# Research Papers — 2026-10 · bundle page 146 of 156 > Full-text markdown of papers #7251–7300 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/146 [← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/145.md) [Next bundle →](https://aipod.dev/papers/archive/2026-10/page/147.md) ### HiPhy: Hierarchical Alignment for Physically-Plausible Multi-Principle Video Generation Authors: Tahira Kazimi, Shubhankar Borse, Munawar Hayat, Fatih Porikli, Pinar Yanardag Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02197v1 Video generation models have achieved remarkable visual fidelity and have strong potential to become general-purpose world simulators. Despite this progress, they still fail to generate videos which adhere to laws of physics. The problem becomes even more apparent in realistic settings where multiple physical principles must work together within the same video; for example, "a balloon floating upward while steam rises from a pot" requires buoyancy and fluid dynamics to unfold coherently and simultaneously. Yet existing methods largely ignore multi-principle interactions, focusing on a single principle per video. We propose HiPhy (Hierarchical Physical Alignment), a reinforcement learning framework that grounds video generation in physical laws through a dual-level objective: locally enforcing the temporal dynamics of individual physical principles, and globally ensuring the physical and semantic coherence of the entire scene. To support multi-principle generation, we construct a 50K-prompt dataset and introduce a prompt benchmark MultiPhyBench, spanning a diverse range of co-occurring physical events. Our experiments show that HiPhy significantly outperforms prior methods and baselines, improving physical commonsense and semantic alignment significantly across various benchmarks, with the largest gains on scenes involving multiple concurrent physical principles where competing methods degrade most sharply. --- ### InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation Authors: Zhuo Lin, Sirui Xu, Liuyu Bian, Yu-Xiong Wang, Liang-Yan Gui Published: 2026-10-01 Categories: cs.RO, cs.CV, cs.GR arXiv: https://arxiv.org/abs/2610.02196v1 We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, without retraining. Our key insight is that a broad controller already holds much of the competence a new task needs, and that this competence becomes accessible through an interface between planning and control that is expressive enough to specify contact-rich, multi-stage interactions, yet executable and measurable enough that execution feedback can guide planning from experience. InterEvolve realizes this interface with two components. First, we develop an object-aware forward-backward (FB) behavioral foundation model, whose object residuals on a frozen body prior turn a new reward about the body or objects into loco-manipulation behavior at test time. Second, we specify tasks as reward programs: staged rewards with completion conditions and tunable constants. A large language model (LLM) agent revises the program structure in context, drawing on execution feedback and a skill library of verified programs, while a numerical optimizer tunes its constants. With every candidate verified across parallel simulation scenarios, the program explores new ways to induce, repurpose, and compose the controller's existing motor competence for the task at hand, and thus improves over iterations. Experiments show that human-designed rewards leave much of the FB model's loco-manipulation competence untapped, whereas the programs InterEvolve evolves release it, sometimes through novel strategies. It further produces behaviors for diverse tasks, complex scenes, and long-horizon compositions in simulation, and evolved skills run autonomously on a physical Unitree G1 from egocentric onboard perception. --- ### Cost-augmented Schrödinger bridges on graphs are exactly solvable: a Feynman-Kac tilt replaces learned control Authors: Akshay Balsubramani Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02195v1 The generalized Schrödinger bridge on a graph moves mass between two distributions while charging a cost for the states visited. It has been approached by learning the rates of a controlled continuous-time Markov chain, with a temporal-difference penalty that restores the cost. A state cost folds into the reference process as a Feynman-Kac tilt. The cost-augmented bridge is then a plain bridge against the tilted reference, and the penalty is unnecessary. The bridge is computed exactly by alternating two endpoint rescalings, each one sparse matrix-exponential application; nothing is discretized in time or learned. The alternation converges at a rate set by the endpoint coupling alone. For a quadratic congestion cost on time-averaged occupancies, damped best response around the exact bridge is gradient descent on a strongly convex function, and its residual bounds its error. On a protein-folding model, a free-energy cost lowers the expected barrier of the folding paths. On the learned approach's road network, roll-outs of the exact bridge match the target within sampling error, and on networks with millions of intersections its memory grows linearly. --- ### Hierarchical Continuous Diffusion Language Models Authors: Hui Ren, Zihan Li, Chang Liu, Huidong Liu, Alexander Schwing Published: 2026-10-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.02193v1 Discrete diffusion language models offer a compelling alternative to autoregressive generation for tasks demanding bidirectional reasoning and global constraint satisfaction. Yet they share a structural bottleneck: when decoding in parallel, each token is sampled independently from its marginal, severing the statistical dependencies among the tokens decoded together. Continuous diffusion language models avoid this by denoising a shared continuous state, but their denoiser sees only that state, so nothing ties it to a valid token configuration until it is finally decoded. To address this, we propose Hierarchical Continuous Diffusion Language Models (HC-DLM), which couple discrete token generation with a continuous latent trajectory in a single, principled denoising process, whose training objective is derived from a variational bound on the token likelihood. In contrast to recent methods that attach continuous context to a self-contained discrete chain, HC-DLM makes the latent the only persistent generative state: tokens are read out from it at every step and feed back as a scaffold for the next latent update. On structured reasoning (Sudoku), mathematical planning (Countdown) and language modeling (LM1B), HC-DLM improves over discrete and continuous diffusion baselines at matched model size, in puzzle accuracy on Sudoku and Countdown and in generative perplexity on LM1B. Project page: https://hc-dlm.github.io/. --- ### The Missing Primitive: Diagnosing and Repairing Mathematical Reasoning in Large Language Models Authors: Shuo Xing, Zilin Dai, Chengyuan Qian, Fangzhou Lin, Wenjing Chen, Ping He, Pan Lu, Alvaro Velasquez, Mohit Bansal, Zhengzhong Tu Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02191v1 While Large Language Models (LLMs) have demonstrated striking capabilities on frontier mathematical problems, it remains unclear whether they possess the structural mathematical understanding underlying their solutions. In this paper, we take a first step toward systematically studying mathematical understanding in LLMs, from diagnosing its distinct capabilities to leveraging these findings to improve post-training. First, we introduce the notion of Mathematical Primitive to probe structural mathematical understanding and propose \hlei{}, a novel benchmark that evaluates mathematical reasoning along four distinct dimensions: Discovery, Generation, Digestion, and Execution. Second, our systematic diagnosis shows that solution accuracy masks distinct capability profiles, primitives unlock substantial latent execution capacity, and Discovery is the dominant bottleneck in mathematical reasoning. Our post-training analysis further shows that discovery-limited failures are particularly amenable to repair. Finally, building on these findings, we introduce \abs{}, a primitive-privileged self-distillation framework that selectively transfers primitive-guided reasoning into the student model. Extensive experiments demonstrate that \abs{} consistently improves mathematical reasoning over baselines across model scales and challenging benchmarks. --- ### Trust the Direction, Search the Step: Zero-and-First-Order Methods for LLM Fine-Tuning Authors: Cristian McGee, El Houcine Bergou, Aritra Dutta Published: 2026-10-01 Categories: cs.LG, math.OC arXiv: https://arxiv.org/abs/2610.02190v1 Step-size selection remains a central challenge in large-scale neural network optimization; conservative steps slow convergence, while aggressive steps can destabilize it. We combine \textbf{Z}ero-and-\textbf{F}irst-\textbf{O}rder optimization~(ZFO) and propose a lightweight framework that decouples direction selection from step-size. ZFO uses a trusted first-order optimizer to determine the direction and performs zeroth-order evaluations only along this one-dimensional subspace to choose how far to move. Using the current {gradient information} and two additional objective function evaluations, ZFO instances construct a local model of the objective function along the proposed direction and select a curvature-aware step within a bounded search interval. This yields an adaptive step-selection mechanism that costs less than a full line search. We provide theoretical guarantees to show that shared-sample evaluations produce reliable finite-difference curvature estimates, that the induced local model selects a near-optimal step along the search interval, and that ZFO converges to a neighborhood of a stationary point. Across the evaluated settings, language models and datasets, ZFO frequently improves optimization and final performance relative to fixed-step first-order baselines, with the magnitude and preferred local model depending on the objective. Our code is publicly available at: https://github.com/nizswan/Zeroth-First-Order-Framework. --- ### Generative modeling of intrinsically disordered protein regions by reinforcing sparse autoencoder features Authors: Jason X. Liu, Sebastian Ibarraran, Frank Hu, Soojung Yang, Xinyu A. Feng, Abigail Park, Anagha Aneesh, Lacramioara Bintu, Alexander R. Dunn, Grant M. Rotskoff Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02189v1 Intrinsically disordered protein regions (IDRs) play central roles in cellular processes such as transcriptional regulation, signal transduction, and subcellular localization, yet their functional design remains challenging. Structure-based design methods do not readily apply to IDRs, and existing protein language models are trained on full-length protein sequences, thus learning a prior that is biased towards folded domains. Here, we present IDiom, an autoregressive protein language model trained on IDiom-DB, a dataset of 54 million predicted IDRs curated from the AlphaFold Database. IDiom generates diverse sequences that recapitulate the composition, patterning, motifs, and predicted disorder of natural IDRs. To control function-associated sequence patterns, we also introduce reinforcement learning with sparse autoencoder features (RL-SAE), a post-training method that rewards the generation of sequences that activate specified feature sets. Across eight IDR design tasks, RL-SAE sequences activate, on average, 90% of 30 targeted features, compared to 24% for activation steering. We demonstrate that RL-SAE improves the predicted subcellular localization and transcriptional activity of generated IDRs compared to steering and supervised fine-tuning, and enables features associated with distinct biological functions to be combined within individual sequences. Thus, IDiom and RL-SAE enable interpretable and composable IDR design through explicit control of function-associated sequence features. More broadly, RL-SAE could extend to other protein design settings where interpretable features provide useful design targets. Code is available at https://github.com/rotskoff-group/idiom. --- ### DMAD: Distribution Matching as Adversarial Distillation for Fast Visual Generation Authors: Zhengming Yu, Junkun Yuan, Haotian Yang, Gordon Guocheng Qian, Yizhi Wang, Angtian Wang, Yiding Yang, Bo Liu, Xin Li, Wenping Wang, Chongyang Ma Published: 2026-10-01 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.02188v1 Distribution Matching Distillation (DMD) trains a few-step student from the difference between separately estimated target and student scores, so it must keep an auxiliary diffusion model fitted to the student's evolving distribution at extra memory and computation cost. We introduce DMAD, Distribution Matching as Adversarial Distillation, which recasts distribution matching as classification and learns the required log-density ratios directly. Two discriminator heads on a shared backbone distinguish real data and teacher samples from the student's, and linear losses on their logits train the student without auxiliary score fitting. We prove that at the discriminator optimum these losses recover the distribution-matching gradient underlying DMD, through the classical identity linking discriminator logits to log-density ratios. We further introduce gap-based reweighting, which adapts teacher supervision across noise levels from the real-data head's empirical logit gap between real and teacher samples. DMAD reaches a Fréchet Inception Distance (FID) of 1.04 with one-step generation on ImageNet-64x64, 14.47 with four-step SDXL on COCO-10K, and a VBench total score of 85.15 with four-step Wan2.1-T2V-14B, the best values among the compared few-step methods and the multi-step teachers. On MiniMax-H3-33B, our four-step student achieves overall human preference rates of 79.1% over DMD2 and 84.6% over rCM for joint audio-video generation, excluding ties. Our code, models and demos are available at https://yzmblog.github.io/projects/DMAD. --- ### Higher-Order Molecular Grammars for Generative and Foundation Models in Chemistry Authors: Yiming Huang, Yujie Zeng, Vijay Prakash Dwivedi, Simone Foti, Jianmin Wang, Jure Leskovec, Tolga Birdal Published: 2026-10-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.02186v1 Molecular learning models are strongly shaped by their underlying representations. Yet standard sequential and graph formalisms struggle to explicitly encode higher-order topology, such as ring systems and recurring motifs. Existing higher-order representations can capture these structures directly, but they are often computationally demanding and difficult to decode into valid molecules. Here, we introduce Higher-order Grammar Representation (HGR), a principled, topology-aware framework that lifts molecules to combinatorial complexes and parses each complex into a compact sequence of production rules under a context-free higher-order grammar. By serialising higher-order topology into rule sequences, HGR makes these structures directly compatible with standard sequence models, avoiding the computational overhead of explicit higher-order encodings while preserving topological expressiveness. To reduce benchmark bias towards simple ring systems, we construct RingDiv, a ring-enriched benchmark containing 1.18 million molecules, including the curated RingDiv300k subset, and introduce the ring diversity index (RDI) to quantify ring-system coverage. In molecular generation, HGR-based models uniquely combine 100% validity by construction with leading distributional alignment, ranking first in FCD on all five generation benchmarks. In representation learning, HGR-FM achieves the highest mean AUC across seven MoleculeNet benchmarks under both transfer protocols, improving on the strongest baseline by 8.3 and 3.3 AUC points under probing and full fine-tuning, respectively. Collectively, these results establish HGR as an efficient higher-order representation for molecular generation and transferable representation learning. --- ### Decoding Looped Transformers Better for (Almost) Free Authors: Weihao Liu, Huangjie Zheng, Tianrong Chen, Rohit Dilip, Richard He Bai, Yizhu Jiao, Yuyang Wang, Ruixiang Zhang Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02185v1 Looped Transformers achieve parameter efficiency by repeatedly executing a shared block across recurrent loops. Each loop yields an intermediate representation decodable for the same next token, yet standard decoding discards earlier states. Because earlier loops embody less computation, recurrence inherently supplies aligned weak-and-strong prediction pairs without auxiliary models or external training. We introduce LoopCD, a training-free contrastive decoding framework that guides token selection by contrasting the final prediction with an earlier recurrent pass, operating either in logit space with one extra output pass (LoopCD-Logits) or in hidden-state space with zero output overhead (LoopCD-Hidden). Across four looped Transformer families, LoopCD delivers substantial, consistent gains at full recurrent depth: LoopCD-Logits raises Ouro-2.6B-Thinking's AIME 2024 pass@1 from 61.88% to 73.33%, while LoopCD-Hidden lifts Huginn's HumanEval pass@1 from 22.56% to 31.71%. Crucially, these performance gains enable halving the number of recurrent loops while still matching or exceeding full-depth unguided baselines, reducing forward FLOPs by 22.5% to 48.2%. By transforming intermediate recurrent states into effective guidance signals, LoopCD achieves superior decoding quality while substantially reducing inference compute. --- ### SoftServe: A Scalable Quasi-Newton Method for Deep Learning Authors: Joohwan Ko, Tetiana Parshakova, Diana Cai, Robert M. Gower Published: 2026-10-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.02182v1 Quasi-Newton (QN) methods have long been among the most effective methods for large-scale unconstrained convex optimization. Two obstacles have limited their use in deep learning: non-convexity and enormous parameter sizes. We introduce SoftServe, a family of QN methods designed to overcome these obstacles without line searches or ad hoc curvature corrections. SoftServe derives positivedefinite curvature estimates from the variational objective of Berglund et al. (2025), even in the presence of negative curvature. We develop diagonal and Kroneckerfactored variants that preserve positive definiteness by construction and scale to massive neural networks. Finally, SoftServe relies on the stable coupled Newton-Schulz iteration for the required matrix operations, replacing costly matrix decompositions with GPU-friendly matrix multiplications. SoftServe excels on problems that are severely ill-conditioned, including tasks such as recurrent networks, deep autoencoders, physics-informed neural networks, and a 136M-parameter physics-informed diffusion model, often achieving lower losses than established baselines including Adam, Muon, and SOAP. --- ### OmniSeek: Native Tool Integration for Multi-turn Audio-Visual Reasoning Authors: Haibo Wang, Jiteng Mu, Jialu Li, Jingru Yi, Yuanjun Xiong, Jianming Zhang, Lifu Huang, Mingze Xu Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02181v1 We present OmniSeek, an agentic framework that transforms an Omni Large Language Model (Omni-LLM) into an active, multi-turn reasoning agent with native tool use. Rather than passively processing an entire audio-visual sequence in a single forward pass, OmniSeek makes evidence acquisition part of the reasoning process: it dynamically decides whether to look or listen, and over which temporal window, to retrieve sparse but critical evidence across different modalities within long contexts. Through an iterative multi-turn protocol, the retrieved raw audio or visual segments are appended back into the context to support subsequent reasoning. To cold-start this capability, we build a data engine that synthesizes OmniTraj-170K, a corpus of multi-hop Chain-of-Thought trajectories with interleaved audio and visual evidence. We first supervise the model on these trajectories to instill multi-turn tool-use behavior, and then further optimize the policy via a two-stage reinforcement learning with verifiable rewards. Moreover, we introduce an Audio-Visual Necessity objective that explicitly rewards successful trajectories whose reasoning depends on both modalities, discouraging single-modality shortcuts. Extensive experiments across a wide range of benchmarks demonstrate that OmniSeek learns adaptive cross-modal evidence seeking and consistently improves audio-visual reasoning performance. --- ### Generative Cinematographer: Composing Camera and Object Motion in 3D Authors: Jiahan Zhang, Chaohao Yang, Namitha Guruprasad, Vivekjyoti Banerjee, Trong-Tung Nguyen, Alan Yuille, Anand Bhattad Published: 2026-10-01 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.02180v1 Current controllable video generation systems often rely on 2D motion trajectories or sparse drag signals for object motion. These controls are ambiguous because the same 2D trajectory can correspond to different 3D motions, especially when the camera and objects move simultaneously. We present Generative Cinematographer (GenCine), a system that lifts a single image into an editable 3D scene scaffold where artists jointly author camera and foreground motion. Artists specify a camera path and move selected foreground regions using local 3D motion handles. Several handles can move different parts of a subject independently, providing a piecewise-rigid approximation to non-rigid motion without a physics simulator or category-specific prior. To communicate these controls to a pretrained video model, we project them into guidance maps. These maps record where the controlled regions appear in each frame, assign each handle a fixed color across frames and encode the current 3D positions of its controlled points in the same world coordinate system as the background. This lets us describe object motion relative to the scene even as the camera moves. For training, we recover controls from the motion observed in real videos and use ground-truth geometry and trajectories from synthetic videos. We train a lightweight guidance branch and LoRA adapters on a pretrained Wan model to follow these controls. Our experiments show consistent camera-relative motion, improved geometric consistency under viewpoint changes, and strong controllability across diverse real-world scenes. --- ### From Gradients to Capabilities: Understanding Multi-Teacher On-Policy Distillation Authors: Siqi Zhu, Suozhi Huang, Kaixuan Zhang, Yuheng Yang, Zhanyang Jin, Yihang Sun, Jiaxuan You Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02179v1 Multi-teacher on-policy distillation (MOPD) aims to combine the strengths of RL-trained teachers in a single student, but how teacher signals affect parameter changes remains underexplored. We study Qwen3-1.7B with four domain teachers trained with RL from the same initialization as the student, comparing gradients, optimizer updates, and task learning curves, with additional SmolLM3-3B diagnostics. We find that several factors influence teacher signals. First, loss averaging implicitly weights responses: token averaging favors longer responses, and equalizing domain contributions retains this weighting within domains. Second, Adam's first moment reduces differences in parameter updates: the cosine similarity is 0.83 between teachers and 0.96 between averaging rules, despite differences in raw gradients. Third, BF16 rounding hides small changes: about 97\% of FP32 master weights differ from initialization, but only 7--11\% of BF16 weights do. Finally, the top-64 intersection KL gradient closely matches Qwen's full-vocabulary gradient, but the effect on task performance depends on averaging: mathematics accuracy is 2.6 points higher than with sampled-token policy-gradient (PG) under response averaging and 2.1 points lower under global token averaging. --- ### Effective Resistance and Graph Neural Network Reliability in Tissue-Specific Interactomes Authors: Jianru Shen Published: 2026-10-01 Categories: cs.LG, q-bio.MN arXiv: https://arxiv.org/abs/2610.02175v1 Protein function annotation needs to know which predictions to distrust, not only what a model predicts. We ask whether tissue-specific interaction structure carries that information. Our candidate signal is effective resistance, used previously to relieve over-squashing by rewiring. Across 24 tissue-specific interactomes it is dominated by inverse degree, and the degeneration deepens as the co-expression filtered network grows, with a Spearman correlation of -0.955. The residual departure from that limit exceeds degree-preserving null graphs in all 24 networks. Controlling for predictive entropy, degree, annotation cardinality, local structure and feature-only difficulty, the residual explains additional per-node loss in 19 of 24 held-out networks once a permutation floor is subtracted, at every depth, and the effect strengthens monotonically with depth. The increment reaches 0.37% of the variance the controls leave unexplained, 5.6 times a permutation floor, against 1.5 times when the model is retrained in a degree-preserving null world. Selective prediction improves negligibly. The signal is reproducible; degree degeneration bounds it. --- ### Every Ablation Is a Dose: Counterweights and the Semblance of Self-Repair Authors: Areeb Ahmad, Pratinav Seth, Vinay Kumar Sankarapu Published: 2026-10-01 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2610.02173v1 Ablate a component of a language model, and other components often appear to adjust and compensate. This phenomenon, termed self-repair, has been observed repeatedly, but its mechanism remains unclear. The most systematic study to date concluded that self-repair is noisy and unlikely to have a single explanation. We argue that it has one: a gain already present before any ablation. Any intervention on a causally important component can be viewed as a point on a coordinate axis $λ$, the signed strength of a counterfactual contrast. Hence, conventional ablation methods are uncalibrated points on this axis. We show that the causal repair response for a fine-grained unit $r$ is governed by an affine law, $E_r(λ)=\mathrm{own}_r+γ_rλ$. The slope $γ_r$ is a fixed coefficient that consistently influences the model, with or without ablation, and its sign determines whether the unit counteracts or reinforces the removed signal. On a factual-verdict task across four models from distinct families (Gemma, Qwen, LLaMA, and Mistral), we identify components including MLP neurons, OV neurons, and singular directions that follow this affine law, 68 of 81 downstream directions in all. Moreover, we can anticipate the magnitude of $γ_r$ from the fixed weights. On the IOI circuit of GPT-2 Small, seven of the ten heads the intervention can reach follow the law, and all seven are counterweights. From this perspective, what may appear as self-repair is a counterweight performing its usual operation when the contrastive signal emerges at the core. --- ### Watch, Infer, Coordinate: Inferring Robot Partner Constraints for Zero-Shot Coordination Authors: Suyu Ye, Zheyuan Zhang, Vaishnav Tadiparthi, Hossein Nourkhiz Mahjoub, Ehsan Moradi Pari, Tianmin Shu, Homanga Bharadhwaj, Nakul Agarwal Published: 2026-10-01 Categories: cs.RO, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2610.02170v1 Robots operating in the physical world will increasingly need to coordinate with other robots, particularly in manipulation tasks where an object may be too large or heavy for a single robot to carry alone. Physical limitations caused by hardware degradation or actuator faults can restrict the actions a robot can reliably execute, yet these limitations may be unknown to its partner. We study whether a helper can infer a robot partner's physical constraints from observing it coordinate with another robot, then use the inferred capability to coordinate with the same partner on a new task. This is difficult because a demonstration shows what the constrained robot did, but not what it could have done. In physically coupled tasks, the other robot may also compensate for its limitations, making those limitations difficult to identify from the constrained robot's behavior alone. Our key insight is that these constraints shape the joint behavior of the team, making the actions of both robots informative about the constrained partner's capability. We introduce Watch, Infer, Coordinate, a benchmark spanning three physically coupled manipulation settings, together with an inference approach that scores candidate constraints using observed joint behavior. Across all three settings, our method substantially improves constraint inference and zero-shot coordination, approaching an oracle with access to the true constraints. --- ### AutoCompact: Learning When to Compact Context in Long-Horizon Coding Agents Authors: Xuan Zhang, Longtao Zheng, Cunxiao Du, Bo An, Xin Dong Published: 2026-10-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.02163v1 Coding agents solve repository-level software engineering tasks through long trajectories of code inspection, search, editing, and testing. As a task progresses, earlier exploration becomes stale, so managing context is more than avoiding overflow: an agent must decide when to compact, what working state to preserve, and how to continue from it. We introduce AutoCompact, which trains a coding agent to make these decisions as part of its policy. To collect training data, we run the base agent on coding tasks and use a judge to review its compaction decisions, summaries, and actions after compaction. Flawed outputs are replaced with corrected ones before being executed in the environment, so each trajectory continues from the corrected decisions. We use these trajectories for supervised fine-tuning, then jointly optimize coding and compaction through reinforcement learning with task-success rewards. Experiments on SWE-bench Verified and SWE-PolyBench Verified show that AutoCompact improves pass rates over the base model by an absolute 9.2\% and 5.0\%, respectively. The improvements hold across all evaluated inference budgets, with a 256K context window that never overflows and with a 16K window whose overflow triggers fallback compaction. --- ### World Observer: Joint Actor-Observer Generation for Persistent World Modeling Authors: Hyunwook Choi, Dahyun Chung, Hyunsung Kim, Siyoon Jin, Jinhyeok Choi, Junyoung Seo, Seungryong Kim Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02162v1 How can a world model continuously observe regions beyond the actor's current view? Video world models simulate how an environment evolves from an agent's actions, yet remain actor-centric. Once an object leaves the actor's view, they lose direct evidence of its evolution, often failing to preserve its state and dynamics upon re-entry. To address this, we introduce World Observer, which decouples observing from acting by jointly generating a perspective actor for the agent-centric view with one or more panoramic observers that watch selected world regions. This allows objects that leave the actor's view to remain visually evolving in an observer, so their updated states are reflected when they re-enter. We ground the actor and observers by warping from a shared panoramic source for explicit geometric correspondence, and introduce an Observer Sink of high-resolution perspective references to restore fine appearance upon re-entry. Since the observers are decoupled from the actor, they can be placed freely across the scene, extended to multiple locations for broader coverage, and driven by control signals to steer out-of-view evolution. To evaluate out-of-view evolution, we further introduce world-space metrics and a benchmark spanning real and synthetic scenes. World Observer substantially improves out-of-view dynamics while remaining competitive in visual fidelity, camera control, and 3D adherence. --- ### DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication Authors: Hanchu Zhou, Dechen Gao, Hang Wang, Brendan Lynch, Boqi Zhao, Qiyao Ma, Raman Goyal, Junshan Zhang Published: 2026-10-01 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2610.02161v1 Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general-purpose robots, yet most progress has focused on single-robot settings. Extending these capabilities to multi-robot systems remains challenging because robots must coordinate long-horizon behaviors while maintaining reliable, fine-grained execution. We introduce DuoMind, a distributed hierarchical framework for multi-robot coordination through semantic communication. Each robot uses a VLA-based action model for low-level execution and a VLM-based orchestrator for high-level reasoning and inter-agent coordination. At each planning step, the orchestrator at each robot reasons over the task instruction, local observations, and messages received from other robots. It then generates low-level instructions for the action model and semantic messages for peer robots. This architecture exploits the complementary strengths of pretrained models by combining the semantic reasoning capabilities of VLMs with the precise action-generation capabilities of VLAs. To address the scarcity of benchmarks for multi-robot coordination, we further develop RoboPoly, a benchmark comprising long-horizon manipulation tasks that require coordinated, closed-loop execution under distributed control. Experiments on RoboPoly and RoboTwin demonstrate that DuoMind improves multi-robot task performance, while ablation studies confirm the contributions of hierarchical orchestration and semantic communication. More details are available on our project page. --- ### 4Director: Controlling Video World Models with Rigid 3D Geometry Authors: Wei Cao, Hao Zhang, Vikram Voleti, Yuqun Wu, Mallikarjun B R, Shimon Vainer, Mark Boss, Yaoyao Liu Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02160v1 Precise control over camera and object motion is essential for professional video production. Existing methods control objects only coarsely, through image-plane cues that are ambiguous in depth and rotation or through 3D tracks and blobs that lack complete geometry and lose consistency across viewpoint changes. We introduce 4Director, a video world model conditioned on an explicit 4D scene representation: each object is reconstructed once from the input image as a canonical mesh and moved by one prescribed rigid transformation per frame. This representation provides an intuitive 3D control interface and prevents unobserved geometry from being regenerated independently in every frame. We render the controlled scene as a depth video and introduce a Motion Adapter that transforms this geometric scaffold into video while synthesizing view-consistent appearance, illumination, and non-rigid dynamics. For training, we construct RealCOD-Rigid, a new dataset of 20,774 clips annotated with rigid 3D scenes by our automatic pipeline. We further introduce Identity-Gated IoU (IG-IoU), which jointly evaluates adherence to prescribed object motion and preservation of object identity. Experiments demonstrate that 4Director consistently outperforms prior methods in visual quality and in camera and object control. --- ### When Do Intrinsic Rewards Lead to Exploration? Authors: Scott W. Viteri, Laura Gomezjurado Gonzalez, Clark Barrett Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02159v1 Intrinsic rewards are designed to guide exploration in reinforcement learning by assigning value to an agent's experience, for example through prediction error or learning progress. However, maximizing these rewards need not produce the most informative experience available. We propose a formal criterion for exploration that compares policies by the counterfactual information they acquire: how well their histories can substitute for experience under alternative policies. We construct a single, simple environment in which specified count-based, prediction-error, empowerment, and information-gain objectives have maximizing policies that are Pareto-suboptimal at acquiring counterfactual information. We explain these failures and establish conditions under which existing intrinsic rewards successfully encourage optimal exploration. We also construct an objective that assigns a higher value whenever exploration strictly improves under our criterion. --- ### Muon meets Tamed Langevin: Momentum Preconditioning beyond Convex and gradient-Lipschitz Potentials Authors: Nikolaos Makras, Sotirios Sabanis Published: 2026-10-01 Categories: cs.LG, math.OC, math.PR, stat.ML arXiv: https://arxiv.org/abs/2610.02158v1 We consider the problem of sampling from Gibbs distributions on matrix spaces whose potential energies are neither convex nor globally gradient-Lipschitz. We introduce a family of non-quadratic kinetic energies that lead to a new underdamped Langevin system with momentum preconditioning, in which the gradient of the kinetic energy acts as a smooth spectral taming of the momentum. We prove that, under these relaxed assumptions on the potential, the resulting dynamics leaves the target Gibbs measure invariant, and we establish exponential convergence to equilibrium in a weighted total variation distance. Finally, we show that the corresponding Euler-Maruyama discretization admits moment bounds that are uniform in time, without any modification of the potential gradient, which ensures the stability of the resulting sampling algorithm. --- ### MosaiChunk: Compositing Spatio-Temporal Memory for Autoregressive Video Generation Authors: Yiwen Zhang, Haocheng Xi, Michael Tian-Yue Liu, Alexei A. Efros, Hadar Averbuch-Elor, Qianqian Wang, Haiwen Feng Published: 2026-10-01 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2610.02153v1 Long-horizon autoregressive video generation is limited by a finite context window. When an object or scene falls out of context, its fine-grained visual details may be lost and difficult to recover upon reappearance. To retain access to such visual details, we introduce MosaiChunk, a spatio-temporal memory mechanism that composes a mosaic of selected historical key-value (KV) entries across space and time. Our approach is motivated by the observation that a frozen video generator can directly consume such non-contiguous historical KV and recover the corresponding visual content. We therefore keep the generator fixed and learn only a lightweight router that determines which historical sections to include in the mosaic under a fixed active-memory budget. We further introduce RememBench, a benchmark of long-horizon revisits with prompt-driven text-to-video (T2V) and camera-driven image-to-video (I2V) splits. Our experiments show that MosaiChunk consistently improves revisit consistency over both sliding-window inference and whole-chunk retrieval under matched memory budgets, across both T2V and I2V settings. --- ### From Knowledge Access to Source Learning: Developing Source-Specific Competence Authors: Lucheng Fu, Kejing Xia, Yiyang Wang, Yiqiao Jin, Jinjin He, Xiyuan Yang, Haoxin Liu, Ye Yu, Haibo Jin, Yijia Xiao, Wenke Lee, B. Aditya Prakash et al. Published: 2026-10-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.02150v1 Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-intensive tasks. Existing methods improve how source content is accessed and organized, while agent-memory systems preserve reusable knowledge from prior interactions, but repeated use of the same source is still largely treated as repeated access rather than an opportunity to progressively improve understanding of that source. We study source learning: developing reusable source-specific competence over a persistent authoritative source. We represent this competence with a persistent source model that captures reusable understanding of the source, including how its knowledge is structured, interpreted, and applied. To construct and progressively refine such models, we propose SourceLearn, which combines two complementary learning mechanisms. Self-Directed Source Learning identifies what remains incompletely understood and adaptively revisits the source, while Task-Guided Source Learning uses downstream experience to reveal local representational gaps and recurring needs in how source knowledge should be organized. In both cases, learning signals determine what should be reconsidered, while persistent updates are reconstructed from the authoritative source. Across five benchmarks and three LLM backends, SourceLearn achieves the best performance in 13 of 15 settings, with gains of up to 22.6 points over Hybrid RAG and substantial overall improvements over static source representations and experience-based memory baselines. --- ### Omni-Embed-Mini: Binding Modalities Without Forgetting via Dense Distillation Authors: Mohammed Irfan Kurpath, Jaseel Muhammad Kaithakkodan, Sahal Shaji Mullappilly, Ivan Laptev, Hisham Cholakkal Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02148v1 Extending a text embedding model to new modalities typically degrades text retrieval quality, and existing omni-modal embedders compensate with multi-billion parameters. We present Omni-Embed-Mini, a 0.9B-parameter model that maps text, speech, audio, images, video, and visually-rich documents into a single shared cosine space without updating any text-side parameter. Our key insight is that the teacher signal requires no separate embedding model: each media sample is paired with a dense cascaded caption, and the teacher target is simply the frozen backbone's own embedding of that caption. Because teacher and student share the same backbone weights, they inhabit byte-identical geometry, and lightweight projectors plus phased LoRA adapters on the modality encoders suffice for alignment. Training combines a Matryoshka SigLIP contrastive loss with an online hybrid hard-negative miner whose negatives sharpen as the encoder improves. The recipe carries over to a 2.3B variant by swapping in a native vision-language backbone. Omni-Embed-Mini-0.9B keeps its text weights bit-identical to the backbone, so training cannot regress text retrieval (49.57 nDCG@10 on MTEB-v2 BEIR-8), while extending it to five additional modalities, and is ~2.7x to 9.5x smaller than every open omni embedder we compare against. The 2.3B variant is competitive with the closed gemini-embedding-2, edging ahead of it on the overall-modality average. Models, code, data and evaluation harness are on our project page: https://omniembed.cvmbzuai.com --- ### Faynt: Scaling and Optimizing Policies for Competitive Melee Authors: Ali Janati, Nikita Kuzmin, Rohit Swamy, Charles Niu Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02144v1 We introduce Faynt, a family of 10M- and 75M-parameter Transformer policies for Super Smash Bros. Melee, each controlling all 26 characters with a single checkpoint. After reinforcement learning (RL), the 10M wins 240 of 244 same-character games (98.4%) against fourteen specialist and multi-character releases on their supported rosters, with a winning record against every release. These opponents retain 21- or 24-frame action delays; Faynt uses no added delay, and we have not isolated the effect of this difference. In a separate evaluation against a privately supplied zero-delay Slippi-AI model, the 10M wins all 68 games across two conditioning settings. We study architecture, optimization, scaling, and hyperparameter transfer to guide pretraining on approximately 840,000 human replays. Post-training combines rank- and outcome-based curricula, 75M-to-10M distillation, and RL restricted to Fox mirror matches. On the initial 152-game benchmark, the supervised 10M wins 69.7% of games, compared with 45.4% for the pretrained 75M, despite higher overall held-out controller-prediction loss. The weighted validation loss used for supervised checkpoint selection agrees with the win-rate ordering of all four pretrained and supervised policies. After supervised post-training, both models take less damage per minute, build larger early leads, and win more often after losing the first life. Optimized inference on recorded game states averages 5.2 ms per decision for the 10M and 8.7 ms for the 75M on an NVIDIA T4, excluding emulator execution and communication. We open-source the weights, both benchmark suites, and a platform for automated model tournaments. --- ### Keyword Harnesses Fail Open: A Cheap Diagnostic Ladder for Tool-Use Claims in Small Language Models Authors: Juan S. Santillana Published: 2026-10-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.02142v1 Keyword-matching benchmarks can credit small models for tool use they never perform. We document such a false positive in a matched-architecture pair of Spanish security language models and propose a ladder of strict, cheap diagnostics. A 661.6M parameter model (approx. 65% code/technical text; no dedicated SFT) and a 1,109M model (web-heavy multi-phase curriculum; 6B-token tool-SFT) share decoder, tokenizer, and special tokens, scoring almost identically on lenient tool-use metrics (B4: 0.660 vs. 0.650). Verbatim-reproduction checks on training examples separate them completely: the 600M emits valid tool calls with generalized arguments on 6/6 examples; the 1B does so on 0/6 across checkpoints. A first-token probe localizes the 1B's failure to a missing prior (prob. $10^{-4}$--$10^{-5}$ on <|tool_call|>), which was erased by its web-heavy training phase. A targeted SFT recipe (diverse corpus, 5x higher learning rate, 2,202 steps, ~3.3 GPU-hours) repairs the 1B using three orders of magnitude fewer tokens than the failed phase. On all 269 corpus rows, valid emission rises from 0.100 to 0.959 (600M: 0.926). On 238 unseen prompts, the repaired 1B passes 0.536 vs. the 600M's 0.428 ($p = 0.004$). Embedding-drift checks show the repair did not move the trigger token's tied embedding (97.7% of the bf16 table remains bit-identical), meaning changes live in the surrounding network. Both models over-trigger, rarely answering negative prompts without a call (0.09 for 600M, 0.17 for repaired 1B). Factorial analyses confirm all repair configurations install the format, though suppression benefits from a diverse corpus remain a hypothesis due to seed sensitivity. This cheap diagnostic ladder costs minutes of CPU time and should gate tool-use claims on small models. --- ### Finetuning with Sampling: SFT Learns Better Than You Think Authors: Aayush Karan, Sitan Chen, Yilun Du Published: 2026-10-01 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.02140v1 Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supervised finetuning (SFT) and reinforcement learning (RL) to this end. Conventional wisdom dictates that RL enables strong generalization on new tasks without losing existing capabilities, while SFT is prone to weak generalization and catastrophic forgetting. At the same time, SFT can learn from off-policy expert data, whereas RL must rely on a model's ability to find successful trajectories with repeated sampling. In our work, we seek to leverage the strength of on-policy learning while utilizing the privileged information contained in off-policy data. However, rather than modifying the learning objective to accommodate this data, we instead tailor the data distribution to better suit the learner. We introduce a Markov chain Monte Carlo (MCMC) sampling algorithm that progressively transforms off-policy traces to be more on-policy given a reference model for finetuning. Across tasks like scientific skill acquisition, mathematical reasoning, and open-ended expertise, our sampling algorithm enables SFT to rival prevailing posttraining techniques, often generalizing better and forgetting less than strong on-policy baselines. In addition, the resulting finetuned models exhibit strong distributional performance and are capable of learning beyond sharpening the base model distribution. At a higher level, our approach presents sampling as a model-native operator that shapes data for learnability, offering broader utility as a general-purpose primitive throughout the posttraining stack. --- ### MIRTO: a registration-gated, multiverse-tested evaluation protocol for unsupervised anomaly segmentation in brain MRI Authors: Negin Kafee Hernashki, Soumick Chatterjee Published: 2026-10-01 Categories: cs.CV, cs.AI, eess.IV, physics.med-ph arXiv: https://arxiv.org/abs/2610.02136v1 Unsupervised anomaly detection (UAD) methods for brain MRI are ranked by a single score, yet that score rests on choices that are rarely reported: how each anomaly map is aligned with the reference, how and on which data the threshold is set, and which false-positive budget, metric, aggregation and lesion definition are used. We present MIRTO, an evaluation protocol that makes these choices explicit and measures their effect. It gates the geometry of every comparison with a registration check and label-free diagnostics of known power, sets thresholds on validation data alone and reports the false-positive volume actually realised on test, repeats each comparison over 15,552 defensible evaluation pipelines, and attaches paired subject-bootstrap intervals with multiplicity control. Applied to four UAD methods trained on the same healthy data and tested on 312 BraTS 2020 subjects, MIRTO showed that an axis-order mismatch between stored maps and the reference lowered a diffusion model's voxel AUROC from 0.873 to 0.583 whilst barely moving its slice-level AUROC. Within each metric, the method explained at least 0.95 of the variance in voxel AUROC and AUPRC and 0.77 in Dice, but only 0.14 in lesion sensitivity, where the lesion definition and hit criterion dominated. A Dice advantage that was significant at validation thresholds vanished at equal realised false-positive burden, and an exact identity attributes it to threshold transfer. A training-free change to REFLECT's latent aggregation raised Dice at equal burden by 0.052. Nine hypotheses were tested against explicit criteria; because the same cohort served to develop the protocol, all inference is exploratory. --- ### Linear Programming Representations and Strongly Polynomial Algorithms for Robust Markov Decision Processes Authors: Han Zhong, Yinyu Ye Published: 2026-10-01 Categories: cs.LG, cs.DS, math.OC arXiv: https://arxiv.org/abs/2610.02131v1 We study linear programming (LP) representations and strongly polynomial algorithms for robust Markov decision processes (RMDPs) with rational polyhedral state-action rectangular uncertainty in rewards and transitions. By encoding a finite sequence of robust policy-iteration steps, we construct a single LP whose optimal solutions recover the robust optimal value and all optimal stationary randomized policies. At fixed discount, the LP has polynomial dimension and encoding length and can be constructed in strongly polynomial time. We also develop a general complexity analysis of robust policy iteration that combines the cost of minimizing over uncertainty sets with the number of iterations needed to evaluate a policy. For a fixed discount factor, we use this analysis to improve the known complexity bounds for $\ell_1$ and $\ell_\infty$ RMDPs and establish new strongly polynomial bounds for general interval, weighted $\ell_1$, and Wasserstein RMDPs, as well as turn-based stochastic games with these uncertainty sets. --- ### Spiking neural networks for streaming qubit readout Authors: Barry M. Dillon, Aqib Javed, Jim Harkin, Patryk Dabkowski, Benjamin Lienhard Published: 2026-10-01 Categories: quant-ph, cs.NE arXiv: https://arxiv.org/abs/2610.02129v1 Fast and accurate qubit-state assignment is essential for feedback, calibration, and error correction in quantum processors. In superconducting platforms, frequency-multiplexed readout makes this task intrinsically multivariate as measured traces can encode crosstalk, qubit-state relaxation events, and other transient nonidealities that are not fully captured by conventional matched filtering. Here, we introduce spiking neural network (SNN) discriminators for superconducting qubit readout. By processing the measurement window in successive time chunks, the networks exploit temporal structure and update classification scores as data arrive, rather than waiting until the end of the readout window. The spiking networks outperform matched-filter discrimination and approach the accuracy of a full-trace artificial neural network. Beyond reaching the performance of artificial neural networks, the key advantage of SNNs is that they provide a streaming, time-resolved estimate of the qubit state that evolves as the readout signal is acquired. Using quantisation-aware training and hls4ml synthesis, we further demonstrate that each FPGA inference update can be completed before the next readout chunk arrives. These results establish spiking neural networks as a promising route to low-latency, real-time qubit readout on FPGA hardware, with broader implications for time-critical quantum-control and scientific-inference applications. --- ### Sample complexity bounds for categorical Markov random fields via Discrete Diffusions Authors: Shivam Kumar, Nabarun Deb Published: 2026-10-01 Categories: math.ST, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2610.02128v1 Many applications in statistics, economics, and physics require sampling from high-dimensional categorical distributions with local dependence structures. Examples include finite memory language models, Ising and Potts systems in statistical physics and protein folding, etc. In modern machine learning, discrete diffusions have emerged as a flexible approach for sampling such data, with strong empirical performance. Motivated by this, we develop learning methods with end-to-end sample complexity bounds for discrete diffusion with uniform noising under local dependence, which we model through low order Markov random fields (MRFs). Our main technical insight is a new \emph{pinning decomposition} of the discrete score. It shows that unlike in continuous diffusions, the score decomposes into components where the dependence on time separates multiplicatively from the dependence on the target. Building on this decomposition, we propose a \emph{weight-sharing neural score learner} and combine it with $τ$-leaping to obtain an end-to-end sampling procedure. Rather than treating score-learning error as a black-box input, as is common in existing sampling analyses, we study the score learning error from finite data and derive optimal sampling guarantees with explicit dependence on the vocabulary size, the interaction order of the MRF, and the sample size. Moreover, our strategy trains a single score network across uniform noise levels while leaving the sampling discretization to be chosen at inference-time. This allows the same trained model to trade accuracy for computational cost as inference-time budgets vary. Numerical experiments on Potts, Ising, and tree-structured models show that weight-sharing score networks outperform fully connected ones for sampling long sequences. --- ### Local Support Learning Authors: Assaf Ben-Kish, Akarsh Kumar, James Glass, Raja Giryes Published: 2026-10-01 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.02126v1 We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of each weight matrix, we uncover a natural retention objective under which updates produced by gradient-based optimizers are suboptimal. Following this observation, we propose Local Support Learning (LSL), a general-purpose framework that augments gradient-based training for retention of prior capabilities without access to prior data. During a new learning phase, LSL pairs two components with distinct roles: a standard weight adapter, trained as usual to minimize the loss, and a gating function that enables the adapter only on input activations from its own training distribution, making the update local to that distribution. The key challenge is that this gate must route data from all learning phases while training only on data from the current one. We address this with a gate based on a Gaussian Mixture Model (GMM), whose likelihood decays rapidly away from its training data, giving it a natural tendency to stay closed on data from prior phases. We show that this post-training approach can resolve forgetting in LLMs of up to 7 billion parameters, retaining both pretrained and finetuned capabilities across multiple training phases, while being efficient in memory and compute, robust to hyperparameter choice, and showing scaling potential. --- ### Harnessing Domain Specialists in Multimodal Mixture-of-Experts for Efficient Adaptation Authors: Damiano Marsili, Raphi Kang, Aditya Mehta, Pietro Perona, Georgia Gkioxari Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02123v1 Mixture-of-Experts (MoE) architectures scale model capacity through sparse computation, routing each token through only a small subset of experts. In this work, we explore whether this sparsity gives rise to emergent intrinsic organization in multimodal MoEs. We find that experts develop strong semantic specialization across modalities and domains despite not being explicitly trained for modularity. Building on this structure, we introduce ExpertLens, a data-free method that identifies domain-specialized experts directly from pretrained model weights by decoding router weights into semantically meaningful vocabulary tokens. We leverage this specialization for efficient multimodal adaptation by selectively fine-tuning experts relevant to a target domain. Across math, medical, and remote sensing tasks, ExpertLens matches or surpasses full fine-tuning while updating only 21.7 - 47.0% of model parameters and achieving a 4.0x average training speedup, and outperforms LoRA in both adaptation performance and training efficiency. These results show that sparsity introduced for efficiency can give rise to semantic modularity that is directly useful for efficient adaptation. --- ### Argo-Bench: Evaluating Data Agents on Enterprise-Scale Workflows Authors: Gabriel Tomitsuka, Arman Raayatsanati, Emma Xing, Duke Gand, Joseph J Ma Published: 2026-10-01 Categories: cs.CL, cs.AI, cs.DB arXiv: https://arxiv.org/abs/2610.02122v1 Real-world enterprise data science and analytics workflows require reasoning across dozens of tables, performing statistical analyses, and acting on the results. Established text-to-SQL benchmarks evaluate query generation alone, and audits have found their answer keys frequently wrong. Because real enterprise warehouses are too sensitive to release, these benchmarks are built on public datasets where a business event fits in a single table. We introduce Argo-Bench, an evaluation framework comprising 210 data science and analytics tasks. Drawing on public data, peer-reviewed industry literature, and regulatory filings, we simulate a food delivery platform in New York City at true scale, with 81 million orders in 2024, grounded economics, fraud patterns, and marketplace incentives. We export this world to an ERP warehouse of 235 tables and 7.5 billion rows, modeled on the Oracle E-Business Suite schema. The simulator's ground-truth state is withheld from the warehouse the agent sees, so tasks require reconstructing facts by navigating the warehouse before acting on them. Argo-Bench goes beyond text-to-SQL: the agent files actions such as banning fraudulent accounts, allocating courier incentive budgets, or issuing back pay, and the grader scores each by its consequences in the simulator. Every task has an executable reference solution that demonstrates solvability using only the warehouse. The strongest of 14 frontier and open-weight models scores 95 or higher on only 34.8% of tasks and averages 59.5 points. We hope Argo-Bench drives progress toward agents that understand, navigate, and act within real data environments. --- ### Where-OPD: Spatially Guided On-Policy Self-Distillation of MLLMs with Synthetic Scenes Authors: Sophia Sirko-Galouchenko, Monika Wysoczanska, Andrei Bursuc, Nicolas Thome, Spyros Gidaris Published: 2026-10-01 Categories: cs.CV, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2610.02117v1 On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supervising students with a frozen or EMA version of themselves that receives privileged information. Its application to multimodal large language models (MLLMs), however, remains largely unexplored. Recent approaches use privileged visual information, such as image crops corresponding to a question, to improve fine-grained perception, but their gains are confined to tasks that benefit from such visual zooming and require either human-annotated grounding data or external teacher models. We introduce a different form of on-policy self-distillation for MLLMs that provides the teacher with textual, spatially grounded guidance identifying the visual elements relevant to a query. We use procedurally generated scenes with automatically available object identities and spatial coordinates, enabling scalable and annotation-free post-training. The teacher uses this spatial guidance to locate and integrate evidence from multiple relevant image regions, while the student learns to reproduce the resulting behavior from the image and question alone. Our approach consistently improves performance on counting, document and chart understanding benchmarks across multiple models. Importantly, although post-training uses only synthetic scenes, the resulting improvements transfer to real-world perception benchmarks, yielding a 3.23-point gain in average performance across CVBench, V*, ZoomBench, BLINK, HR-Bench, and MME-RealWorld. These results show that spatially grounded privileged information can induce broader perceptual capabilities through on-policy self-distillation, enabling substantial synthetic-to-real transfer beyond the task and data distribution used for post-training. Project page: https://github.com/sirkosophia/Where-OPD --- ### A Comparative Explainability Framework for DeBERTa-v3 in Zero-Shot Medical Abstract Classification Authors: Javier Diaz Esteban-Herreros, David Muñoz-Valero, Raquel Martínez-España, Jose M. Juarez, Juan Moreno-Garcia Published: 2026-10-01 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.02116v1 A comparative explainability framework is presented to audit DeBERTa-v3 under zero-shot classification of medical abstracts. The work addresses the disagreement problem in Explainable Artificial Intelligence, where different attribution methods produce divergent explanations for the same input and prediction. A natural language inference engine is implemented over the Medical Abstracts corpus with five enriched hypotheses per diagnostic category and a balanced sample of one thousand texts per class. Five explanation methods are compared: SHAP and LIME as model-agnostic approaches, occlusion and Input x Gradient as deep-learning-specific approaches, and Attention x Gradient as a transformer-specific approach. Explanations are standardized through top-token attribution, and pairwise agreement is quantified using the Jaccard index. High predictive accuracy is achieved across well-defined clinical domains, whereas performance degrades under high semantic ambiguity. Explanatory stability directly mirrors predictive certainty, exhibiting strong convergence in univalent categories and a marked drop under diagnostic uncertainty. Furthermore, qualitative error auditing uncovers three systemic failure mechanisms: lexical hypersensitivity, semantic overlap, and loss of attribution coherence. The results support the combined use of several explanation methods and quantitative agreement metrics when auditing transformer-based models in medical text classification, and suggest prioritizing specific clinical ontologies over broad diagnostic labels. --- ### Surface-volume self-supervised representation learning of brain MRI for genetic discovery Authors: Tian Xia, Nuo Chen, Zihao Zhu, Huiwen Han, Ziqian Xie, Zhiwen Fan, Degui Zhi Published: 2026-10-01 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.02114v1 Existing genome-wide association studies (GWAS) of brain imaging provide predefined or deep-learning-derived imaging phenotypes, yet these phenotypes come from either volumetric scans or cortical surface meshes, so each captures only part of the heritable variation in brain anatomy. Here we introduce MEVA (Mesh-Enhanced Volumetric Autoencoder), a self-supervised framework that encodes voxel-level image intensity together with cortical mesh geometry, including curvature and cortical thickness at each surface vertex, into one shared set of imaging features. Combining the mesh and volumetric inputs in MEVA yields modest performance gains in age and sex prediction over models that use either input alone. When these features serve as phenotypes for GWAS in the UK Biobank, they reveal more genome-wide significant loci than features learned from volumes alone or from meshes alone. These results suggest that adding cortical surface geometry to volumetric self-supervised learning captures additional heritable variation and so increases the number of loci detected. --- ### Are We Recovering Mechanisms? Objective-Level Recovery Gaps in Mechanistic Interpretability Authors: Chuqin Geng, Li Zhang, Haolin Ye, Mark Zhang, Luke Zhang, Xujie Si Published: 2026-10-01 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.02098v1 Mechanistic interpretability aims to recover the internal computations responsible for model behavior. Progress in automated circuit discovery is often framed as a search problem: better attribution or optimization should identify better mechanisms. This assumes that the evaluation objective can recognize a better circuit once it is found. We show that intervention-defined faithfulness can instead prefer an equally sized circuit that reproduces the model's behavior less well, creating an objective-level recovery gap. Across four human-reference tasks and InterpBench, we compare validation faithfulness with behavior on held-out prompts under fixed ordinary resampling. The behavioral criterion is agreement with the intact model, including its mistakes, except on Greater-Than, where we use semantic accuracy. Controlled reference edits reveal misranking without any discovery algorithm, and outputs of EAP, EAP-IG, ACDC, and Edge-SP exhibit the same failure. Under resampling, KL misranks 9.4%-41.2% of candidate pairs across these methods on the human-reference tasks. We investigate context distortion as an explanation: replacing excluded signals changes the inputs on which retained components operate. Restoring selected signals from the recipient's intact-model execution repairs 96 of 100 persistent KL misrankings from the discovery pool on both validation and held-out prompts. The circuits and their original behavioral scores remain unchanged. These findings show why better discovery alone is insufficient when its objective rewards the wrong candidate. --- ### Scalable, Transferable Meta-network for Data Selection Requires a Different Loss (and Why the Obvious Choice is Problematic) Authors: Zilin Du, Bowen Yang, Boyang Albert Li Published: 2026-10-01 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.02092v1 Data selection is critical for training large language models on massive and heterogeneous corpora. Meta-learning for Training-data Selection offers a principled alternative to heuristic scoring by learning data weights from a target validation objective, but existing methods face a trade-off between fine-grained valuation and transferability to unseen data. A natural solution is to replace per-sample weights with a selection network. However, we find that directly incorporating such a network into existing MTS objectives leads to unstable optimization and poor generalization, caused by weight suppression and persistent reliance on easy-to-learn features. To address these issues, we propose Transferable Example Scoring and Selection (TESS), a scalable data-selection framework built on a Pointwise Value Matching objective (PVM). Experiments on LLM safety and targeted instruction tuning demonstrate strong transfer across datasets, from subsets to full corpora, and from smaller to larger models. --- ### GeoLatent: Geometry-Guided Latent Structuring with Routed Optimization for 3D Reasoning Authors: Yakun Zhu, Yi Bin, Yujuan Ding, Zheng Wang, Pengpeng Zeng, Duo Peng, Jingkuan Song, Heng Tao Shen Published: 2026-10-01 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.02091v1 Despite progress in vision-language models, 3D spatial reasoning from 2D images remains challenging. Text-based methods describe intermediate geometry with discrete tokens, limiting fidelity for continuous spatial relations. Continuous latents offer richer representations, but a single latent type does not explicitly separate the cues needed across spatial tasks. Decomposed spatial latents address this by representing position, direction, and global geometry separately under geometric supervision. Yet the geometry representation can still collapse toward one dominant direction, and unrestricted attention can leave the latents underused during answer learning. We introduce GeoLatent, combining Common--Residual Geometry Alignment (CR-GEO) with routed optimization to structure the geometry states while promoting latent-mediated answer learning. CR-GEO separates shared from residual teacher geometry; routed optimization jointly trains geometry and language, temporarily directs visual answer learning through the latents, and restores full attention with geometry supervision. In controlled comparisons, CR-GEO raises geometry effective rank from 1.00 to 3.87, while blocking latent readout at the bottleneck lowers direction accuracy from 89.1% to 25.8% on 128 fixed questions. After recovery, the differentiated geometry representation and latent-mediated visual route remain available alongside direct image access. GeoLatent achieves 73.0% on SPAR-Bench and 72.1% on SPBench, outperforming previously reported methods on both. --- ### HumanoidToolBench: Benchmarking Humanoid Tool Use from Selection to Mobile Execution Authors: Kyochul Jang, Seohyeon Park, Ohchul Kwon, Sangjun Park, Junhyeok Choi, Seungyeop Yi, Chaeyun Kim, Sangkyu Lee, Idan Szpektor, Avi Caciularu, Jongmin Park, Youngjae Yu Published: 2026-10-01 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2610.02089v1 As robotic hardware and learning methods advance, humanoids need tools to perform tasks beyond their inherent physical limits. Successful tool use requires selecting a suitable tool and coordinating manipulation and, when needed, locomotion to complete the task. Existing benchmarks do not jointly evaluate these capabilities on a humanoid. We introduce HumanoidToolBench, an 18-task benchmark spanning three scenarios, three execution levels, and two tool-set modes, together with ToolBook, a dataset of 3.1k demonstrations collected in simulation and on a real Unitree G1. Evaluation of seven policies in simulation and three on the real robot reveals substantial gaps between selecting a suitable tool and completing the task. Focused GR00T N1.7 probes show reduced selection accuracy on unseen tools and continued task execution under unrelated instructions. Code and data are available at https://snu-pi.github.io/HumanoidToolBench/. --- ### Kolmogorov-Arnold Networks for Free-Boundary Partial Differential Equations Authors: Tan Phuong Dong Le Published: 2026-10-01 Categories: math.NA, cs.LG arXiv: https://arxiv.org/abs/2610.02084v1 We study free-boundary problems within a physics-informed framework using Kolmogorov-Arnold network (KAN) approximations. The proposed approach incorporates obstacle constraints, partial differential equation (PDE) inequalities, complementarity conditions, and boundary conditions through residual-based loss functions. We consider a linear elliptic obstacle problem, a nonlinear $p$-Laplacian obstacle problem, and a time-dependent one-phase Stefan problem. The proposed KAN solver is compared with physics-informed neural network (PINN) and residual-network baselines. Numerical experiments show that KANs achieve low relative $L^2$ and $L^\infty$ errors while accurately resolving contact regions and moving interfaces. The results indicate that KAN representations provide an effective alternative for solving free-boundary PDEs. --- ### Wasserstein Gradient Flows and Forward-Only Diffusion Are Not Enough for Multimodal Sampling Authors: Daniel McBride, Pratik Khandagale, Cristina Garcia-Cardona, Yen Ting Lin Published: 2026-10-01 Categories: stat.ML, cs.LG, math-ph, math.DS, math.PR, math.SP arXiv: https://arxiv.org/abs/2610.02081v1 There has been a proliferation of sampling algorithms based on Wasserstein gradient flows (WGF) and forward-only diffusion processes (FODP), often accompanied by theoretical guarantees of exponentially fast convergence to the target distribution. These guarantees are frequently interpreted as evidence that such methods can efficiently sample complex multimodal distributions, often supported by empirical results. In this work, we argue that this interpretation is fundamentally misleading. By invoking the Jordan-Kinderlehrer-Otto (JKO) scheme and Otto calculus, we establish that the canonical WGF sampling dynamics and overdamped forward diffusion share the same density evolution and therefore inherit the same metastability and slow-mixing phenomena long understood in nonequilibrium statistical physics. We analyze this family of samplers using two complementary tools -- spectral analysis and mean first-passage time (MFPT) analysis -- and show that well-separated multimodality can induce exponentially long mixing times associated with small spectral gaps and rare inter-mode transitions. For the commonly adopted log-linear annealing schedule studied here, we find that introducing intermediate distributions does not remove the exponential scaling of the total transport time. The limitation is structural rather than implementation-specific: purely local, gradient-driven transport mechanisms can require exponentially long times to transport probability mass across well-separated modes. We argue that this represents a fundamental limitation of WGF- and FODP-based sampling in their standard forms, and motivates future development of fundamentally nonlocal mechanisms for efficient multimodal sampling. --- ### LLM2Jev: LLMs Are Already Jev-Style Decision Models -- When and How to Fine-Tune Them Authors: Yinheng Li, Justin Wagle Published: 2026-10-01 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.02076v1 Jev-style decision models return categorical probability distributions over predefined options without generating free-form text, enabling software systems to act on their outputs directly. In this work, we investigate the extent to which general-purpose LLMs already possess this capability out of the box, and when fine-tuning is actually necessary. We present LLM2Jev, an architecture-preserving framework that extracts calibrated decisions directly from next-token probabilities over bracketed numeric identifiers. LLM2Jev provides both a training-free inference recipe and a fine-tuning objective that optimizes candidate selection via a tree-factorized listwise loss while anchoring auxiliary predictions to the base model using KL divergence penalties. Evaluating on Qwen3.5-4B and Qwen3-0.6B, we find that modern LLMs are inherently effective decision models: without training, the 4B model matches community Jev-style models built on the same backbone, outperforms letter-logit readouts, supports arbitrary option counts, and natively handles multimodal decisions over images. Fine-tuning provides targeted rather than universal benefits -- substantially improving weaker models and specific tasks (such as many-option intent routing), but offering diminishing returns for strong backbones. Crucially, our KL anchors prevent behavioral degradation in conversational text generation, with LoRA delivering the strongest performance on capable models. --- ### Homomorphic Advantage Operator: Stabilizing Reinforcement Learning Under Fully Homomorphic Encryption Constraints Authors: Abid Mohamed Nadhir, Ahmad Al Hanbali, Beggas Mounir Published: 2026-10-01 Categories: cs.AI, cs.CR arXiv: https://arxiv.org/abs/2610.02074v1 Privacy-preserving machine learning presents significant deployment challenges on the cloud for intelligent systems with confidential data. Fully Homomorphic Encryption (FHE) offers a compelling solution for secure computation, preserving data confidentiality of cloud computations. However, applying FHE to reinforcement learning (RL) requires replacing non-linear operations with polynomial approximations, which diverge catastrophically due to a unique recursive error phenomenon known as the Bellman drift. This article introduces the Homomorphic Advantage Operator (HAO), a stabilization framework designed to prevent polynomial approximation divergence in FHE-based deep RL. HAO adapts the zero-mean centering projection from advantage-based value estimation directly to temporal-difference (TD) targets. This linear projection annihilates the uniform state-value baseline that drives the Bellman drift, maintaining per-state action rankings while requiring zero additional non-linear multiplicative depth and avoiding expensive ciphertext bootstrapping. The proposed HAO framework was evaluated using a three-tier experimental methodology, including a tabular Markov Decision Process (MDP), an encrypted CartPole environment using real CKKS cryptographic operations, and a 20-node logistics routing benchmark with dense continuous features. The results demonstrate that the proposed HAO strictly bounds network pre-activations within the safe polynomial approximation domain. The proposed HAO RL agents achieved 0% boundary breaches across all random seeds used, whereas regularization alone (L2 weight decay and gradient clipping) breached the bound on 3 of 5 seeds and the unstabilized baseline did so in 83.8% of episodes. Finally, HAO agents improve optimal policy accuracy by 18.0 percentage points in tabular domains and remain stable when DP-SGD-style Gaussian noise is added to the clipped gradients. --- ### PyPottery: an AI-powered end-to-end suite for pottery processing and publication Authors: Lorenzo Cardarelli Published: 2026-10-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.02072v1 The study of ceramic materials constitutes a cornerstone of archaeological research, yet the post-production workflow for pottery documentation remains labor-intensive and creates significant publication bottlenecks. This paper presents PyPottery, an open-source, AI-powered suite designed to semi-automate the complete ceramic documentation pipeline. The suite comprises four integrated modules: PyPotteryScan for automated image extraction and handwriting recognition; PyPotteryInk for automatic inking of pencil drawings; PyPotteryTrace for semantically-aware vectorization; and PyPotteryLayout for automated layout generation. Evaluated on 50 hand-drawn sheets containing 240 pottery drawings from the Terramara di Montale (Italy), the framework achieved substantial time savings confirmed by usability study participants, who reported a median perceived speedup of 40$\times$ over traditional workflows (range: 17.5$\times$--120$\times$). These results highlight the potential of AI-assisted tools in archaeological documentation, while the paper addresses the strategic redistribution of cognitive labor toward augmentation rather than automation. --- ### Causal Memory Policy: Making Memory Utility Identifiable by Intervening on Retrieval Authors: Arman Behnam, Binghui Wang Published: 2026-10-01 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.02070v1 Memory-augmented large language models must decide which memories to retain, and recent systems do so by estimating each memory's effect on task performance. However, these estimates rely entirely on retrieved memories. When a memory is never retrieved, store-level interventions produce identical outcomes, leaving its utility unidentified. This is a retrieval-level positivity violation, invisible to diagnostics that examine only memory operations. We introduce Causal Memory Policy (CMP), a causal framework that restores identification by intervening on retrieval itself, reserving a fixed number of context slots for memories sampled with known propensities. CMP estimates memory utility by self-normalized inverse propensity weighting under a balanced assignment design. We prove the causal factorization of memory utility through retrieval, the unbiasedness and exact variance of the estimator, and the optimal decision rule under irreversible operations. Empirically, identification fails for 54% of required memories on LongMemEval and 67% on LoCoMo, and the failure persists in a deployed memory system. CMP improves discrimination between required and non-required memories from 0.54 to 0.66 AUC. Finally, we show that identified memory utility alone is insufficient for retention decisions: per-query utility reaches 0.78 AUC on the query for which it is estimated, yet no aggregation available to a retention policy predicts a memory's value on unseen queries. Code is available at: https://anonymous.4open.science/r/cmp-release-D0C3/. --- ### AI Emulation of Stochastic Sudden Stratospheric Warming with Interpretable Latent Structure Authors: C. Daniel Boscu, Daniel Hernandez, Fabio Alvarez Ventura, Justin Finkel, Ashesh Chattopadhyay, Pedram Hassanzadeh, Dorian S. Abbot Published: 2026-10-01 Categories: physics.ao-ph, cs.LG arXiv: https://arxiv.org/abs/2610.02069v1 Rare weather regime transitions pose a challenge for data-driven modeling due to class imbalance. In this study, we develop a probabilistic deep learning emulator for a prototypical system with regime transitions, the stochastic Holton--Mass model of stratospheric variability, and analyze the structure of its learned latent space. The Holton--Mass model exhibits two metastable regimes, a strong and a weak polar vortex, maintained by nonlinear wave--mean flow interactions, with weak stochastic forcing intermittently triggering rare transitions between these regimes that qualitatively represent SSW events. We employ a ResNet-inspired Conditional Variational Autoencoder with six-layer encoder and decoder layers and explicit current-state conditioning to model the distribution of the system's state at the next time step (one day). The emulator accurately reproduces short-term dynamics, steady-state probability distributions, regime persistence statistics, rare transition rates, the transition committor function, and the transition expected lead time of the physical model. Beyond emulation fidelity, we interrogate the learned latent representation to understand how the model internalizes the underlying metastable structure of the dynamics. Principal Component Analysis of the 32-dimensional latent space reveals a clear and unsupervised separation into four physically interpretable clusters corresponding to strong versus weak vortex regimes and stable versus transition-prone configurations. Such emergent regime separation in latent space is hard to identify for deep generative models applied to high-dimensional stochastic systems. Our results show that carefully designed probabilistic emulators can uncover physically meaningful manifolds governing extreme-event dynamics, potentially aiding the development of improved operational advanced warning systems. ---