jointly predicts phase arrivals, first-motion polarities, source location and magnitude, and the association of arrivals into events. On a Northern California test dataset and on continuous waveforms from Ridgecrest and the San Francisco Bay Area, EQNet2 recovers more weak arrivals, produces fewer false detections, and associates more picks into located events. Its learned attention links information across stations along time--distance patterns consistent with seismic moveout. As seismic networks grow denser, exploiting multi-station coherence and event-centric prediction offers a path to more complete and reliable catalogs. --- ### Evaluating Temporal Reasoning in Large Language Models: Event Temporal Relation Extraction with Allen's Interval Algebra, Logical Consistency, and Relative Time in Narratives Authors: Adib Sakhawat, Fardeen Sadab, Tamjid Hasan Fahim, Multazam Mahmud, Tahsan Ferdous Lihan, Md Isfak Iqbal Chowdhury, Nafis Ahnaf Jamil, Md. Atik Shahriar, Sieam Shahriare, Wasi Omar Syed, M Safwan Hasan Khan, Tabib Hassan Published: 2026-10-12 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.17402v1 Understanding a narrative means knowing how every event relates in time to every other, yet large language models (LLMs) are mostly tested on isolated event pairs or coarse labels. We ask five LLMs to reconstruct the complete network of Allen's 13 interval relations for 360 English narratives whose events are placed relative to one another rather than by dates, and evaluate the outputs under a pre-registered protocol for agreement with our gold annotations, converse consistency and logical satisfiability. End to end, the best model clearly beats trivial baselines on macro-F1, our primary metric, yet under our pre-registered parser no model matches the accuracy of always answering "before", which the skewed labels inflate. Most of that accuracy gap is a parsing artifact: formatting errors in large matrices cost whole rows, and a post hoc parser that recovers the determined cells of malformed rows raises coverage from 0.60 to 0.84 and closes the gap to the baseline for the best model. On the cells they do answer, models recover precedence and containment but rarely overlap, and they fail most on flashbacks, where both the narration and the event list present the later event first. Logical consistency varies widely yet says little about agreement with gold: enforcing it raises accuracy mainly by filling empty cells, and is beaten there by the trivial control of guessing the majority label. Reliable structured output and robust handling of relative time are prerequisites for temporal reasoning over whole narratives. --- ### Generalizable field of view extension for diffusion MRI via conditional flow matching Authors: Gaurav Rudravaram, Lianrui Zuo, Ema Topolnjak, Michael E. Kim, Adam M. Saunders, Elyssa M. McMaster, Paula Trujillo, Ely E. Wesley, Colin D. McKnight, Pratik P. Pandharipande, Mayur B. Patel, Cayce Strength et al. Published: 2026-10-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.17401v1 Incomplete field-of-view (FOV) diffusion MRI (dMRI) acquisitions compromise whole-brain white matter analysis by excluding brain regions from tractography and connectivity studies. While deep generative models for dMRI FOV extension have been proposed, their generalizability remains limited by the acquisition-dependent nature of dMRI, where signal intensity varies with b-value, gradient direction, and scanner protocol, and the inherent 4D complexity of the data, making it difficult to train models that transfer across sites or acquisition schemes without retraining. We propose a flow-matching generative model that extends the dMRI FOV in both the superior and inferior directions by predicting apparent diffusion coefficients (ADC) in six directions, conditioned on the two nearest acquired slices and the full-FOV T1w image to enforce both local diffusion continuity and anatomical consistency. The ADC representation decouples the model from acquisition-specific b-values and gradient directions, providing a middle-ground across protocols. The model was trained on 3,103 subjects (5,105 sessions) across five different datasets spanning diverse acquisition schemes (b = 500 and 1000 s/mm$^2$; 21--192 gradient directions), four single-shell (b = 1000) and one multi-shell (b = 500/1000). Evaluated on 299 subjects across five independent datasets spanning b = 750--1500 s/mm$^2$ and 32--90 gradient directions, with single- and multi-shell schemes unseen during training, the proposed method achieves significantly improved bundle Dice score on downstream tractography (p < 0.01). --- ### Learning to Detect Shifts in Unstructured Data Streams via Monitoring-by-Betting Authors: Siyu Zhou, Jialin Ouyang, Fan Xia, Adarsh Subbaswamy, Drew Prinster, Suchi Saria, Augusto Garcia-Agundez, Sarah Pollet, Jinoos Yazdany, Patrick Vossler, Jean Feng Published: 2026-10-12 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2610.17399v1 Continuous monitoring of generative AI systems requires detecting distribution shifts in the unstructured data they ingest and generate. Existing monitoring methods typically control false alarms by requiring the discrepancy measure used to detect shifts to be *pre-specified*. However, for unstructured data, pre-specification creates a sensitivity--breadth trade-off: a narrowly defined measure is sensitive to the anticipated shift but may miss unexpected changes, whereas a broadly defined measure may detect a wider range of shifts but with reduced sensitivity. Although adaptively hypothesizing the shift direction can overcome this limitation, such data-dependent adaptation generally invalidates conventional false-alarm guarantees. We reconcile adaptivity with statistical validity through testing-by-betting, reframing sequential changepoint detection as a betting game in which the bettor continually updates its hypothesized shift direction, or witness function, as new observations arrive. We introduce *asymptotic e-detectors*, which approximate the pre-change distribution using a calibration set of size $m$, and prove that they asymptotically control the average run length to a false alarm as $m$ increases. Empirically, calibration sets as small as $m=30$ provide effective control. We then develop adaptive monitoring strategies ranging from black-box bettors that learn from embeddings to interpretable bettors based on LLM-proposed concepts. Across three real-world datasets, including data from a clinical AI scribe deployment, adaptive monitors detect shifts faster than fixed monitors at the same false-alarm level. Further, when the true shifts are sparse and interpretable, LLM-concept monitors provide the fastest and most interpretable detection, while also guiding root-cause diagnosis for complex shifts. --- ### JustOPD: On-Policy Distillation from Autoregressive to Diffusion Language Models Authors: Zikun Zhang, Jiayuan Sheng, David D. Yao, Wenpin Tang Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17397v1 On-policy distillation (OPD) is a post-training approach for large language models (LLMs) that enables a student model to learn a teacher model's distribution from sequences sampled from the student's own policy. Despite its promising capability, OPD for diffusion language models (DLMs) remains underexplored and existing studies primarily focuses on DLM-to-DLM distillation. The limited availability of strong teacher DLMs and the flexible generation order of DLMs impose additional technical challenges as well. However, it remains unclear how to effectively distill knowledge from a strong autoregressive language model (ARM) into a weaker DLM, given the mismatch between the left-to-right decoding of ARMs and the flexible decoding order of DLMs. Inspired by Ni et al. (2026), we propose JustOPD, an ARM-to-DLM on-policy distillation method that uses autoregressive rollouts during training to align the student's generation process with the ARM teacher's left-to-right decoding. This alignment enables effective distillation from ARM teachers while preserving the DLM's flexible generation-order capability at inference time. Experiments using Qwen3-14B as the teacher and Dream-7B as the student demonstrate that JustOPD consistently improves mathematical reasoning performance over the base DLM across multiple divergence choices. --- ### A Method for Exact Enforcement of Boundary Conditions on General Curved Triangular Domains: Algorithm and Application to Scientific Machine Learning Authors: Suchuan Dong, Yuchuan Zhang Published: 2026-10-12 Categories: math.NA, cs.LG, physics.comp-ph arXiv: https://arxiv.org/abs/2610.17392v1 We present a method for enforcing exactly Dirichlet, Neumann, and Robin conditions on general triangular domains with arbitrarily curved boundaries. The algorithmic constructions developed here complements those in a previous work for general quadrilateral domains. Our method combines (i) the exact mapping between a general curved triangular domain and the standard triangle, (ii) TFC constrained expressions, and (iii) Barnhill transfinite interpolations, to provide explicit closed-form trial ansatze that mathematically satisfy the imposed boundary conditions exactly. Corner compatibility induced by the boundary conditions is pivotal, and enforcing exactly the imposed boundary conditions hinges on the exact enforcement of the induced compatibility constraints. The corner compatibility associated with the acute vertices of the standard triangle and the coupling between gradient components when the hypotenuse corresponds to a Neumann/Robin boundary pose particular challenges. Two strategies are critical to resolving these issues, by reformulating conditional corner constraints into equivalent unconditional forms and by introducing a separate TFC-style auxiliary edge construction together with an associated linear solve on the hypotenuse. We analyze the corner compatibility constraints and present formulations for the general form of functions that exactly satisfy the imposed Dirichlet, Neumann, or Robin conditions on curved triangular domains. The proposed method has been implemented using extreme learning machine type randomized neural networks for solving PDE problems. We present ample numerical experiments with linear/nonlinear PDE problems on a variety of domains with complex boundary geometries. Numerical results show that the current method has enforced the Dirichlet, Neumann, and Robin conditions on curved domain boundaries to the machine precision. --- ### Learning to Play: Reinforcement Learning and Games Authors: Aske Plaat Published: 2026-10-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.17391v1 Amazing breakthroughs in reinforcement learning have taken place. Computers teach themselves to play Chess and Go and beat world champions. There is talk about expanding application areas towards general artificial intelligence (AI). The breakthroughs in Backgammon, Checkers, Chess, Atari, Go, Poker, and StarCraft have shown that we can build machines that exhibit intelligent game playing of the highest level. These successes have been widely publicized in the media, and inspire AI entrepreneurs and scientists alike. Reinforcement learning in games has become a mainstream AI research topic. It is a broad topic, and the successes build on a range of diverse techniques, from exact planning algorithms, to adaptive sampling, deep function approximation, and ingenious self-play methods. Perhaps because of the breadth of these technologies, or because of the recency of the breakthroughs, there are few books that explain these methods in depth. This book covers all methods in one comprehensive volume, explaining the latest research, bringing you as close as possible to working implementations, with many references to the original research papers. The programming examples in this book are in Python, the language in which most current reinforcement learning research is conducted. We help you to get started with machine learning frameworks such as Gym, TensorFlow, and Keras, and provide exercises to help understand how AI is learning to play. --- ### Cross-Market Meta-Learning for Electricity Price Forecasting Authors: Delaram Pirhayatifard, Terrence Alsup Published: 2026-10-12 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.17387v1 Day-ahead (DA) electricity price forecasting is usually posed per market, with a separate model fitted to each grid's own history. This caps accuracy: statistical and tree-based baselines saturate, while higher-capacity sequence models need more training data than a single market's few years of hourly prices can supply. Such methods also treat each node as an independent series, discarding the spatial dependencies induced by network topology and congestion. Because price formation across independent system operators (ISOs) is driven by shared physics and largely common market design, we instead treat cross-market history as a pretraining corpus. We present MetaGrid, a topology-aware transformer, meta-trained across markets so that a single zone-count-independent parameter set adapts to an unseen grid. On a leak-free benchmark of five U.S. ISOs, where every model receives the same prices, DA load forecasts and calendar features, MetaGrid attains the lowest average MAE, improving on the best baseline by 3.1\% and on the same architecture trained from scratch by 7.7\%. --- ### Measure Neural Mappings Authors: Ricardo Baptista, Edoardo Calvello, Bohan Chen, Andrew Stuart, Urbain Vaes Published: 2026-10-12 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2610.17384v1 We introduce and define measure neural mappings (MNMs), which combine measure-dependent neural transport with ensemble consistency for learning operators on probability measures. Ensemble consistency requires outputs from growing ensembles sampled from an input law to converge to the specified population output in probability. We establish quantitative continuum limits for attention on noncompact laws under strong sub-Gaussian and tail-integrability assumptions. These laws accommodate unbounded state spaces in scientific machine learning, including Bayesian inference and uncertainty quantification. For fixed self-attention transformers, the expected output Wasserstein error is bounded by the empirical input Wasserstein error plus an additional $N^{-1/2}$ attention term. These self-attention transformers are MNMs under the stated assumptions: pre-norm architectures follow from compact stability after normalization, while post-norm architectures are covered under the stated tail conditions. We prove corresponding cross-attention estimates and extend the results to multihead attention and grouped transformer evaluation, with convergence as the group size grows. For Set Transformers viewed as measure-to-vector mappings, we obtain an $N^{-1/2}$ bound on the expected Euclidean output error under the stated assumptions. Experiments on seven operators in dimensions $d=1,2,3,5,10$ support the predicted particle-count dependence, using one-dimensional Wasserstein computations and Sinkhorn approximations in higher dimensions. The observed Wasserstein and sliced Wasserstein decay rates are close to their respective theoretical sampling benchmarks. --- ### Variable Input Masked Autoencoders for Wrist-Worn Accelerometry Authors: Harish Haresamudram, Wanting Mao, Aditya Radhakrishnan, Maxwell A. Xu, James M. Rehg Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17383v1 Large-scale repositories of wearable sensor data have enabled the development of foundation models for human activities and behavior. However, existing models are pretrained using fixed input durations and sampling rates, despite substantial heterogeneity in recording conditions, downstream tasks, and deployment constraints. We introduce VarMAE, a variable input masked autoencoder that learns a single representation across sampling rates and temporal contexts. During pretraining we construct two views of the same signal: a long duration high frequency view, and a short duration lower frequency view, obtained via random cropping and downsampling. We perform masked reconstruction within each view and introduce a cross-resolution masked reconstruction objective that uses visible tokens from the long view to predict masked tokens from the short view. Cropping and downsampling impose complementary information bottlenecks, encouraging the model to learn representations that remain effective across input durations and sampling rate changes. We pretrain VarMAE on the large-scale NHANES dataset, producing a single pretrained encoder matching or outperforming controlled foundation model baselines on eight benchmark datasets. With only 3.82M parameters, VarMAE is parameter efficient and shows that a single foundation model can achieve broad transfer capabilities across sensing configurations. --- ### Mitigating Black-Box Score-Based Query Attacks via Gradient Confusion Authors: Ali Javeri, Javad Hezareh, Radin Shahdaei, AmirMahdi Sadeghzadeh Published: 2026-10-12 Categories: cs.LG, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2610.17382v1 Score-based black-box adversarial attacks pose a serious threat to deployed machine learning systems by exploiting repeated model queries to craft adversarial examples, requiring neither access to model parameters nor architecture. Existing defenses predominantly rely on stochastic output perturbations, which fail to prevent determined attackers from recovering stable gradient estimates by increasing query budgets or averaging over randomness. In this work, we propose the Gradient Confusion Defense (GCD), a retraining-free, deterministic test-time wrapper that actively shapes the probability vector returned to querying adversaries. By solving a constrained optimization problem grounded in the model's Jacobian, GCD aligns the exposed logit-gradient with a target confusion direction that is maximally misaligned with the true adversarial loss gradient---increasing the query complexity of attacks without sacrificing predictive utility or top-1 accuracy. We provide rigorous theoretical analysis demonstrating that GCD provably slows gradient-estimation-based attacks and induces systematic accept/reject errors in random search-based attacks, with the confusion effect amplifying as the fraction of defended queries increases. Extensive experiments on CIFAR-10 and ImageNet across multiple architectures (ResNet-18, ResNet-50, ViT-B/16) and four state-of-the-art attacks confirm these predictions: GCD achieves state-of-the-art robustness, while incurring zero clean accuracy degradation. %Crucially, unlike randomization-based defenses, GCD's deterministic structure provides resilience against adaptive attacks that exploit stochasticity, making it well-suited for security-critical deployment settings. --- ### Robustly Reliable Learning for Regression Authors: Avrim Blum, Anh Tuan Nguyen Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17381v1 We introduce the problem of robustly reliable learning for the regression setting in adversarial environments. Given a test point $x_{test}$, a training dataset $S$ of $n$ data points, a corruption budget $b \ll n$, and a real-valued hypothesis class $\mathcal H$, our goal is to output the optimal prediction range $I^*_{x_{test}}(S, b)$, such that any $L$-Lipschitz predictor consistent with all but at most $b$ points in $S$ must predict a value $h(x_{test}) \in I$. We then instantiate the analysis for $L$-Lipschitz hypothesis classes, a natural and widely studied class in the regression setting, focusing on two key scenarios. First, for the restricted case where the corrupted dataset is guaranteed to be $L$-Lipschitz consistent, we propose a simple and efficient $b^{th}$-best heuristic algorithm and prove its correctness in computing the tightest possible range $I^*_{x_{test}}(S, b)$. We then establish its computational complexity of $O(n)$ and show that it is computationally optimal in this scenario. Building on this foundation, we address the general scenario where the dataset $S$ is no longer $L$-Lipschitz consistent. We demonstrate that computing the optimal range in this setting reduces to analyzing a violation graph, which models the pairwise $L$-Lipschitz violations among data points. By leveraging the structural properties of this graph's kernel, we propose a Fixed-Parameter Tractable (FPT) algorithm that computes the exact optimal range in $O(n^2 + b^4)$ time. Furthermore, under the Strong Exponential Time Hypothesis (SETH), we establish a computational lower bound of $Ω(n^{2 - o(1)})$, proving that our algorithm is computationally optimal. Finally, we discuss open directions for the problem of robustly reliable learning for regression. --- ### EPOCH: Coverage-Aware Evolution with Experiential Preference Optimization for LLM-based Automated Heuristic Design Authors: Oguzhan Gungordu, Ayse Tursucular, Siheng Xiong, Faramarz Fekri Published: 2026-10-12 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.17379v1 LLM-based automated heuristic design (AHD) discovers heuristics for combinatorial optimization problems (COPs), where an LLM generates candidates in an evolutionary search. Existing frameworks freeze the generator or fine-tune it on a scalar fitness signal hiding which instances each heuristic covers. Reflection-based frameworks improve heuristics with natural-language feedback, but it stays in the prompt and never reaches the generator's weights. We propose EPOCH, which makes instance coverage the objective of the evolutionary search and fine-tunes the generator to internalize reflection. Parent selection, operator allocation, and stagnation recovery all optimize how well the heuristics jointly cover the training instances, so the search returns a complementary heuristic ensemble. The generator learns from its own search by experiential preference optimization. A teacher LLM returns a natural-language insight on the generated heuristics, and the generator produces new heuristics with this insight. These improvements join the preference pairs for direct preference optimization (DPO). Since DPO only teaches the generator to rank these improvements, a reflective self-distillation (RSD) term makes it more likely to generate them on its own, and the gains accumulate over the long horizon of the search. Across COPs and open-source LLM backbones, EPOCH's heuristic ensembles outperform inference-only and fine-tuned baselines and generalize across problem scales. --- ### Get In, Decide, Prove: A Merchant-Side Readiness Benchmark for Agentic Commerce Authors: Adrian de Valois-Franklin, Alex Bogdan, Anas Ibrahim, Ibrahim Shaer, Tamaz Andguladze Published: 2026-10-12 Categories: cs.AI, cs.CY, cs.HC arXiv: https://arxiv.org/abs/2610.17378v1 AI agents now browse and buy on people's behalf, and benchmarks measure how well the agents perform. Far less is known about the other side of the transaction: whether a merchant's website lets an agent in, publishes terms precise enough to decide a purchase, and leaves evidence of what was agreed. This paper presents a merchant-side benchmark built on those three questions and reports a pilot of it. The central measurement, decision sufficiency, rates each criterion of a concrete buyer mandate (total cost, cancellation or return terms, service date, variant identity) as determinate, indeterminate, or unknown from the merchant's own published evidence. Readiness scores carry their evidence coverage and missingness bounds; access is reported separately for a declared bot, a declared headless browser, and a user-present browser; and each observed term is mapped to the RAILS admissibility class of the evidence that could support it in a dispute. Four large merchants received the full assessment: three refused the declared profiles at their offer pages, and no mandate could be confirmed as met, because each merchant left a total or a refund term undetermined. An access census of 555 sites of merchants that two payment processors publicly name as clients found the home page unreadable to both declared profiles at 105 sites (18.9%). A declared assessment of 330 merchants, scored by deterministic rules without a rater, found that evidence coverage rather than the score decides who can be ranked: 115 merchants (35%) reached the ranking threshold, rankable profiles scored a median of 89 to 90, and material terms was the one pillar well below the others, because the all-in total was decidable at 18 merchants. The pilot shows that the measurements can be taken; it does not validate them or rank merchants, and the weights and the projected clearing band await validation against outcomes. --- ### What Makes a Latent Space Shared? Direct and Recoverable Compatibility in Motion VAEs Authors: Iago Brito, Walcy Rios, Lucas Fernandes, Kaiki de Oliveira, Arlindo Galvão Filho Published: 2026-10-12 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2610.17376v1 A shared latent space is often assumed to make representations from different schemas directly interchangeable, yet sharing parameters, dimensionality, or a prior does not determine how corresponding instances are expressed in latent coordinates. We show that this assumption conflates recoverable correspondence, direct compatibility, and decoder reuse. In a controlled multi-schema motion VAE on HumanML3D, even identical inputs yield only $5.9$--$17.7\%$ raw Recall@1 and $0.221$--$0.484$ paired-token cosine, despite correspondence being almost perfectly recoverable after linear calibration. Cycle consistency strengthens recoverability but still leaves direct retrieval and token agreement near zero. In contrast, only $64$ paired clips raise raw Recall@1 to $88.4$--$90.3\%$ and token cosine to approximately $0.82$, making most post-hoc alignment unnecessary, although crossed decoding remains substantially worse than native reconstruction. We further find that reducing measured marginal discrepancy does not imply schema invariance. These results show that a shared backbone or prior does not ensure directly compatible latent coordinates, while limited paired correspondence can substantially improve direct compatibility in the tested setting. --- ### PIG: Parallel Inconsistency Guidance for Accelerating Diffusion Language Models Authors: Quentin Fruytier, Anton Xue, Sanjay Shakkottai, Sujay Sanghavi Published: 2026-10-12 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.17375v1 Diffusion language models accelerate generation by decoding multiple masked positions per forward pass. However, accuracy is known to drop when parallel decoding trajectories deviate from their sequential counterparts, which are slow and expensive "oracles" that adhere to the joint distribution of language. In this paper, we propose Parallel Inconsistency Guidance (PIG), a new paradigm that uses a trained auxiliary model to directly estimate the inconsistency between parallel and sequential generation from a single model forward pass. During inference, masked positions are traversed by decreasing confidence, and tokens are added to the parallel decode set one-at-a-time until the total estimated inconsistency between the parallel and sequential generation exceeds a threshold. Surprisingly, it suffices to train a lightweight ~2M parameter auxiliary model on a small mixture drawn from OpenCoder, GSM8K-train, MATH-train, and MBPP-train. For LLaDA-8B Instruct and NemotronDiffusion-3B Instruct, PIG extends the accuracy-TPF frontier on GSM8K, MATH500, MBPP, HumanEval, and MMLU, when compared to the industry standard Fast-dLLM and measures favorably on LLaDA-8B Instruct against DAPD, KLASS, EB-sampler, and other learned unmasking policies. --- ### HarnessTTT: Learn to Optimize Harness at Test-Time for Scientific Discovery Authors: Yingzi Ma, Yulong Cao, Nanxi Li, Fang Wu, Yizhe Zhang, James Zou, Yejin Choi, Chaowei Xiao Published: 2026-10-12 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.17370v1 Scientific discovery with language models requires finding the best valid solution within a finite computational budget. We study how feedback from this search can improve the procedure used to conduct it. We introduce HarnessTTT, a test-time learning framework that trains a harness proposer while keeping the executor frozen. The proposer generates executable harnesses comprising system prompts, reusable skills, tools, middleware, and execution controls. Each harness guides a budgeted discovery rollout, whose outcome provides feedback for updating the proposer. A discovery-oriented policy-gradient objective combines gap-normalized rewards, leave-one-out advantages (RLOO), and max-sharpening to favor high-performing proposals. Generated components are validated before execution, and the search retains the best valid solution and selected harnesses across rounds. Across fourteen discovery tasks in mathematics, systems, GPU kernels, heuristic programming, astrodynamics, and biology, HarnessTTT with a frozen 9B executor improves on OpenEvolve with the same executor on all eleven benchmark tasks (e.g., from 1.17 to 2.64 on circle packing), improves on the discovery-fine-tuned Finch-9B by up to 92%, holds the best open-weight sub-10B result on nine, and reaches 0.380919 on Erdős minimum overlap, better than the best human reference of 0.380927. Scaling the proposer and frozen executor to 27B improves on eleven of fourteen tasks, and transferring the learned harnesses to GPT-5.5 matches or exceeds the strongest reported system of any backbone on nine of fourteen tasks. Together, these results show that harness search can itself be learned at test time: a trained proposer lets a small frozen executor search effectively, and the harnesses it finds carry over to stronger executors. --- ### HASTE: Hierarchical and Attribute-aware Supervision for Generative Recommendation Authors: Tripti Shukla, Shwetha Somasundaram, Akshat Gupta, Rohit Garg, Paul Duff, Ian Anderson Published: 2026-10-12 Categories: cs.IR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.17368v1 Generative recommendation models have shown promising results by representing items as semantic IDs and generating these IDs autoregressively. We introduce an approach showing that explicitly incorporating hierarchical and attribute-level supervision into generative recommendation can simultaneously improve training efficiency and long-tail recommendation performance. Existing approaches typically optimize semantic ID prediction with a standard causal cross-entropy objective, which does not explicitly exploit the hierarchical structure of semantic IDs and can require substantial training to learn effective item representations. Moreover, data-sparse tail items often suffer from poor representations due to limited supervision, resulting in weaker long-tail recommendation performance. We address these challenges with HASTE (Hierarchical and Attribute-aware Supervision for generaTive rEcommendation), a unified training framework that introduces hierarchical and attribute-specific supervision. HASTE's Hierarchical Loss explicitly exploits the multi-level structure of semantic IDs, providing supervision at each level of the item hierarchy and substantially accelerating training convergence. Its Attribute-Specific Losses encourage the model to preserve fine-grained item attributes, leveraging the language model's inherent semantic knowledge to improve representations and recommendations for tail items. We extensively evaluate HASTE on a fashion recommendation benchmark and across multiple domains of the Amazon dataset. HASTE improves recommendation performance across datasets and item-frequency regimes. Notably, the Hierarchical Loss achieves approximately 2x faster convergence than the standard causal cross-entropy objective, substantially improving training efficiency while maintaining strong recommendation quality. --- ### When Goals Leak Across Relationships: How Cached Goal Representations Shape Cross-Client Decisions Authors: Saloni Gupta, Sébastien Krier, Glenn Matlin Published: 2026-10-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.17365v1 As LLM agents take on more autonomous roles, they may interact with multiple clients and counterparties whose goals and incentives conflict. If an agent maintains shared inference state across these interactions, information associated with one relationship may influence decisions made for another. We test OLMo-3-7B-Instruct in a controlled two-client environment with 20 market scenarios, where each client is assigned an opposing objective. With shared conversational memory, the model deviates from the current client's objective in 22% of eligible decisions. We then investigate how these goals are represented in the KV cache. Linear probes show that client-goal ownership becomes highly recoverable in middle layers and is concentrated in particular attention heads. We use these results to identify where to intervene in the cache and test whether these representations affect the model's decisions. Ablating the competitor's cached goal representations at these locations shifts the model toward the current client. The effect is larger for heads identified by probing than for randomly selected heads, and larger when ablating goal tokens than matched non-goal tokens from the same client. Our results show that client-specific goal representations persist in the KV cache and can causally influence decisions made for other clients, highlighting the importance of relationship-specific state isolation in persistent LLM agents. --- ### UMCR: Universal Medical Concept Recognition through Guideline Integration and Agentic Linking Authors: Fabrice Lamarche, Amal Zouaq Published: 2026-10-12 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.17363v1 Medical Concept Recognition is the task of extracting and grounding spans in medical text to entries in a standard ontology. Current methods are generally developed towards a single ontology and source text type (clinical notes, drug labels, etc.) The central challenge is therefore to build methods that adapt to any target ontology without requiring ontology-specific retraining. More recent approaches like REAL and AutoPCR make direct use of Large Language Models and can theoretically adapt to any ontology choice but suffer from a lack of generalization. We propose UMCR, a pipeline that allows extraction and linking to any target ontology using any source text variety, utilizing direct guideline integration and an agentic linking module. We evaluate our method on the recent MedPath benchmark, covering a wide variety of medical text types and vocabularies, and show that our method achieves strong performance across these tasks. This ability unlocks interoperability in the healthcare domain, allowing for extraction and alignment between different medical standards. Our method also reduces dependence on parametric LLM memory, thanks to improved interaction with the target ontology compared to other proposed approaches. --- ### claimcheck: Measuring and Improving Specification Faithfulness Authors: Martin Amster, One An, Holly Baker, Fernanda Graciolli, Simon Henniger, Nada Amin Published: 2026-10-12 Categories: cs.CL, cs.AI, cs.PL arXiv: https://arxiv.org/abs/2610.17358v1 Formal verification proves that a program satisfies its specification, but says nothing about whether the specification captures what the author meant. As LLMs increasingly write both code and specifications, a gap between the author's natural-language requirement and the specification becomes a live risk. A machine-checked proof of a property narrower than the one requested is worthless but passes every check. Closing this gap is solving the spec formalization problem: given a natural-language requirement, produce a formal artifact whose meaning matches it. To measure the gap, we curate an end-to-end benchmark for spec formalization spanning four formal languages: Dafny contracts and Lean 4, Isabelle and Rocq theorem statements. Each of its 3,234 rows pairs a natural-language requirement with a formal artifact and a label established by execution, by proof, or by the benchmark it is drawn from. A third of the rows are meaning-preserving rewrites of the artifact that keep the faithful label, so a judge cannot score by detecting that a row was edited. We introduce claimcheck, a tool that detects incorrect formalizations: the formal artifact is translated back into plain English by a model that never sees the requirement, and a separate model, the judge, decides whether the informalization and the requirement express the same guarantee. On all four languages it gains 8.2 to 9.8 points of macro recall over a single call shown the requirement and the artifact, and 4.2 to 6.0 over that call handed the same list of meaning-preserving rewrites. Showing the judge the artifact as well costs 1.4 to 4.5 points. Across four judges that also see the artifact, the gain shrinks as the judge gets stronger, to within a point of the informed single call for the strongest. --- ### Brain2Face: A Study on the Relation of EEG Signals and Spatio-temporal Head Avatars Authors: Camilo Andrés Martínez Martínez, Olaf Dünkel, Kartik Teotia, Maurice Rekrut, Christian Theobalt, Marc Habermann Published: 2026-10-12 Categories: cs.CV, cs.HC arXiv: https://arxiv.org/abs/2610.17356v1 Driving digital head avatars is classically done by hand-crafted artist design or by image-based facial performance capture. In this work, we ask whether spatio-temporal tracking of the humans's head geometry and appearance can be performed solely from brain signals, i.e., Electroencephalography (EEG). Since prior work mostly focused on activity and emotion recognition, this task is relatively unexplored without even having publicly available high-quality datasets and benchmarks. Thus, we contribute a dataset featuring paired EEG and human head geometry as well as appearance annotations. We synchronize 64-channel EEG with per-frame expression latent codes of a personalized Gaussian head avatar. Second, we introduce a baseline to the task of expression recovery of a head avatar from EEG signals. To this end, we propose a dedicated EEG-to-expression model that takes a temporal sequence of EEG signals as input and regresses the latent expression parameters of the 3D Gaussian head avatar. Lastly, we validate our capture and algorithmic design choices through a benchmark protocol that we introduce to this task. We believe this work paves the way towards better understanding of the relation between EEG signals and facial surface as well as appearance changes. --- ### WIRE: Weight Space Implicit Representation Encoding of Frozen Video LLMs Authors: Ameen Ali, Tamim Zoabi, Lior Wolf, Chaim Baskin Published: 2026-10-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.17355v1 Video language models represent a video as a long sequence of visual embeddings in a decoder's input. Every visual embedding is processed by every decoder layer during prefill, stored in the key value cache, and attended by every generated token, so a long video raises prefill, memory, and generation cost at once. We present WIRE (Weight Space Implicit Representation Encoding), which reduces this recurring cost by moving part of the video conditioned computation of a frozen video language model from its input activations into a temporary low rank update of its weights. WIRE keeps a short explicit visual context, so the update acts as a correction rather than a replacement of the video. WIRE first compresses the full resolution embeddings of each frame into a small set of carrier tokens using a question conditioned bidirectional selective scan, and physically contracts the decoder input so every decoder layer, cache entry, and decoding step operates on the shorter sequence. Hypernetworks at selected decoder layers read the compressed forward pass together with a readout of the full resolution features removed by contraction and generate LoRA updates to the attention and feedforward projections. The updates are merged after prefill, so autoregressive decoding runs the unmodified frozen backbone. On Qwen3-VL-8B, WIRE with 16 carrier tokens per frame contracts the visual stream by $3.5\times$, reduces decoder prefill time by $3.35\times$ at 256 frames and $4.14\times$ at 1024 frames, shrinks the key value cache proportionally, and keeps per token decoding latency approximately constant. Across nine benchmark settings at 256 frames, it obtains the highest average accuracy among efficient video language models we evaluate under a common 256 frame budget. --- ### MID: Mask-Image Distributional Divergence for Evaluating Medical Image Segmentation Authors: Vincenzo Marcianò, Xiaoming Zhang, Gianluca Guglielmo, Michela Antonelli, Sebastien Ourselin, Maria A. Zuluaga Published: 2026-10-12 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.17354v2 Current evaluation protocols for medical image segmentation remain largely anchored in sample-wise comparisons, e.g., Dice and Hausdorff distance, which require each prediction to be paired with a ground-truth mask. Yet two limitations arise in practice: they are inapplicable when predictions are made on unlabeled data; and even when annotations are available, they cannot assess whether a collection of predictions matches the distributional structure of the ground-truth population. We introduce MID (Mask-Image Distributional Divergence), a distribution-level evaluation metric for medical image segmentation. MID builds upon the observation that segmentation quality is a property of the joint image-mask distribution: a mask may be anatomically plausible in isolation yet incorrect for the image it accompanies. MID compares the joint distribution of images and masks between a reference set and a set of predictions, without requiring sample-wise correspondence. This captures both mask realism and image-mask consistency. We evaluate MID across 13 publicly available datasets spanning CT, MRI, and MRA. Under anatomically realistic corruptions across 56 organ groups, MID tracks mean Dice and Hausdorff distance without per-sample ground-truth pairing. MID also detects distributional shifts across modalities and anatomies, identifies image-mask misalignment through controlled shuffling experiments, and produces reliable model rankings across real segmentation models. Official implementation available at github.com/erc-caravel/pytorch-mid --- ### HairFlow: Modeling 3D Hair Strands with Multi-Modal Conditioned Diffusion Authors: Yating Wang, Kui Wu, Ran Yi, Zixuan Lu, Tongtong Wang, Zhongtian Zheng, Lizhuang Ma, Yuefan Shen Published: 2026-10-12 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.17348v1 Modeling 3D hair strands is challenging due to complex hairstyle geometry, scarce high-quality training data, and incomplete or noisy input observations. Foundation models offer broader hairstyle priors, but their outputs are not directly usable for strand-based rendering and simulation. Existing approaches require costly per-instance optimization to fit hair priors to meshes from foundation models and 2D orientations from specialized predictors trained on limited hair data, restricting efficiency and generalization. We present HairFlow, a multimodal conditioned diffusion framework for efficient and controllable hair strands modeling. HairFlow conditions strand generation on two complementary representations: a coarse mesh that constrains global shape and multi-view sketches that specify local strand flow. HairFlow combines an Orientation VAE with a conditioned diffusion transformer to generate volumetric fields that are traced into explicit strands. To overcome the lack of paired sketch-strand training data, we introduce an automatic synthesis pipeline that generates geometrically aligned multi-view sketches with diverse styles representative of inference-time inputs. Experiments show that HairFlow generates plausible 3D strands that faithfully follow the input conditions across diverse hairstyles, substantially reduces inference time, and supports flexible multimodal control. --- ### On the Statistical Mechanics of Attention Authors: Jerry Yao-Chieh Hu, Yu-An Chen, Maojiang Su, Han Liu Published: 2026-10-12 Categories: cs.LG, cond-mat.dis-nn, cond-mat.stat-mech, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2610.17347v1 We revisit the statistical mechanics of Transformer attention via modern Hopfield models. We present a free-energy formalism to characterize parameter regimes and critical thresholds for robust attention inference. The resulting free-energy landscape incorporates randomness in query states and attention weights. Attention then emerges as the equation of motion of the corresponding statistical-mechanical system (i.e., modern Hopfield model) through the standard free-energy minimization principle. This allows us to identify first-order phase transitions in attention between glassy and robust-inference phases. We further extend to attentions with stochastic queries and keys. Together, our results provide a rigorous statistical-mechanics framework for characterizing phase structure and inference reliability across a family of Transformer attention mechanisms. --- ### Monitoring for Shifts at Unknown Times: Parameter-Free Conditional Conformal Martingales Authors: Tuo Liu, Drew Prinster, Francesco Orabona Published: 2026-10-12 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2610.17343v1 Conditional conformal test martingales provide sequential tests for detecting distribution shifts, using a fixed reference dataset to avoid contamination from test-time samples. Prior work used a penalized betting function via online optimization with smoothing, clipping, and other hyperparameters, which can adapt slowly when a shift occurs after a long stationary period. In this work, we introduce an approach that requires no hyperparameters and provably adapts quickly to shifts occurring at unknown times. Our key idea is to frame the penalized betting problem as portfolio selection in a three-asset market. Then, we use a restarting universal portfolio, with a heavy-tailed wealth allocation over time. We establish anytime false-alarm control and derive optimal finite-sample bounds on detection delay under detectable alternatives. We also develop an FFT-based implementation that computes the portfolio sequence over $T$ rounds in $\mathcal{O}(T^2\ln T)$ time, instead of $\mathcal{O}(T^3)$ of the naive implementation. Synthetic and real-world experiments suggest that the proposed approach adapts rapidly after long periods of null samples, with detection delays depending substantially less on shift time than prior baselines. --- ### Budgeted KV-Cache Repair for Evolving Contexts Authors: Han Yu, Wenhui Zhu, Jincheng Cao, Rui Cai, Yuanda Xu, Zhengze Zhou, Xiwen Chen, Hao Wang, Xuanzhao Dong, Han Shi, Menglin Zhou, Shayan Mohajer Hamidi et al. Published: 2026-10-12 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.17338v1 Language-model agents revise their working context as information becomes obsolete, but deleting text does not erase its influence on cached computation. Later key-value (KV) states remain conditioned on the deleted content, while recomputing the entire retained suffix can be costly. We introduce budgeted KV-cache repair: recovering task performance by recomputing a user-chosen fraction of the retained suffix. Comparing repair allocations under identical relevance scores and token budgets, we find that refreshing coherent regions can substantially outperform refreshing isolated tokens. This finding motivates Semantic-Unit Refresh (SUR), which allocates the budget to query-relevant text regions using inexpensive embedding scores, without training or an auxiliary transformer forward. To execute these repairs efficiently, a layer-packed executor processes selected states together at each layer while preserving full causal cache access. Across five models, SUR achieves strong recovery in controlled deletion tasks while recomputing only 2-5% of the suffix. On four natural multi-hop QA datasets, it achieves near-full-recompute exact match at a 5% budget with lower post-edit latency. The executor accelerates fragmented refresh by up to 16.1x over interval-wise execution. --- ### Accounting for model error in ensemble variational data assimilation through learned generative dynamics Authors: Marc Bocquet, Tobias S. Finn, Sibo Cheng Published: 2026-10-12 Categories: stat.ML, cs.LG, physics.ao-ph arXiv: https://arxiv.org/abs/2610.17333v1 Many approaches have recently been proposed to combine data assimilation with machine learning for improved forecasts with geophysical models. Here, we focus on combining advanced sequential data assimilation schemes with a generative forecasting model which accounts for complex model errors. This generative model provides a stochastic approximation of the conditional distribution of the successor state given the current state. The proposed variational formulation is a novel one-step weak-constraint 4D-Var scheme in which a learned conditional stochastic forecast replaces the prescribed additive Gaussian model error representation. We derive and implement this formulation as an ensemble variational algorithm, the generative iterative ensemble Kalman filter (IEnKF-G), which generalises the iterative ensemble Kalman filter with additive Gaussian model error (IEnKF-Q). We extend the method to a smoother with an assimilation window spanning multiple time steps and propose a simplified variant. The IEnKF-G is evaluated on a stochastic variant of the Lorenz 1963 model to assess its robustness. It is then tested on a multi-scale variant of the one-dimensional Lorenz 1996 model, in which the resolved variables are affected by model error originating from the hidden small-scale variables. The algorithm is then compared to an ensemble Kalman filter that adaptively mitigates model and sampling errors, and to the IEnKF-Q. The IEnKF-G is generally more accurate in the tested configurations, with substantial gains under partial observations and significant model error. This demonstrates the potential of generative data-driven models when paired with classical data assimilation methods, particularly when model error is substantial. ---