# Research Papers — 2026-09 · bundle page 97 of 266 > Full-text markdown of papers #4801–4850 of 13298 published in 2026-09, newest first. AI-readable mirrors of the full 13298-paper 2026-09 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-09/page/97 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/96.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/98.md) ### Generalizable Robotic Insertion with World Models Authors: Nicklas Hansen, Iretiayo Akinola, Yijie Guo, Jie Xu, Bingjie Tang, Hao Su, Xiaolong Wang, Abhishek Gupta, Dieter Fox, Yashraj Narang Published: 2026-09-23 Categories: cs.RO, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.28258v1 Robotic assembly in high-mixture settings requires adaptable systems that can handle diverse parts, yet current approaches typically rely on policies specialized to each insertion task. Although this can reach high success rates, it makes the process of deploying systems for new problems tedious and time consuming. We present a framework for generalizable insertion using world models that combine robot proprioceptive information with raw visual observations captured by a wrist-mounted camera. Our model-based approach trains a single world model on up to 90 insertion tasks with geometrically diverse parts, achieving 56% zero-shot success on unseen objects with unknown geometry compared to just 7% with a model-free baseline. Importantly, performance improves as more objects are included in the training dataset, demonstrating strong scalability. Lastly, finetuning the generalist model on held-out objects significantly enhances data-efficiency compared to training from scratch and, in some cases, achieves better asymptotic performance. To our knowledge, this is the first system capable of assembling unseen objects in an entirely data-driven manner, and thus represents a significant step toward scalable, generalizable robotic assembly systems. --- ### MemBodied: Recurrent Associative Memory for Vision-Language-Action Models Authors: Tej Deep Pala, Navonil Majumder, Bryce Goh, Raphael Yee, Jianfei Yang, Liming Chen, Soujanya Poria Published: 2026-09-23 Categories: cs.RO, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2609.28256v1 Vision-Language-Action models provide a strong foundation for general-purpose robot control, yet a vast majority of policies do not preserve and leverage episode-level information beyond the current observation. This limitation is consequential in history-dependent manipulation tasks that depend on information available only in past observations. Retaining past observations in context can aid in recovering this information, but at the significant cost of ever-growing, bloated context and inference latency. We thus introduce MemBodied, a fixed-size episodic memory with two complementary components: an associative state that records interactions across policy calls and an episode anchor that preserves a compact representation of the initial scene as a reference. At each policy call, the model conditions action generation on the current input and the memory components, rather than directly using past observations. Across five evaluated RMBench tasks requiring memory, MemBodied achieves $7.81\times$ the mean success rate of a stateless policy and $2.98\times$ of vanilla recurrent memory, while outperforming the strongest memory-augmented baseline by $1.3\times$ with $10\times$ fewer added parameters. On the fully observable LIBERO-Long suite, it reached 90.6%, a 5.4% improvement over the stateless $π_0$ policy. These findings support MemBodied as a practical alternative to expanding the policy context for history-dependent manipulation. --- ### Complementary Roles of Activation and Parametric Memory in Few-Shot Learning Authors: Miaohe Niu, Runsong Zhao, Xinyu Liu, Bo Jin, Yucheng Qiao, Chunliang Zhang, Jingbo Zhu, Tong Xiao Published: 2026-09-23 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.28250v1 At test time, large language models (LLMs) can encode historical information in activation memory (i.e., KV caches) and parametric memory (i.e., updated parameters). While activation memory is generally considered effective for factual recall and parametric memory for learning new tasks, their interplay remains unclear. In this work, we systematically investigate the role of memory in few-shot learning through controlled experiments. We find that activation memory is superior for recalling facts, whereas parametric memory does not consistently outperform activation memory in task learning. Moreover, our experiments show that the composite task, Conditional Arithmetic, requires the synergy of both memory types. Through neuron-level analysis, we find that the model activates distinct sets of neurons when accessing the same historical information through activation versus parametric memory. When both memory types are combined, the model recruits neurons from both sets, which is crucial for solving Conditional Arithmetic. These findings suggest that neither memory mechanism alone is sufficient for this composite task, highlighting the importance of their collaboration. --- ### hyperbolix: Hyperbolic Deep Learning in JAX Authors: Timo Klein, Thomas Lang, Yllka Velaj, Sebastian Tschiatschek Published: 2026-09-23 Categories: cs.LG, cs.MS arXiv: https://arxiv.org/abs/2609.28248v1 We present hyperbolix, an open-source library for hyperbolic deep learning in JAX, built on Flax NNX. To our knowledge, it is the first comprehensive, general-purpose hyperbolic deep learning library in JAX. It includes six manifolds with a common interface: Euclidean space, the Poincaré ball, the hyperboloid, the $κ$-stereographic model, mixed-curvature product spaces, and the proper velocity space. We implement layer families that cover linear layers, convolutions, attention, normalization, positional encoding, regression, and vector quantization. These building blocks span methods ranging from Ganea's original hyperbolic neural networks to recent fully hyperbolic architectures such as Hypformer and Lorentzian ResNet. Additionally, hyperbolix contains Riemannian optimizers implemented as optax transformations, wrapped distributions, and hyperbolic dimensionality-reduction techniques. Its API uses idiomatic JAX: Manifolds are stateless, with curvature being passed at call time, while manifold operations act on single points, with jax.vmap enabling batch operations. The precision of every checked operation is tested against a closed-form NumPy/SciPy transcription from the source paper or a finite difference, for both float32 and float64. On the hyperboloid, standard formulas for two-point operations, such as the distance, lose precision far from the origin, because they subtract two large, nearly equal terms. hyperbolix replaces these subtractions with cancellation-free formulas that stay accurate in float32 at distances where prior implementations return NaN. hyperbolix is available under the MIT license at https://github.com/timoklein/hyperbolix . --- ### Controlling Collectives of AI Agents in Reasoning Space with Spatial Transformers Authors: Frederic Vatnsdal, Roshan Gopal, Romina Garcia Camargo, Vijay Kumar, Alejandro Ribeiro Published: 2026-09-23 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.28247v1 Large Language Models (LLMs) introduce an exciting new paradigm for planning and navigation in robotics, but fail on even simple multi-robot tasks as team sizes grow. We propose COMPASS, a scalable, decentralized multi-robot architecture for controlling large collectives of agentic robots with reasoning space feedback control. Feedback is generated locally on each robot by a spatial transformer which aggregates multi-hop messages across the fleet into a learned feedback token. Our experiments find that collectives of language models demonstrate performance gains from structured diversity of the input command, which can cancel biases; an advantage that is held across scale. Compared against a centralized frontier LLM policy and a language-only communication ablation, we find that the coupled design of COMPASS decisively produces cohesive flocking formations that accurately fly the commanded intent. We show that reasoning feedback works best when composed with a compact learned token. Our ablations show that hand engineered feedback with raw state appearing in the language channel obliterates cohesion. COMPASS generalizes zero-shot to unseen instructions of ambiguous meaning while commanding flocks up to 16 times its training scale, flying up to 1024 robots under natural language commands. --- ### Beyond Poetry: Can Large Language Models Generate Classical Arabic Maqamat? Authors: AbdulRahman A. Morsy, Aya Zirikly Published: 2026-09-23 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.28245v1 Large language models (LLMs) have shown strong performance in creative text generation, yet their ability to produce culturally grounded and stylistically constrained literary forms remains underexplored. Prior work has focused largely on modern language varieties and poetry, while classical prose traditions such as maqama remain largely unstudied. The maqama is a classical literary genre characterized by rhymed prose (saj), dense rhetorical ornamentation, and episodic narrative structure, making it a challenging testbed for evaluating whether LLMs can move beyond surface fluency toward deeper literary competence. In this paper, we present the first controlled evaluation study of maqama generation with LLMs, comparing five models under zero-shot, few-shot, and rule-based prompting, and evaluating outputs through both human annotation and an LLM-as-a-judge framework across dimensions such as rhetorical richness, saj density, structural coherence, and stylistic authenticity. Our results show that prompting strategy plays a strong role in stylistic quality: few-shot prompting most consistently improves saj density, while its effects on rhetoric and coherence vary by model, with the strongest models (GPT-4o and GPT-5.4-mini) benefiting most from rule-based prompting on these dimensions, though zero-shot prompting yields the highest aggregate scores across all five models. We further observe systematic differences between models in stylistic alignment with Arabic maqama conventions, and corroborate our findings with a second independent LLM judge, paired statistical significance testing, and non-LLM proxy measures of saj. --- ### ODPure: Backdoor Purification for Object Detection via Ensemble Corruption Consensus Authors: Li Zeng, Mingcheng Duan, Longfei Fan, Hangtao Zhang, Xianlong Wang, Yanchun Li, Xia Wen, Leo Yu Zhang Published: 2026-09-23 Categories: cs.CV, cs.CR arXiv: https://arxiv.org/abs/2609.28239v1 With the development of applications like autonomous driving, object detection has gained significant attention, while also highlighting critical vulnerabilities like backdoor attacks that severely compromise model integrity. Specifically, such attacks involve altering the categories of objects (i.e., object misclassification), removing bounding boxes (i.e., object disappearance), or generating bounding box proposals for non-existent objects (i.e., object generation) when a predefined trigger is present in the input. Although backdoor defenses for image classification are well-established, the research for object detection remains comparatively underexplored. Existing defenses address these threats by scanning outputs or models for potential backdoors but require discarding either malicious data or models. This remedy fails to enable a continuous and accurate perceptual stream for the object detection pipeline. To address such limitations, we propose ODPure, a novel input-stage black-box defense for object detection, which is based on input purification that ensures stable perception flows. Tailored to the dense prediction nature of object detectors, our Corruption-Reconstruction-Selection (CRS) paradigm operates by neutralizing triggers through a diverse portfolio of corruptions to generate a massive pool of redundant proposals, then recovering fine-grained structural cues via generative priors, and finally employing voting to reach a consensus on the resulting detections. Comprehensive experiments demonstrate that our method provides robust defense against diverse backdoor attacks and trigger types while preserving baseline accuracy. Our code is available at https://github.com/Alex66366/ODPure. --- ### EmbodiedMemory-Bench: Benchmarking Embodied Memory for Long-Horizon Embodied Tasks Authors: Lizhou Liang, Xinyu Zhong, Miao Pan, Xiaohe Zhou, Xuanyu Liu, Qinfeng Li, Peng Li, Jintao Chen, Xuhong Zhang, Wenqi Zhang Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28236v1 Long-horizon embodied interaction requires agents to retain and continually update information about the environment as they observe, act, and encounter change. Yet current agents struggle to maintain such memory reliably. Our analysis traces this limitation to four key deficiencies: weak fine-grained visual memory, unreliable dynamic world-state tracking, failing to record world state revealed by interaction outcomes, and limited generalization from prior experience. However, existing benchmarks do not directly assess these memory capabilities during long-horizon embodied interaction. To address this gap, we introduce EmbodiedMemory-Bench (EMem-Bench), comprising 2,554 interactive episodes across four task families. EMem-Bench requires agents to build and update memory from interaction history, then use it to complete a later task by acting in the environment. We further present Embodied-Memorizer (EMem), an external memory system that organizes embodied experience into spatial, event, and scene memories. We also train EMem-8B, an 8B policy that manages and uses these memories. We evaluate a diverse range of open-source and proprietary MLLMs and representative multimodal memory systems. Results show that current models remain weak and uneven across the four challenges. Under matched backbones, EMem achieves the best overall performance among the evaluated memory systems and improves both open-source and proprietary models, while EMem-8B further improves over its backbone. Project page: https://zju-omniai.github.io/EmbodiedMemoryBench/ --- ### Diff-RF: Mutually Reinforced Image Registration and Fusion via Degradation-Aware Learning Authors: Xunpeng Yi, Zaixi Du, Qinglong Yan, Yibing Zhang, Han Xu, Jiayi Ma Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28235v1 Image registration and fusion aim to establish spatial correspondences from misaligned multi-modal source images, and integrate complementary information. However, in real-world imaging scenarios, source images are often affected by complex and diverse degradations, such as low illumination, noise, etc., which severely hinder the effectiveness of registration and fusion. To address this issue, we propose a mutually reinforced image registration and fusion diffusion framework via degradation-aware learning, termed Diff-RF. It explores the intrinsic coupling between registration-fusion and information restoration in the degradation conditions, enabling high-quality fusion of unregistered images under complex degradation conditions. First, the intra-modal restoration module is designed to alleviate modality-specific degradations by leveraging information within each modality, thereby providing more reliable structural representations for registration and facilitating subsequent cross-modal fusion. Second, we develop a cross-modal diffusion registration and fusion module that establishes bidirectional interaction between registration and fusion. By integrating fusion-derived visual cues and correspondence-based geometric conditions into the diffusion process, the proposed framework progressively refines spatial alignment and exploits cross-modal complementary information to achieve collaborative enhancement. Rather than treating them as independent components, degradation-aware information restoration and the collaborative optimization of registration and fusion are tightly coupled, achieving overall performance improvements. Extensive experiments on multiple extended datasets demonstrate that Diff-RF achieves superior registration accuracy and fusion quality under various degraded scenarios, exhibiting strong robustness and generalization ability. --- ### Do Center Biases Propagate? Robustness of Pathology Foundation Models in Whole-Slide Image Classification Authors: Ilán Carretero, Pablo Meseguer, Rocío del Amor, Valery Naranjo Published: 2026-09-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.28231v1 Pathology foundation models (PFMs) have transformed computational pathology through powerful representation learning from histopathological images. PFMs provide rich, discriminative representations for whole slide image (WSI) analysis, enabling tasks such as slide-level classification under multiple instance learning (MIL). However, these representations may also encode non-biological signals associated with acquisition centers, potentially introducing spurious shortcuts into downstream predictions. In this work, we evaluate center-associated robustness in WSI classification using a controlled training setting with increasing class-center correlations quantified by Cramér's V. We benchmark six PFMs across four datasets and two MIL aggregators, while evaluating ComBat as a robustification strategy. We further introduce the Area Under the Cramér's V Curve (AUCC) to jointly capture absolute classification performance and its degradation as spurious correlation increases. Results show that center-related information encoded by PFMs propagates to WSI-level predictions, with robustness depending on both the PFM representation and MIL aggregation strategy. Additionally, ComBat harmonization does not provide consistent robustness gains across datasets. --- ### A Unified Framework and Dataset for Oriented Object Visual Grounding in Remote Sensing Authors: Zeyu Ding, Yong Zhou, Jiaqi Zhao, Wen-Liang Du, Xixi Li, Hancheng Zhu, Rui Yao, Abdulmotaleb El Saddik Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28230v2 Visual grounding in remote sensing images aims to locate objects described by referring expressions. Most existing methods predict horizontal bounding boxes, which are often inaccurate for objects with arbitrary orientations. To address this limitation, we introduce O$^2$-VG, a family of models for oriented object visual grounding with three complementary designs. Specifically, O$^2$-VG-Trans is a cross-modality transformer for oriented object visual grounding. It establishes a strong discriminative foundation for the model family. Building upon it, O$^2$-VG-Uni predicts universal oriented proposals for possible foreground objects without specific text prompts. It also supports object retrieval through cached proposal embeddings. Using these universal oriented proposals as input prompts, O$^2$-VG-VLM is an autoregressive vision-language model. It generates oriented box token blocks in parallel through multi-token prediction. In addition, we construct DIOR-R-RSVG, a dataset for oriented object visual grounding in remote sensing images. It provides image, expression, and oriented box triplets for training and evaluation. Together, the O$^2$-VG family provides a flexible framework that spans discriminative transformers and generative vision-language models. It achieves superior performance across multiple benchmarks. Code is available at https://github.com/wokaikaixinxin/ai4rs and https://github.com/wokaikaixinxin/Eagle_o2_vg. --- ### A Unified Framework and Dataset for Oriented Object Visual Grounding in Remote Sensing Authors: Zeyu Ding, Yong Zhou, Jiaqi Zhao, Wen-Liang Du, Xixi Li, Hancheng Zhu, Rui Yao, Abdulmotaleb El Saddik Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28230v1 Visual grounding in remote sensing images aims to locate objects described by referring expressions. Most existing methods predict horizontal bounding boxes, which are often inaccurate for objects with arbitrary orientations. To address this limitation, we introduce O$^2$-VG, a family of models for oriented object visual grounding with three complementary designs. Specifically, O$^2$-VG-Trans is a cross-modality transformer for oriented object visual grounding. It establishes a strong discriminative foundation for the model family. Building upon it, O$^2$-VG-Uni predicts universal oriented proposals for possible foreground objects without specific text prompts. It also supports object retrieval through cached proposal embeddings. Using these universal oriented proposals as input prompts, O$^2$-VG-VLM is an autoregressive vision-language model. It generates oriented box token blocks in parallel through multi-token prediction. In addition, we construct DIOR-R-RSVG, a dataset for oriented object visual grounding in remote sensing images. It provides image, expression, and oriented box triplets for training and evaluation. Together, the O$^2$-VG family provides a flexible framework that spans discriminative transformers and generative vision-language models. It achieves superior performance across multiple benchmarks. Code is available at https://github.com/wokaikaixinxin/ai4rs. --- ### From Alignment to Fusion in 3D Vision-Language Authors: Xueqi Qiu, Xingyu Miao, Jingjing Deng, Haoran Duan, Yang Long, Ling Shao Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28222v1 Unified 3D vision-language systems must combine complementary geometry, scale, and appearance cues while supporting tasks from instance segmentation to language-guided reasoning. Existing methods often process point clouds, voxel grids, and multi-view images independently; directly combining these heterogeneous representations may leave substantial feature discrepancy unresolved, while subsequent unconstrained adaptation may distort their internal geometry. We propose an align-then-fuse framework that first applies triple pairwise cosine alignment to establish segment-level correspondence across the three representations and then retrieves task-conditioned features with a prompt-guided query decoder. Before fusion, representation-specific query features are transformed by learnable mappings constrained to the special orthogonal group. These mappings preserve inner products and Euclidean distances within each representation, permitting controlled representation-specific re-parameterisation without arbitrarily distorting its internal geometry. The transformed features are subsequently combined through Adaptive Fusion under downstream task supervision. Experiments cover eight datasets for instance segmentation, visual grounding, question answering, and dense captioning. Compared with PQ3D, the model improves average precision by 3.2 points on ScanNet200 and grounding accuracy by 2.9, 10.6, 4.6, and 4.1 points on ScanRefer, Nr3D, Sr3D, and Multi3DRefer, respectively, while also improving performance on ScanQA, SQA3D, and Scan2Cap. Ablations further support the complementary roles of alignment and orthogonal re-parameterisation and the effectiveness of Adaptive Fusion. --- ### From Agent Output to Authorized Transition Authors: Christopher Koch Published: 2026-09-23 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2609.28216v1 Agentic engineering systems can edit repositories, run tools and tests, build firmware, synthesize schematics, and prepare deployable or manufacturable artifacts. The assurance problem is therefore shifting from whether an agent can produce an output to whether an engineering lifecycle is justified in acting on claims about that output. Current products and standards provide sandboxes, approvals, hooks, traces, policy enforcement, attestations, bills of materials, and assurance representations, but these capabilities remain fragmented. This paper presents the Agile-V Assurance Spine, a cross-domain transition contract for software, firmware, and PCB engineering. Evidence is admitted only when it establishes required properties through an authoritative source profile, is bound to the exact artifact and frozen policy baseline, remains current with respect to declared dependencies, and satisfies risk-appropriate independence and authority. Gate decisions are recorded as receipts; approvals and exceptions are exact-scope and time-bounded; and authorization is rechecked at the effect boundary before merge, deployment, flashing, release, or fabrication. A bounded review of contemporary research, commercial platforms, open-source infrastructure, and standards positions the model relative to evidence-gated lifecycle control, continuous assurance, runtime admission, provenance, and AI/ML inventories. The paper contributes a precise vocabulary, compositional architecture, domain profiles, mapping to open-source implementations, and an adversarial evaluation agenda. It does not claim regulatory conformity or demonstrated production superiority. --- ### Log-Depth Recurrent Language Modeling Authors: Yiqin Wang, Nuri Cingillioglu, Charles Pert Published: 2026-09-23 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.28212v1 Language modeling using Transformers has become commonplace despite their fixed computational depth and quadratic runtime with respect to input tokens. Recurrent models on the other hand offer linear depth but no parallel execution. In this work, we extend balanced-tree recursive operators from sequence encoding to autoregressive prediction, enabling all prefix representations to be computed with logarithmic depth and linear runtime. Our experiments provide an initial characterization of this model class, demonstrating robust length extrapolation and performance approaching that of ALiBi-based Transformers, highlighting its potential as an alternative architecture for language modeling. --- ### Support-Compiled Feature Folding: More Evidence at Lower Memory Across Tabular Foundation Models Authors: Tian Zhou, Beverly Jin, Xue Wang, Linxiao Yang, Wenwei Wang, Bingqing Peng, Mengni Ye, Jinjie Gu, Liang Sun Published: 2026-09-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.28208v2 Wide tables offer tabular foundation models more evidence, but accessing it can exhaust their memory: full-width pairwise mixing grows quadratically with the number of columns, while feature selection makes inputs affordable by discarding evidence. We ask whether using more features requires interacting over all of them at once. We introduce Support-Compiled Feature Folding (SCFF), a training-free inference framework that encodes wide tables through bounded calls to a frozen backbone. SCFF organizes support-ranked features into a strong Core and a candidate Tail, folds them into narrow feature groups, and support-checks the Tail's added evidence before a single contextual prediction. This converts quadratic feature-interaction work into linear-in-width work with a bounded local working set, without ensembling predictions or training new parameters. On the exhaustive 18-dataset wide-table slice of fixed AMLB-29, TabZilla, and TabArena snapshots, SCFF improves dataset-macro accuracy and NLL on all six evaluated backbones. All four matched-width comparisons retain favorable 95% dataset-bootstrap intervals on locked folds, with relative error reductions up to 26.1%. Median paired GPU-memory savings are 2.09-2.36x, and the ratio of separately observed maximum peaks reaches 34.3x. Under a measured peak-memory ceiling, SCFF uses the saved budget to preserve more support-selected evidence, improving accuracy by 4.06 and 3.72 points over the widest feasible single leaf on predeclared wide-Core strata of TabICLv2 and TabPFN-3. --- ### Support-Compiled Feature Folding: More Evidence at Lower Memory Across Tabular Foundation Models Authors: Tian Zhou, Beverly Jin, Xue Wang, Linxiao Yang, Wenwei Wang, Bingqing Peng, Mengni Ye, Jinjie Gu, Liang Sun Published: 2026-09-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.28208v1 Tabular foundation models face a feature-side scaling dilemma: full-width pairwise mixing grows quadratically with the number of columns, whereas feature selection saves memory by discarding evidence. We introduce Support-Compiled Feature Folding (SCFF), a training-free inference framework that resolves this dilemma without changing the frozen backbone. SCFF routes support-ranked features through bounded leaves of the native feature encoder, support-checks the residual evidence, and merges the encoded messages before a single contextual prediction. It thereby converts quadratic feature-interaction work into linear-in-width work with a bounded local working set, without ensembling predictions or training new parameters. On the exhaustive 18-dataset wide-table slice of fixed AMLB-29, TabZilla, and TabArena snapshots, SCFF improves dataset-macro accuracy and NLL on all six evaluated backbones. All four matched-width comparisons retain favorable 95 percent dataset-bootstrap intervals on locked folds, with relative error reductions up to 26.1 percent. Median paired GPU-memory savings are 2.09x to 2.36x, and the ratio of separately observed maximum peaks reaches 34.3x. Under a measured peak-memory ceiling, SCFF uses the saved budget to preserve more support-selected evidence, improving accuracy by 4.06 and 3.72 points over the widest feasible single leaf on predeclared wide-Core strata of TabICLv2 and TabPFN-3. --- ### Transferable Evidence Reconstruction for Longitudinal Glucose Representations Authors: Tian Zhou, Bingqing Peng, Linxiao Yang, Wenwei Wang, Mengni Ye, Beverly Jin, Zuyi Zhu, Jinjie Gu, Liang Sun Published: 2026-09-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.28199v2 Long physiological recordings contain many routine measurements, while predictive information often lies in rare events, sustained burden, and recurring patterns. These properties can be computed as label-free evidence, but directly using them as features leaves limited labeled data to separate reproducible associations from sample-specific ones. Learning to reconstruct evidence can exploit unlabeled recordings, yet joint reconstruction does not explicitly require the decoding rule to transfer across individuals. We introduce transferable evidence reconstruction (TER): a Ridge regressor fits evidence from representations in one group and predicts it in an identity-disjoint group without refitting. The transfer error trains the encoder through the differentiable fit. For continuous glucose monitoring (CGM), clock-aware encoding preserves the multi-day content and timing needed for evidence recovery. Matched interventions connect the gains to reduced fitting-group sensitivity, with structured targets improving on raw recovery. Across ten leading CGM and time-series baselines, TER sets a new best metric on 12/14 phenotype tasks and exceeds the strongest prior overall PR-AUC/ROC-AUC/Macro-F1 by 4.95/4.43/0.66 percentage points; the PR-AUC and ROC-AUC gains are $2.6\times$ and $2.2\times$ the respective gaps between the two strongest baselines. Meal-response and future-CGM studies further demonstrate predictive utility. TER thus uses meaningful signal properties to supervise not only what a representation preserves, but how reliably it can be read across individuals. --- ### Transferable Evidence Reconstruction for Longitudinal Glucose Representations Authors: Tian Zhou, Bingqing Peng, Linxiao Yang, Wenwei Wang, Mengni Ye, Beverly Jin, Zuyi Zhu, Jinjie Gu, Liang Sun Published: 2026-09-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.28199v1 Long physiological recordings contain many routine measurements, while predictive information is often concentrated in rare events, sustained burden, and recurring temporal patterns. Masked autoencoding recovers measurements; contrastive learning aligns views. We study self-supervision that explicitly prioritizes structured signal evidence. We introduce transferable evidence reconstruction (TER), which constructs evidence from unlabeled recordings, fits a fresh low-capacity reader on one recording group, and requires that reader to recover the same evidence in another group without refitting. Differentiating through this cross-group test learns representations with transferable evidence-decoding rules; the evidence guides self-supervision but is not used as a downstream feature. For continuous glucose monitoring (CGM), an observation-aware daily encoder and clock-aware multi-day memory bind glucose level and change to recorded time while organizing up to seven days of history. On the 14-task leaderboard, TER improves the strongest prior overall PR-AUC/ROC-AUC/Macro-F1 scores by 5.51/4.43/2.80 percentage points and sets a new best metric on 12/14 tasks. These leaderboard gains are 2.0-2.9 times the respective gaps between the two strongest baselines. With public pretraining data, folds, and the linear probe matched, TER outperforms our GlucoFM reproduction by 6.09/5.52/2.72 points. Target-reader ablations, same-history controls, and cross-person readouts support the combination of structured evidence, cross-group reader fitting, and learned multi-day organization. --- ### PASTABench: Proactive Assessment of Sequential Trajectories for Agent Safety Authors: Jiapeng Sun, Yujin Zhou, Han Zhu, Pengcheng Wen, Jiayi Zhou, Sirui Han, Yike Guo Published: 2026-09-23 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.28197v1 As Large Language Models (LLMs) evolve into autonomous agents that alter real-world states, ensuring operational safety across multi-step workflows has become a critical challenge. While recent work has moved beyond single-turn evaluation toward multi-turn paradigms, key limitations persist: step-level methods treat actions in isolation, missing how risks accumulate, while trajectory-level evaluations operate post-hoc, offering no opportunity for timely intervention. To address these limitations, we formalize Decoupled Proactive Safety Monitoring along three dimensions: whether to intervene, when to intervene, and what the risk is. We introduce PASTABench, a benchmark of 1,139 multi-turn trajectories spanning 5 risk categories and 13 subcategories. We further propose the Optimal Intervention Window (OIW), anchored by annotated Earliest-Signal and Trigger turns, to quantify intervention timeliness. Evaluation of 16 LLMs reveals that proactive intervention remains largely unsolved, with the best model achieving only 40.74% optimal-timing interventions. Fine-grained diagnosis further uncovers pervasive lexical overfitting: competitive safety scores of smaller models mask keyword hypersensitivity rather than genuine risk comprehension, as their proactive capability largely collapses once hazard vocabulary is neutralized. --- ### Geospatial embeddings detect old-growth forests but buffered spatial validation narrows their advantage over Sentinel features Authors: Thomas Ratsakatika, Mihai Zotta, Srinivasan Keshav, Emily R. Lines Published: 2026-09-23 Categories: cs.LG, cs.CV, eess.IV arXiv: https://arxiv.org/abs/2609.28194v1 Old-growth forests develop over centuries under minimal anthropogenic disturbance, producing structurally complex and biodiverse stands. In Europe, protecting them requires mapping that is accurate for individual forest parcels yet deployable continent-wide. Geospatial foundation model (GFM) embeddings enable label-scarce land classification, but their value for old-growth detection remains unknown. Here, we map old-growth forests across 211,893 ha of Romania's Southern Carpathians, a beech-spruce landscape typical of the Alpine Biogeographic Region. We construct high-confidence, expert-informed reference labels for old-growth and non-old-growth parcels. We add AlphaEarth, TESSERA v2 and Sentinel-1/2 features to a common baseline of topographic and human-access predictors, then compare them under spatially blocked validation with and without 10 km train-test buffers to limit residual autocorrelation. With buffering, GFM and Sentinel-1/2 predictors increase precision-recall AUC by 0.21-0.25 (95% CIs: 0.15-0.34) relative to baseline, indicating spectral data contain a spatially robust old-growth signal. With a PR-AUC of 0.84 (0.79-0.88), TESSERA outperforms Sentinel-1/2 (+0.08 (+0.05 to +0.11)) and AlphaEarth (+0.08 (+0.04 to +0.12)) under unbuffered spatial validation. At a 10 km buffer, however, this advantage narrows to +0.04 (-0.01 to +0.11) and +0.03 (-0.04 to +0.10), intervals consistent with no difference. At 10 m resolution, convolutional neural networks add no benefit over pixel-based XGBoost. Comparisons with four national- and continental-scale products show the importance of non-old-growth labels, and reveal 81% agreement between our predictions and a field-calibrated map. We conclude that buffered spatial validation is vital when transferring old-growth detection models to unseen landscapes, and provide our labels and predictions for future work. --- ### From Change Captions to Change Detection: Semantic-Appearance Agreement Framework for Remote Sensing Change Detection Authors: Yuan Qian, Jie Ma Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28192v1 Remote sensing change detection (RSCD) is essential for monitoring land-cover changes and urban development. However, most methods demand pixel-level change masks, which are costly and time-consuming to annotate. Weakly supervised methods reduce this cost by using image-level change labels. Yet these labels indicate only whether a change occurs, leaving models to recover the location of the change and semantic meaning through additional and complex mechanisms. This missing information can be supplied directly by change captions, which describe what changes, what it becomes, and where it occurs. Therefore, we introduce change-caption-guided RSCD, using change captions as the sole task-specific supervision to learn change masks without manually annotated change masks. Our framework has two components: a caption-driven generation pipeline that produces bi-temporal remote sensing image pairs at scale with controlled changes matching each caption, and a change detector guided by the caption's transition semantics. The detector uses our Semantic-Appearance Agreement Framework (SAAF) to combine caption-grounded semantic responses with RGB differences for change localization, while text conditioning guides dense prediction. Experiments on our newly constructed Flair-RSGen dataset and WHU-CDC show that SAAF outperforms the closest reproduced limited-supervision baselines in macro-averaged IoU and F1 under the evaluated protocols. Code is publicly available at https://github.com/qianyuancs/SAAF. --- ### Two Global Crops Suffice: Locating Semantic Emergence in DINO-Style Self-Supervised Learning Authors: Basavaraj Sunagad, Artur Jesslen, Adam Kortylewski Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28187v1 Self-supervised vision transformers trained with DINO-style objectives exhibit striking emergent semantic representation quality across visual tasks, yet the mechanisms underlying this behavior remain unclear. We present a systematic empirical dissection of the DINO family and show that semantic representations arise primarily from enforcing consistency between geometrically distinct global views of the same image instance. This instance-specific global alignment acts as the semantic anchor of DINO-style learning. Across controlled retraining experiments evaluated on semantic correspondence and a diverse suite of 2D and 3D downstream tasks, we find that patch-level masking objectives enhance semantics only when trained jointly with this global alignment, indicating that the iBOT objective refines and densifies existing semantic structure rather than creating it independently. In contrast, local-to-global view alignment does not substantially improve semantic qualities at fixed compute beyond a purely global alignment. Beyond training design, we revisit how semantic representation quality should be evaluated: while classification accuracy is the standard validation score, semantic correspondence provides a complementary axis that more reliably predicts downstream task performance. Together, these findings provide a functional decomposition of DINO-style learning and represent an important step toward understanding how semantic representations emerge in self-supervised vision models. --- ### VLMs Can Describe, But Not Measure: Object-Centric Scene Understanding for Robotic Manipulation Authors: Enrico Saccon, Tommaso Faraci, Iñigo De La Ossa Zarzuelo, Luigi Palopoli, Marco Roveri, Matteo Saveriano Published: 2026-09-23 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.28184v1 Robotic operation in previously unseen environments requires both semantic understanding and reliable metric information. While vision--language models (VLMs) provide strong semantic capabilities, their geometric estimates remain less reliable. In this paper, we propose a VLM-driven, modular perception framework for scene understanding using off-the-shelf approaches. Starting from a single RGB-D observation, the scene is segmented into object-level regions, annotated by a VLM, and grounded with depth information to construct a task-independent object-centric representation. Experiments on 151 tabletop scenes show that the proposed decomposition preserves strong semantic performance while substantially improving localization and depth estimation over direct VLM inference. The resulting representation is also integrated with a task-planning framework for robotic execution. --- ### From ECG Signals to Representative-Morphology Heatmaps for Biometric Recognition Authors: Athanasios Angelakis, Marta Gomez-Barrero Published: 2026-09-23 Categories: cs.CV, eess.SP arXiv: https://arxiv.org/abs/2609.28183v1 Electrocardiography (ECG) contains subject-specific morphology that supports biometric recognition, yet image-based performance depends on how the waveform is rendered. We introduce representative-morphology heatmaps, a deterministic ECG-to-image representation adapted from ECGXtractor. Within each block of ten aligned beats, the five beats closest to the block mean are averaged into a 400 by L matrix and rendered either as a conventional trace or as a dense cardiac-time-by-lead heatmap. Since both representations contain identical physiological samples, their comparison isolates the effect of rendering. We evaluate verification and closed-set identification on PTB, ECG-ID, and MIMIC-IV-ECG-DEMO. Five compact models, including ZACH-ViT, are trained from scratch, while six ImageNet-pretrained CNN and transformer backbones assess model scale and visual transfer. Heatmaps improve both FNMR operating points and both identification ranks in all 15 compact model-dataset comparisons, while EER improves in 14. Across the matched experiments, EER decreases by 9.59 percentage points and Rank-1 increases by 24.69 points on average. ConvNeXt-Tiny reaches 2.43% EER on PTB and 5.79% on ECG-ID, whereas DeiT-Base reaches 14.92% on MIMIC-DEMO. ImageNet initialization clearly benefits the two multilead datasets but has a mixed effect on ECG-ID, and performance does not increase monotonically with model size. The best heatmap systems approach the strongest signal-domain EER on PTB and ECG-ID, while DeiT-Base provides the strongest evaluated performance on MIMIC-DEMO. Lead-channel ablation further shows that useful channel combinations depend on the cohort and biometric task. Overall, representative-morphology heatmaps provide an effective image representation for ECG verification and identification. --- ### Finite-Sample Probabilistic Safety Certification for AI-Based Grid-Edge Coordination Authors: Yihong Zhou, Hanbin Yang, Thomas Morstyn Published: 2026-09-23 Categories: cs.AI, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2609.28182v1 Coordinating large population of flexible grid-edge devices can alleviate the need for time-consuming and capital-intensive network upgrades, and AI-based control methods such as multi-agent reinforcement learning or imitation learning are promising in their real-time decision scalability. However, system operators still need an independent and rigorous way to decide whether a given AI system is safe enough for deployment. This paper develops a finite-sample probabilistic safety certification framework for black-box AI decision models in closed-loop grid operation. The central idea is to reduce the complete input--AI--grid evaluator workflow to a binary unsafe outcome under an operator-defined safety specification, and then use exact binomial inference to certify the corresponding unsafe operation probability. Given a set of held-out calibration scenarios, the framework returns the tightest one-sided upper certificate and an accept/reject deployment criterion that controls the probability of false safety certification. Because the certification is for the calibration distribution that may deviate from the future operation, we further combine the nominal certificate with physically interpretable sample-space adversarial attacks, a concept widely used in AI to investigate the fragility of AI models. Case studies on grid-edge flexibility coordination with 1{,}000-agent AI models (independent parameters) verify the finite-sample safety guarantee and the value of integrating adversarial attacks into a rolling-window training-certification-deployment flow. --- ### How Sensitive Are LLM Leaderboard Claims to Hidden Model Selection? Authors: Chen Yang, Jun Chen Published: 2026-09-23 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.28177v2 LLM leaderboard gains can reflect selection among privately evaluated model variants, yet neither the number of variants nor their dependence is public. We ask how many hidden variants a published margin can support while retaining statistical evidence of a provider's advantage over a fixed comparator. For a fixed candidate family under a Gaussian margin model, we derive a sensitivity curve that reports this maximum count as a function of a lower bound on within-family correlation. The relevant correlation must match the score used for ranking and the sampling model: in a controlled family, pooled item correlation is 0.90, whereas composite-score correlation is 0.46 under item resampling and 0.92 when MMLU subjects are resampled. An item-based audit of 394 adjacent-rank claims on the Open LLM Leaderboard finds that 391 lack statistical support even before accounting for selection. Among claims that pass the uncorrected test, certification can depend on assumptions about the hidden family's correlation. The resulting curves make these assumptions explicit without estimating the unobserved search size. --- ### How Sensitive Are LLM Leaderboard Claims to Hidden Model Selection? Authors: Chen Yang, Xianyang Zhang, Jun Chen Published: 2026-09-23 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.28177v1 LLM leaderboard gains can reflect selection among privately evaluated model variants, yet neither the number of variants nor their dependence is public. We ask how many hidden variants a published margin can support while retaining statistical evidence of a provider's advantage over a fixed comparator. For a fixed candidate family under a Gaussian margin model, we derive a sensitivity curve that reports this maximum count as a function of a lower bound on within-family correlation. The relevant correlation must match the score used for ranking and the sampling model: in a controlled family, pooled item correlation is 0.90, whereas composite-score correlation is 0.46 under item resampling and 0.92 when MMLU subjects are resampled. An item-based audit of 394 adjacent-rank claims on the Open LLM Leaderboard finds that 391 lack statistical support even before accounting for selection. Among claims that pass the uncorrected test, certification can depend on assumptions about the hidden family's correlation. The resulting curves make these assumptions explicit without estimating the unobserved search size. --- ### Confidence Falls Short: Asymmetric Certainty Gains from Optimization Hinder Multimodal Classification Authors: Longfei Huang, Xiangyu Wu, Yang Yang Published: 2026-09-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.28165v1 Multimodal learning (MML) falls into the optimization dilemma due to the modality imbalance phenomenon, leading to suboptimal overall performance in practice. While many attempts primarily focus on balancing the optimization dynamics across modalities to address this issue, we identify a subtle yet critical flaw: optimization yields asymmetric gains in predictive certainty, with the strong modality more confident than the weak one, driving imbalanced modality contributions. In this paper, our analysis reveals that this flaw stems from unimodal characteristics rather than multimodal learning, and this confidence discrepancy can be corrected by positive cross-modal intervention. Based on this insight, we propose multimodal Max Confidence Regularization (MaxCR) to dynamically intervene in modality semantic confidence. Specifically, the semantic confidence of each modality is tracked using a nonlinear sparsity measure. We then design max suppression and max excitation based on this measure to regularize strong and weak modalities, respectively. They penalize and encourage the top-1 confidence, thereby constraining multimodal prediction. To this end, strong and weak modalities are expected to make calibrated confidence, thereby improving the overall performance. Empirical experiments on widely used datasets reveal the superiority of our method through comparison with various state-of-the-art (SOTA) multimodal learning baselines. --- ### Rank-One Signal Recovery in Sparse Wishart Noise Authors: Preben Forer, Urte Adomaityte, Pierpaolo Vivo Published: 2026-09-23 Categories: cond-mat.dis-nn, cond-mat.stat-mech, math-ph, stat.ML arXiv: https://arxiv.org/abs/2609.28163v1 We study the high-dimensional recovery of a signal vector $\mathbf{x}$ in the presence of sparse Wishart-like noise. We define an $N \times N$ matrix $A = J+(θ/N)\mathbf{xx}^{\top}$, where $\mathbf{xx}^{\top}$ is the rank-one deformation of the random noise matrix $J$. We consider a Wishart-like matrix $J={X}^{\top} X$, where $X$ is a sparse $M \times N$ random matrix with entries $X_{ij} = c_{ij}W_{ij}$, with $c_{ij}$ regulating the density of non-zero elements, and $W_{ij}$ the bond weights. Using the replica method, we compute analytically the top eigenpair statistics of $A$, and their dependence on the signal strength $θ$, the rectangularity ratio $α=\sqrt{M/N}$, and the average connectivity of the noise. The spectral observables are expressed in terms of a system of Recursive Distributional Equations, which are efficiently solved via a Population Dynamics algorithm. They allow us to compute the average largest eigenvalue $\langleλ_1\rangle_{A}$, the average top eigenvector component density, and the average overlap between the top eigenvector of $A$ and $\mathbf{x}$. We identify a critical threshold $θ_{\mathrm{crit}}$--depending on the average connectivity of the noise--that marks a BBP-like phase transition: below this value, $\langleλ_1\rangle_{A}$ is unaffected by the signal, and the overlap vanishes. Thus, the signal is not recoverable from the top eigenvector of $A$. For $θ>θ_{\mathrm{crit}}$, the signal-related outlier eigenvalue becomes $\langleλ_1\rangle_{A}$ and the overlap is nonzero, allowing for recovery of the signal. The results are in excellent agreement with numerical diagonalisation. We show that in the dense limit, the recovery threshold and eigen-statistics converge to the results predicted by the classical BBP transition for additive rank-one deformations of dense Wishart matrices. --- ### Depth-Guided Contrastive Learning for 2D Representations with 3D Spatial Awareness Authors: Liang Zeng, Maarten Vergauwen Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28159v1 Standard contrastive learning frameworks are mainly designed from a semantic perspective, yet learning 2D visual representations that preserve 3D spatial structure is also important for scene understanding. In this work, we propose Depth-Guided Contrastive Learning (DGCL), a simple auxiliary objective that injects 3D spatial awareness into 2D contrastive representation learning. Our key idea is to use depth to convert local 3D proximity into contrastive similarity: pixels that are closer in 3D space are encouraged to have more similar representations than pixels that are farther apart. Instead of relying on absolute depth values, DGCL formulates supervision through relative 3D distance comparisons among randomly sampled pixels, making the objective invariant to depth scale, efficient to compute, and easy to integrate into existing contrastive frameworks. Experiments across different datasets and models show that DGCL consistently improves 2D representation learning and benefits semantic downstream tasks by stronger spatial and geometric understanding. The code is available on https://github.com/LeungTsang/DGCL. --- ### A comparative assessment of global building and settlement datasets across geographic and settlement contexts Authors: Rufai Omowunmi Balogun, Caroline Margaux Gevaert, Capucine Riom, Derrick Mirindi, Aaron Opdyke, Hamed Alemohammad, Pierre Chrzanowski, Edward Charles Anderson Published: 2026-09-23 Categories: cs.CV, eess.IV arXiv: https://arxiv.org/abs/2609.28154v1 Global building and settlement datasets increasingly support population mapping, exposure assessment, urban monitoring, and other analyses of the built environment, yet comparative evidence remains fragmented across products, geographic regions, reference datasets, spatial scales, and evaluation methods. We benchmark seven global or near-global products, including Overture Maps, Global Building Atlas, 3D-GloBFP, Google Open Buildings 2.5D Temporal (OBT), Microsoft TEMPO, GHSL, and WSF Tracker, against harmonized reference footprints across 135 study areas. The evaluation combines complementary measures of detection, geometric agreement, and aggregate quantity accuracy, together with stratified analyses of settlement characteristics and diagnostic experiments on error size and temporal alignment. Overture achieved the highest median city-level vector F1 (0.786). Raster rankings were resolution-dependent: OBT achieved the highest median F1 at 10m (0.642), whereas WSF Tracker led at 100m (0.862). However, WSF Tracker substantially overestimated built-up area, emphasizing that when using raster products, it is important for the user to understand whether the raster identifies only buildings or includes additional impervious surfaces. Raster accuracy increased consistently with building density (Spearman \r{ho} = 0.58-0.75), while small candidate buildings were disproportionately associated with false positives in the vector products. Temporally aligning WSF Tracker with reference imagery increased mean F1 by 0.060 (median +0.037), indicating that the reported accuracies are conservative in rapidly growing areas. The study establishes a reproducible benchmark for comparing heterogeneous global urban and settlement layer datasets across geographic and settlement contexts. --- ### Exact Feedback Is Not Control: Evaluating Text-based Closed-Loop Revision in LLMs Authors: Haitong Jiang, Chunlin Liu, Yile Wang, Yuhong Feng Published: 2026-09-23 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.28150v1 Closed-loop revision is increasingly used in large language model (LLM) applications, but failures may reflect incomplete feedback or ineffective responses to correct feedback. We introduce a fixed-budget revision protocol with deterministic verifiers that report all remaining violations across exact-length, lexical, and compositional constraints. Fixing feedback correctness and completeness isolates model-side revision behavior. Across 19 open- and closed-source models, controller-level mean final joint success ranges from 17.4% to 99.8%, with substantial cross-model gaps persisting under identical initial drafts. Controlled experiments reveal reproducible model-specific responses to exact feedback. Post-training and scale reshape these responses without consistently bringing them closer to exact correction. Across all constraint families, failed trajectories often repeat earlier outputs, and prior recurrence is associated with lower subsequent recoverability. Matched-state interventions show that removing earlier dialogue while holding the current draft and feedback fixed changes recurrence escape without reliably improving final success; effects depend on the model, task, and trigger-state composition. Exact feedback makes revision errors observable, but does not make the closed loop reliable. Code and reproduction instructions: https://github.com/kevinjiang0121-cyber/exact-feedback-code. --- ### EvEMTBench: An Open Benchmark for Machine Learning in Power System Protection Authors: Julian Oelhaf, Georg Kordowich, Christian Bergler, Andreas Maier, Johann Jäger, Siming Bayer Published: 2026-09-23 Categories: eess.SY, cs.LG arXiv: https://arxiv.org/abs/2609.28149v1 Studies of machine-learning-based power system protection are difficult to compare because task definitions, measurement access, data partitions, metrics, and generalization conditions often differ. EvEMTBench addresses this gap with an open, executable, and versioned benchmark that fixes these evaluation choices while leaving model design open. Across four grids spanning 20-345 kV, it defines 12 protection and event-analysis functions instantiated as 24 scored tasks and supports structured evaluation across observability conditions, predefined distribution shifts, and zero-shot and fine-tuned cross-grid transfer. Committed partitions, leakage controls, and reproducible reporting provide a common basis for comparing future methods. A reference evaluation spanning trivial, conventional, feature-based, and deep-learning baselines shows that wider observability is not uniformly beneficial, shifted conditions can reveal failures not apparent in-distribution, and cross-grid transfer is substantially stronger for fault detection than for fault localization. Protection-relevant diagnostics identify failure modes not apparent from primary metrics alone. EvEMTBench therefore makes generalization in machine-learning-based protection an explicit and reproducible evaluation problem. --- ### RL Starts before RL: On Policy Distillation for Better Reinforcement Learning Authors: Shuai Dong, Yongfu Zhu, Yuqi Xu, Weichu Xie, Liuwenpu, Ziyue Wang, Kaiwen Tuo, Congcong Wang, Siyuan Wang, Wenqi Shao, Shuai Yang, Ji Zhao et al. Published: 2026-09-23 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.28145v1 Reinforcement learning (RL) improves reasoning, but its performance depends on the policy from which training begins. We study on-policy distillation (OPD) as a preparation stage for RL and ask whether its benefits extend beyond improvements in the distilled model's initial accuracy. Under shared RL settings, students initialized with OPD reach higher final performance than those trained with direct RL or supervised fine-tuning followed by RL. This advantage can emerge even when OPD produces little immediate improvement in accuracy. Pre-RL Pass@k does not fully explain the benefit: similar or even higher values do not necessarily lead to better performance after RL. Behavioral analyses point to alignment with the teacher's distribution beyond top-1 agreement as a possible explanation. Such alignment may favor higher-quality reasoning paths while retaining alternatives that RL can further refine using outcome feedback. We further examine how trajectory sources and divergence objectives affect the value of distillation for subsequent RL. Standard reverse-KL OPD performs better before RL, but forward-KL OPD overtakes it afterward; with teacher-generated distillation trajectories, reverse KL remains ahead at both stages. These findings suggest that the preferred distillation objective depends on both the trajectory source and the training that follows. Our results support evaluating OPD as preparation for RL and selecting distillation choices by the performance achieved after subsequent training. --- ### "We'll Fix It Later": Education, AI, and the Deferral of Privacy in EdTech Authors: Meghna Manoj Nair, Rachel Greenstadt Published: 2026-09-23 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2609.28137v1 Educational technology (EdTech) platforms collect highly sensitive student data, including behavioral logs, disability records, and academic histories. However, privacy considerations are often postponed rather than treated as a foundational design requirement. We present a mixed-methods study combining 12 semi-structured interviews with EdTech professionals and a privacy policy audit of 48 platforms coded across five dimensions, with strong inter-rater reliability (mean Cohen's Kappa = 0.781). Our interviews reveal a recurring organizational pattern in which privacy is recognized as important but deferred across the product lifecycle as organizations prioritize product functionality, growth, funding, and immediate educational outcomes. Responsibility is often delegated to cloud providers, policy documents, or downstream institutions, while limited privacy-related feedback gives organizations little pressure to change these practices. The policy analysis reflects these patterns: platforms describe what data they collect relatively well but provide substantially less information about how that data is subsequently governed. Thirty-three percent make no meaningful Artificial Intelligence (AI) disclosure despite visible AI features, and 73% provide only generic accountability and breach-response language. K-12 platforms perform better on children's consent where regulation creates explicit requirements, but this advantage does not extend to AI governance or accountability. These findings suggest that meaningful improvement requires enforceable institutional and regulatory mechanisms rather than voluntary privacy commitments alone. --- ### Scenario-Driven Neuroevolution: Using Models to Guide Test Generation for Games Authors: Gijs van Cuyck, Patric Feldmeier, Jan Tretmans, Gordon Fraser Published: 2026-09-23 Categories: cs.SE, cs.NE arXiv: https://arxiv.org/abs/2609.28130v1 Automatically generating test inputs for games is challenging, as test generators must master the game to reach advanced program states while also ensuring robustness against the heavy program randomisation inherent to games. The test generator Neatest therefore optimises test suites consisting of neural networks that reach advanced program states and are robust to program randomisation, as they generate test inputs dynamically based on the current program state. Neatest is a white-box testing approach that aims to generate a network agent for each yet-uncovered statement or branch of the code using neuroevolution. Due to this iterative test generation approach, the algorithm does not scale well to larger programs that may contain thousands of branches. Furthermore, covering every statement or branch in a game often does not correspond to playing the game as intended. To alleviate these shortcomings, we propose combining Neatest with a model-based testing approach that allows game testers to define test scenarios via abstract game models. The test generator then no longer optimises networks to reach all branches or statements of a program, but instead trains networks to replicate the concrete desired testing behaviour expressed by the abstract game model. An evaluation on 13 Scratch games across varying genres demonstrates that Neatest, combined with model-based testing, is able to optimise agents that replicate the desired gameplay behaviour defined in the game models while increasing achieved branch coverage by 7% compared to the traditional code-guided Neatest approach. --- ### NPBoost: Neural Processes with Gradient-Boosted Fixed Effects Authors: Andrea Nava, Ken Rölli, Armin Begic, Fabio Sigrist Published: 2026-09-23 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.28122v1 Neural Processes (NPs) are model-based meta-learners that implicitly learn a stochastic process and adapt to a new task from a small context set. Most extensions of NPs focus on improving the neural network architecture. We instead develop an extension motivated by the shared hierarchical interpretation of meta-learning and mixed-effects models. Specifically, we introduce Neural Process Boosting (NPBoost), which decomposes structured response variability into tree-boosted fixed effects shared across tasks and NP random effects that capture stochastic task-to-task variation. We propose to train the two components jointly using a boosting algorithm in which an NP learns residual task-specific structure and a tree ensemble estimates common patterns across tasks. Across synthetic and real-world tabular meta-learning problems, this decomposition improves over a standard NP when the shared structure contains discontinuities or other irregular patterns that boosted trees can represent effectively. --- ### Scaling Attention Head Analysis via Gradient-Based Attribution in Context-Aware Machine Translation Authors: Paweł Mąka, Yusuf Can Semerci, Jan Scholtes, Gerasimos Spanakis Published: 2026-09-23 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.28117v1 In this paper, we introduce a gradient-based head attribution strategy where the Token-level Max-Margin loss is backpropagated to the attention maps. This framework enables a large-scale causal analysis of attention heads, making it suitable for LLMs. We evaluate our method on the task of disambiguation in Context-aware Machine Translation, where we analyze 50 phenomena across 4 models and 4 language directions. We empirically show the alignment of our method with the effects of increasing the attention scores of token-to-token relations on three models and two language directions, ensuring the robustness of our method. Our analysis reveals the presence of the "general-purpose" attention heads that improve the model's performance when attending to different relations. We find that the average attention a head assigns to a relation does not necessarily relate to the model's performance, which suggests that the models developed redundancies during training in terms of the head functions. --- ### Probabilistic and Geometry Aware Neural Surrogate of Scrape Off Layer Plasma Simulations Authors: Gabriele Gianuzzo, Stefan Dasbach, Fleur Hendriks, Sven Wiesen, Vlado Menkovski Published: 2026-09-23 Categories: cs.LG, physics.plasm-ph arXiv: https://arxiv.org/abs/2609.28116v1 Fast surrogates for tokamak boundary-plasma simulation are typically deterministic regressors mapping a global operating point to a flattened vector of cell values. Near the divertor detachment transition the steady state is not reliably single-valued. A point estimate must average over qualitatively different plasma states, and it arrives with no statement of confidence. Moreover, the flattened vector representation discards the geometric structure of the SOLPS-ITER mesh. This work addresses both problems. We unroll the curvilinear mesh into three fixed-size image tensors whose layout preserves cell adjacency and inverts exactly, letting a convolutional network act on the geometry without loss of information. A conditional flow matching model, well suited to highly sensitive systems, is then trained on this representation. The result is an efficient, scalable surrogate that captures multiple plausible outcomes even at sensitive operating points. Along a gas-puff scan, the predictive distribution splits into a hot and a cold mode across an early regime transition. A further check on synthetic data with an injected bifurcation of known size confirms the model recovers both branches rather than their average. --- ### Field-of-View Extension in Dental Cone-Beam CT via Implicit Neural Representations and Diffusion Model-Based Refinement Authors: Susanne Schaub, Florentin Bieder, Matheus L. Oliveira, Yulan Wang, Buyanbileg Sodnom-ish, Dorothea Dagassan-Berndt, Michael M. Bornstein, Philippe C. Cattin Published: 2026-09-23 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.28110v1 Dental cone-beam computed tomography (CBCT) systems often employ detector configurations that provide a truncated field of view (FOV) that only captures a small part of the patient's anatomy. In this work, we aim to reconstruct an extended FOV using projections of truncated FOV scans. To this end, we propose a three-stage framework that consists of (1) an implicit neural representation (INR) for estimating missing parts of the truncated projection data, (2) an iterative reconstruction for generating a secondary volumetric image with improved anatomical consistency and (3) a fast diffusion model for image enhancement. The proposed approach combines the strengths of continuous representations, physics-based reconstruction and generative refinement within a unified pipeline for truncated CBCT imaging. Experimental results demonstrate that the method effectively reduces truncation artifacts, improves the reconstruction of structures extending beyond the original FOV and produces images with enhanced quality. Our code is publicly available at https://github.com/SusanneSchaub/CBCT-FOV-Extension. --- ### Distillation for Efficient Multitask Manipulation Policies via Conditional Flow Matching Authors: Shreya Deshmukh, Imen Mahdi, Nick Heppert, Abhinav Valada Published: 2026-09-23 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.28107v2 Advances in generative modeling have recently been extensively employed in robotics for policy learning. In particular, Conditional Flow Matching (CFM) trained with expert demonstrations has been shown to outperform existing methods on robot manipulation benchmarks. While prior work has mainly focused on single-task settings, we study the problem from a multi-task perspective, as training independent models for each task is computationally expensive. Multi-Task policy learning comes with its own set of challenges, as naively training on a concatenated dataset of demonstrations would either require increased model capacity to accommodate the added complexity or result in drops in performance. We propose to distill knowledge from single-task CFM experts into a shared multi-task policy by transferring their learned velocity fields. We combine this distillation signal with the original CFM objective to retain fidelity to the demonstrations. Experiments on RLBench show that our approach improves multi-task policy performance over naive training while maintaining a fixed model size. --- ### Distillation for Efficient Multitask Manipulation Policies via Conditional Flow Matching Authors: Shreya Deshmukh, Imen Mahdi, Nick Heppert, Abhinav Valada Published: 2026-09-23 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.28107v1 Advances in generative modeling have recently been extensively employed in robotics for policy learning. In particular, Conditional Flow Matching (CFM) trained with expert demonstrations has been shown to outperform existing methods on robot manipulation benchmarks. While prior work has mainly focused on single-task settings, we study the problem from a multi-task perspective, as training independent models for each task is computationally expensive. Multi-Task policy learning comes with its own set of challenges, as naively training on a concatenated dataset of demonstrations would either require increased model capacity to accommodate the added complexity or result in drops in performance. We propose to distill knowledge from single-task CFM experts into a shared multi-task policy by transferring their learned velocity fields. We combine this distillation signal with the original CFM objective to retain fidelity to the demonstrations. Experiments on RLBench show that our approach improves multi-task policy performance over naive training while maintaining a fixed model size. --- ### Fed-ReMasker: Federated Tabular Imputation under Feature-Level Missingness Authors: Ioannis Papathanail, Rooholla Poursoleymani, Lubnaa Abdur Rahman, Stavroula Georgia Mougiakakou Published: 2026-09-23 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.28105v1 Multi-center clinical studies and biomedical research collaborations increasingly seek to utilize data across centers to build models that generalize beyond any single center. This creates two distinct challenges: data protection regulations may restrict the sharing of raw patient data across institutions, while centers may collect only partially overlapping sets of features under different protocols. Federated learning enables collaborative model training without centralizing raw data. However, existing federated imputation methods rarely evaluate feature-level missingness, in which entire features are unobserved at some centers. To address this setting, we adapt the ReMasker masked autoencoder to federated learning (Fed-ReMasker), enabling centers to impute features never observed locally by leveraging knowledge learned across collaborating centers. We evaluate Fed-ReMasker in a benchmark spanning synthetic datasets with linear and nonlinear relationships and real-world tabular datasets, including clinical data. The benchmark varies the number of centers, the missingness ratios, and client heterogeneity. Fed-ReMasker achieves the lowest imputation error in 93.2% of value-level and 96.7% of feature-level scenarios in the homogeneous benchmark. It also remains robust to client heterogeneity using simple federated averaging, outperforming all baselines in all 36 value-level scenarios and each baseline in at least 35 of 36 feature-level scenarios, and comes within 3.0% on average of a centralized model trained on the pooled data. --- ### Visual Tripwires: Anticipating Failure in Deep Vision Systems Authors: Anoushka Harit, Rehan Zuberi, William Prew, Florian Markowetz Published: 2026-09-23 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.28099v1 Deep vision systems remain vulnerable to corruption, occlusion, and distribution shift despite strong benchmark performance. Existing reliability methods typically evaluate uncertainty at individual time steps and do not explicitly model how a system progresses toward failure. We introduce Visual Tripwires, a predictive reliability framework that uses temporal instability in model behaviour to anticipate impending failure. Our central hypothesis is that predictive degradation develops progressively through measurable changes in latent representations, prediction trajectories, and attention structure. Visual Tripwires captures these changes using representation drift, prediction oscillation, trajectory curvature, and attention entropy. A lightweight tripwire predictor aggregates these signals over a temporal window to estimate the probability of failure within a future prediction horizon. Experiments across multiple datasets, architectures, and progressive perturbation settings show that the proposed instability signals emerge before predictive degradation and provide earlier and more accurate failure warnings than conventional uncertainty estimation methods. These results demonstrate that temporal instability contains useful information about future model reliability and provides a practical basis for early warning in deep vision systems. --- ### MotionSpec: Spectral Trajectory Supervision for Motion-Consistent Video Generation Authors: Ziqi Ni, Rui Li, Shiqi Jiang, Wei Zhou Published: 2026-09-23 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.28095v1 Recent advances in text-to-video generation have enabled high-fidelity visual synthesis, yet realistic motion remains challenging. Generated videos may exhibit temporal discontinuities, inconsistent action progression, and structural distortions during complex movements. Even when individual frames appear realistic, the underlying motion may evolve in inconsistent or implausible ways. Standard generative objectives provide limited motion-specific supervision, leaving motion evolution insufficiently constrained. In this paper, we propose MotionSpec, a motion supervision framework centered on Spectral Trajectory Consistency (STC). STC constructs dense anchor-relative motion trajectories and transforms them into motion spectral volumes via a temporal Fourier transform. By aligning the spectral amplitude and phase of predicted and target trajectories, STC constrains both motion strength across temporal frequencies and the temporal organization of motion. To complement this trajectory-level supervision, we introduce Local Flow Consistency (LFC), which aligns consecutive-frame optical flow between predicted and target videos to stabilize local motion transitions. Experiments demonstrate that MotionSpec consistently improves motion consistency, temporal coherence, and plausibility while preserving visual fidelity. --- ### Can LLMs Catch a Rigged Backtest? A Clean-Control Calibration Benchmark Authors: Makar Ulesov, Vladislav Smirnov, Omar Ibrahim, Arsenii Bobovnikov Published: 2026-09-23 Categories: cs.CL, cs.AI, cs.CE, cs.SE arXiv: https://arxiv.org/abs/2609.28090v1 Backtest auditing is a calibration problem: high flaw recall is not useful when the model falsely flags matched clean strategies. We build a 96-item paired benchmark in which every flawed backtest has a clean control that holds strategy, dates, code style, labels, and reporting scaffold fixed while changing one methodology detail. A deterministic scorer separates flaw recall, clean-control false positives, evidence localization, and fix relevance. Over 1440 cached audits from four text endpoints, the primary DeepSeek auditor reaches 100.0\% closed and clean-aware code recall, but open prompts over-flag 93.8\% of clean code controls, and clean-aware all-three specificity is 87.5\% even where recall saturates. A clean-aware warning drops DeepSeek code false positives from 20.8\% (95\% CI 11.7--34.3) to 0.0\% (0.0--7.4) at unchanged recall, while the budget anchor still flags 38/48 clean controls under the same prompt. Reporting recall alone would rank three of these four models identically; reporting the clean-control rate separates them by 79 points. --- ### Discovery of fully efficient fault indicators along a data-based diagnosis process Authors: Igor Bezmaternykh, Louise Travé-Massuyès, Elodie Chanthery Published: 2026-09-23 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.28087v1 The integration of model-based and data-driven paradigms provides a powerful framework for fault diagnosis by combining the interpretability of analytical redundancy relations, i.e., input-output relations that are used as diagnosis indicators in model-based diagnosis, with the adaptability of learning techniques. DT4X is a recent diagnosis algorithm that uses symbolic regression to generate multivariate relations leveraging some properties of analytical redundancy relations and uses them as split functions in a decision tree. However, its symbolic regression procedure optimizes only the separation between two selected classes at each node, often fragmenting the remaining classes and degrading both interpretability and diagnosis performance. This paper introduces DT4X+, an enhanced version of DT4X that modifies the construction of training sets and the symbolic-regression loss so that expressions separate the target classes while preserving the coherence of non-target classes. The resulting relations become fully consistent with ARR properties and lead to more informative splits, improved robustness, and better performance on dynamic-system datasets. Experiments conducted on several benchmark systems demonstrate the benefits of this enhanced formulation. --- ### LAYERSCOPE: A Layerwise Characterization of Video and Multimodal Learned Representations Authors: Sandra Arcos-Holzinger, Debashish Chakraborty, Rohita Mocharla, Will Walden, Andrew Yates, Reno Kriz, Sarah M. Erfani, James Bailey, Vishal M. Patel, Sanjeev Khudanpur Published: 2026-09-23 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2609.28086v2 We propose LAYERSCOPE, a label-free, layerwise framework that aims to characterize a model's learned representations in video and multimodal settings. Evaluating downstream performance using representations from final or intermediate layers typically requires large amounts of labeled data, repeated task-specific evaluations, and substantial computation. To address these limitations, LAYERSCOPE uses local, global, distributional, and correspondence-based geometric metrics to compare layerwise representation structure within and across models without requiring task-specific labels. We evaluate seven architecturally diverse models across video and multimodal classification, clustering, and text-to-video retrieval tasks from MVEB/MVEB+. We find that intermediate-layer representations can outperform final-layer and model-default outputs. We also find that no single geometric metric consistently predicts downstream performance, but note that distinct layerwise geometric signatures emerge across model families. LID shows task-dependent relationships with performance, while RankMe provides the strongest measure for classification and clustering, but is not a universal layer selector. We also find that pairing-aware metrics explain retrieval better than distributional distances alone. LAYERSCOPE therefore offers a framework for comparing representations across models and layers, enabling a more systematic evaluation in video and multimodal settings. --- ### LAYERSCOPE: A Layerwise Characterization of Video and Multimodal Learned Representations Authors: Sandra Arcos-Holzinger, Debashish Chakraborty, Rohita Mocharla, Will Walden, Andrew Yates, Reno Kriz, Sarah M. Erfani, James Bailey, Vishal M. Patel, Sanjeev Khudanpur Published: 2026-09-23 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2609.28086v1 We propose LAYERSCOPE, a label-free, layerwise framework that aims to characterize a model's learned representations in video and multimodal settings. Evaluating downstream performance using representations from final or intermediate layers typically requires large amounts of labeled data, repeated task-specific evaluations, and substantial computation. To address these limitations, LAYERSCOPE uses local, global, distributional, and correspondence-based geometric metrics to compare layerwise representation structure within and across models without requiring task-specific labels. We evaluate seven architecturally diverse models across video and multimodal classification, clustering, and text-to-video retrieval tasks from MVEB/MVEB+. We find that intermediate-layer representations can outperform final-layer and model-default outputs. We also find that no single geometric metric consistently predicts downstream performance, but note that distinct layerwise geometric signatures emerge across model families. LID shows task-dependent relationships with performance, while RankMe provides the strongest measure for classification and clustering, but is not a universal layer selector. We also find that pairing-aware metrics explain retrieval better than distributional distances alone. LAYERSCOPE therefore offers a framework for comparing representations across models and layers, enabling a more systematic evaluation in video and multimodal settings. ---