# Research Papers — 2026-09 · bundle page 48 of 266 > Full-text markdown of papers #2351–2400 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/48 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/47.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/49.md) ### MASCIT: A Mask-Aware State Space Classifier for Naturally Irregular Time Series Authors: Yoo-Min Jung, Hyeon-Gi Kim, Jonghun Park Published: 2026-09-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.34409v1 Naturally irregular time series combine asynchronous observations, missing values, unequal lengths, and nonuniform sampling, while dense adapters can discard temporal structure. We propose a mask-aware state space classifier for irregular time series (MASCIT), which supplies observation masks to the encoder and excludes invalid steps from gated temporal aggregation. Across 34 irregular time series datasets, MASCIT yielded the strongest aggregate point estimate and was the only evaluated neural model with three-seed results on every dataset. MASCIT retained the lowest point rank across six overlapping irregularity indicators, while factorial ablations favored partial over full selectivity. These results support selective state space models as effective, executable backbones for naturally irregular time series classification. --- ### Distilling Visual Reasoning into Text Space Authors: Wenhan Yang, Nilay Naharas, Ali Payani, Baharan Mirzasoleiman Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34408v1 Large Vision-Language Models (LVLMs) have shown strong promise for multimodal reasoning, yet often struggle with tasks requiring concepts beyond what is directly observable in the input image. Existing methods generate intermediate images or latent visual tokens to guide reasoning, but these representations can introduce errors and increasingly interfere with textual reasoning as reasoning progresses. We propose Visual-to-Text Chain-of-Thought Distillation (V2T), a framework that enables LVLMs to internalize visual reasoning without generating intermediate visual representations at inference time. V2T first trains a teacher LVLM using interleaved visual and textual chains of thought, and then uses knowledge distillation to train a student LVLM using the teacher's logits and cross-entropy supervision from ground-truth textual reasoning. When reasoning images can be mapped to the original image, V2T can additionally distill the teacher's attention to corresponding regions, while ground-truth bounding boxes can further guide a subsequent reinforcement learning stage. Experiments across multiple multimodal reasoning benchmarks show that V2T consistently outperforms the teacher and existing baselines, improving average accuracy by 14.3% on a held-out set and 2.7% on the broader visual evaluation suite. Moreover, lightweight SFT and substantially reduced RL make V2T up to 42x faster to train than state-of-the-art baselines. --- ### Unlocking Few-Step Diffusion for Faithful Previews Authors: Jing Jia, Sifan Liu, Guanyang Wang Published: 2026-09-28 Categories: cs.LG, cs.CV, stat.ML arXiv: https://arxiv.org/abs/2609.34406v1 Sampling latency compounds in diffusion workflows, where users generate and discard many candidates before keeping one. Surprisingly, the poor outputs of standard few-step samplers do not reflect a lack of reconstruction capacity: by optimizing only the initial noise, frozen 3-4-step samplers can closely reproduce their corresponding full-step outputs. Building on this finding, we learn corrections to the initial noise and denoising updates using endpoint supervision, improving correspondence with full-step outputs generated from the same noise and prompt. The resulting previews allow users to screen candidates cheaply and reserve full-step generation for promising ones. Input correction also transfers across sampling budgets without retraining. Experiments show substantial improvements in reference fidelity, including 53-78% lower reconstruction MSE than retrained LD3 on unconditional benchmarks, alongside improved ranking preservation and candidate selection on SD1.5, SDXL, and FLUX.1-dev. --- ### Eval4DiRec: A Unified and Systematic Evaluation Framework for Diffusion-based Recommender Systems Authors: Cong Wang, Shoujin Wang, Yishuo Li, Qi Zhang, Liang Hu, Wenpeng Lu Published: 2026-09-28 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2609.34404v1 Leveraging the strong generative capabilities and stable training dynamics of diffusion models, diffusion-based recommender systems (RSs) have recently emerged as a novel recommendation paradigm, attracting increasing attention from both academia and industry. However, despite the rapid growth of diffusion-based RSs, a critical issue has emerged: the lack of a unified and systematic quantitative evaluation benchmark, which often results in irreproducible experimental results and unfair comparisons across studies due to inconsistent data processing, training configurations, inference procedures, and evaluation protocols. To address this challenge, we propose Eval4DiRec, the first unified and open-source evaluation framework specifically designed for diffusion-based RSs. Eval4DiRec supports 14 representative diffusion-based RS models across five different recommendation scenarios, providing consistent and reproducible experimental settings to systematically assess their performance. Built upon this framework, we conduct extensive empirical studies to benchmark these models under unified protocols. The results highlight the strong potential of diffusion models for recommendation while also revealing key factors and practical challenges that substantially affect their performance, thereby establishing a solid foundation to facilitate fair evaluation and guide future research in this promising field. Our code and data are available at: https://github.com/wangcong2001/Eval4DiRec. --- ### GeoCFM: Positive-Only Conditional Flow Matching for Mineral Occurrence Sampling Authors: Moshe Eliasof, Eldad Haber Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34398v1 Critical mineral discovery is a positive-only problem: deposits are observed as sparse locations, while unlabeled regions are not reliable negatives, and similar geophysical signatures can arise from different subsurface states. We therefore model mineral targeting as learning a conditional spatial distribution over occurrence locations, $π(p\mid d)$, given geo-images $d$, rather than predicting a deterministic per-pixel score map. We introduce GeoCFM, a conditional flow-matching model that generates mineral occurrence point sets conditioned on multi-channel geo-images; GeoCFM learns a point-wise transport field in $\mathbb{R}^2$, using UNet features with point-conditioned velocity prediction to bridge dense rasters and sparse supervision without pseudo-negatives. On a synthetic magnetics--geochemistry benchmark with latent activation and on USGS Earth MRI data with a spatially disjoint tile split, GeoCFM improves geometric agreement with observed occurrences over score-map and non-conditional baselines, while representing epistemic uncertainty through conditional sampling. --- ### SkillFocus: Evolving Agent Skills via Capability Decomposition Authors: Ning Wang, Zhiren Gong, Bingdong Li, Peng Yang, Aimin Zhou Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34397v1 Agent skill evolution seeks to improve reusable procedural guidance for large language model (LLM) agents through iterative revision. Existing methods base each revision mainly on execution trajectories or feedback, leaving recurring behavioral requirements across tasks implicit and tying revision to the behavior of the current skill. We introduce SkillFocus, which decomposes recurring task requirements into a capability space that remains fixed as the skill evolves, separating what tasks require from how the current skill behaves. SkillFocus maps current task outcomes to this space to identify the capability that leaves the most tasks unresolved, then uses that capability to determine what to revise and which evidence to use. Across four benchmarks spanning heterogeneous tasks, SkillFocus achieves the best held-out accuracy on all four, outperforming the strongest competing result by 5.7 points on average while using 24\% fewer evolution tokens on average than the closest iterative baseline. Controlled studies further show that capabilities derived from recurring task requirements outperform task-semantic and execution-derived alternatives, while randomizing task--capability assignments reduces final accuracy by up to 20.2 points. Matching evidence to the selected capability increases candidate gain by 4.4 points under prioritized revision. --- ### HyperDAM: Hyperspectral Distractor-Aware Memory with Amodal Expansion for SAM 3 Tracking Authors: Ryoga Yuzawa, Tasuku Takagi Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34396v1 Hyperspectral video provides material cues that can disambiguate targets with similar false-color appearance, yet foundation-model trackers update memory primarily from spatial and appearance evidence. We present HyperDAM, a DAM4SAM3-based hyperspectral tracker with three principal contributions. First, HOTC2026-Modal adds human-verified frame-wise modal masks and mask-tight boxes to all 481 organizer-provided HOTC 2026 videos. Second, a frame-zero-calibrated HSI gate rejects spectrally inconsistent updates to the distractor-resolving memory (DRM) without altering the current prediction. Third, a causal spatiotemporal expander adds outward-only amodal corrections from frozen SAM features. Static-scene recovery and empty-mask RTS smoothing address target switches and full occlusion. Model selection prioritizes cross-domain robustness over leaderboard-specific optimization. The final system ranked second in HOTC 2026, achieving 68.0093% AUC and 87.7703% DP@20 in the organizer's private evaluation. --- ### Org-Agent: Beyond Personal Assistants Towards Organizational Agents Authors: Luyao Zhuang, Yujing Zhang, Zijin Hong, Yilin Xiao, Xiao Huang Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34392v1 Language model agents serving organizations must coordinate requests from multiple users while using knowledge distributed across their interactions. We identify two complementary capabilities for this setting, namely cross-user interaction and decision-making, as well as cross-user memory and knowledge use. Both capabilities are governed by organizational constraints across three aspects: user identity, authority, and access permissions; the attribution and temporal validity of information; and rules for resolving conflicting requirements across users and completion requirements for joint decisions. These constraints shape what information or decisions must be obtained before an action can proceed and what conditions must be satisfied during its execution. Motivated by this, we introduce Org-Agent, a unified constraint-centric reasoning framework that organizes task execution in three stages. Specifically, Org-Agent decomposes a task into atomic subtasks and constructs a task dependency graph whose edges encode the dependencies among them. Building on this graph, it schedules the subtasks in dependency order through topological sorting. It then executes each subtask while accounting for the task's constraints, supported by evidence-acquisition and memory-management tools. Experiments on MUSES-Bench and GroupMemBench demonstrate the effectiveness of Org-Agent on both capabilities, and ablations further support the contributions of dependency modeling and tool use. --- ### P2P: Cross-View Population Denoising for Unpaired Single-Cell Perturbation Response Prediction Authors: Haojie Yang, Ran Su Published: 2026-09-28 Categories: cs.ET, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.34391v1 AIVC (AI Virtual Cell) is a learned simulator of cellular behavior across conditions. Predicting how a cell population responds transcriptionally to a genetic perturbation is a core task. Perturb-seq records that response by destructive sequencing, so a control cell and a perturbed cell are never observed as a pair, and cells under one condition remain heterogeneous and noisy. Regression on individual cells absorbs sampling variation into the estimated effect, whereas interpretation requires the reproducible population effect. P2P (Perturbation-to-Perturbation) takes a stochastic cell-set view as its supervision unit. Two views drawn from the same condition share a reproducible population effect and differ by view-specific variation. A permutation-invariant set encoder summarizes the control population, a structured encoder represents perturbation tokens, cellular context, dose, and combination interactions, and a gate blends empirical condition-effect memory with a neural residual. A heteroscedastic head predicts the population mean and gene-wise response variance. Under one protocol and five seeds, P2P attains the lowest expression RMSE and the highest Effect Pearson, DEG F1, and DEG average precision on each of Adamson, Norman, Replogle K562, and Replogle RPE1 relative to GenePert, LinearPert, SLIM, Scouter, and scPILOT. On Replogle K562, Effect Pearson rises from 0.643 to 0.702 and DEG F1 rises from 0.067 to 0.178 relative to Scouter, the strongest baseline on both metrics. --- ### VastMAT: A Large-Scale Multi-Category Benchmark for Multi-Animal Tracking Authors: Zhizhen Li, Zan Wang, Huidong Peng, Bohan Tan, Shimin Shan, Yu Liu, Liang Peng Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34390v1 Multi-animal tracking (MAT) supports the study of animal movement, behavior, and group interactions. However, general multi-object tracking (MOT) benchmarks primarily focus on pedestrians and vehicles, whereas dedicated MAT benchmarks remain limited in jointly supporting broad animal coverage, large-scale video data, and extensive within-video multi-instance association. To address this gap, we introduce VastMAT, which has four key characteristics: (1) Large scale. It comprises 2,947 videos with 1,002,562 annotated frames, totaling 27.85 hours. (2) Broad category coverage. These videos cover 337 animal categories with diverse morphologies and motion patterns. (3) Extensive instance annotations. It provides 3,663,248 bounding boxes and 22,883 identity trajectories---to our knowledge, the largest numbers of both among dedicated MAT benchmarks. (4) High-quality annotations. To ensure reliability, annotations undergo iterative expert review and correction, and quality is assessed through an independent reannotation audit. To systematically assess tracking performance and cross-category generalization, we establish Seen-category and category-disjoint Unseen-category protocols, and evaluate eight representative MOT methods under both protocols. Under these protocols, the highest baseline HOTA scores are 66.37\% and 52.90\%, respectively, highlighting the challenge of tracking unseen animals. To address the low-overlap association challenge revealed by our analysis, we propose Center-Distance-Augmented Association (CDA), a lightweight module that adaptively combines IoU with center similarity normalized by the boxes' own scales. Without additional training, CDA improves TrackTrack's HOTA by 1.58 and 1.31 percentage points under the two protocols, respectively. To facilitate further MAT research, we will publicly release our benchmark and code. --- ### Reciprocal Guidance: Orchestrating Draft and Verify Budgets for Advancing the Diffusion-AR Self-Speculation Frontier Authors: Linye Wei, Shutian Zheng, Haoyu Zeng, Meng Li Published: 2026-09-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.34388v1 Diffusion drafting with autoregressive (AR) verification has emerged as a promising paradigm for efficient speculative decoding. Recent self-speculation models, represented by Nemotron-Labs-Diffusion, further simplify the speculative pipeline by unifying drafting and verification within a shared backbone, while enabling longer acceptance lengths. However, the Pareto frontier between aggregate and per-request throughput remains underexplored. At low concurrency, sequential draft-verify execution requires two model forward passes per round, limiting the effective tokens per forward (TPF). By contrast, at high concurrency, longer drafts incur increasingly expensive computation, forcing individual requests to operate under constrained speculation budgets and preventing full exploitation of the full-backbone drafter. Our key observation indicates that drafting and verification exhibit reciprocal predictability. Draft logits can anticipate likely verification mismatches, while recent verification outcomes predict future drafting utility and suitable block sizes. Building on this observation, we introduce Reciprocal Guidance (RecGuide), a runtime draft-verify orchestration framework that adapts speculative decoding to varying serving loads. RecGuide exploits spare compute capacity through verification-overlapped drafting at low concurrency, while dynamically allocating request-specific draft block sizes as the workload becomes increasingly compute-intensive. Experiments across a wide range of concurrency levels demonstrate consistent throughput improvements over vanilla self-speculation, achieving up to $1.8\times$ speedup. --- ### CAR-VLA: Complexity-Aware and Risk-Adaptive Reasoning for Autonomous Driving Authors: Xiaolei Chen, Zhuolin He, Yuxuan Liang, Xu Li, Haotian Chen, Fan Shi, Mengyang Zhao, Wenjuan Meng, Zisheng Chen, Zhihao Zhu, Zhounan Jin, Hengli Wang et al. Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34387v2 Existing adaptive reasoning methods for driving Vision-Language-Action (VLA) models primarily focus on whether to reason, overlooking how reasoning should differ across driving situations. Our key insight is that while scene complexity informs reasoning depth, dynamic risk is equally critical for deciding how to reason in time-critical situations. We therefore propose CAR-VLA, a unified driving VLA model that jointly considers scene complexity and dynamic risk to guide reasoning depth, urgency, and focus. CAR-VLA maps four complexity--risk categories to three reasoning modes: \textit{Fast Intuition} for direct trajectory generation in simple low-risk scenes, \textit{Slow Thinking} for deliberate reasoning in complex low-risk scenes, and \textit{Reflex Response} for compact, hazard-focused reasoning in high-risk scenes regardless of complexity. Rather than merely shortening deliberation, Reflex Response centers reasoning on the most critical hazard and the immediate safe response. We train CAR-VLA through progressive supervised learning that links scene assessment, reasoning-mode selection, and trajectory generation, followed by reasoning-augmented reinforcement learning to improve driving quality and reasoning behavior. Experiments on NAVSIM v1(91.1 PDMS), NAVSIM v2(90.3 EPDMS), and Navhard(35.0 EPDMS) demonstrate competitive driving performance. Qualitative comparisons on navtest and in-house high-risk scenarios further illustrate risk-aware reasoning and hazard-responsive trajectory generation. The code for this paper will be released publicly at: https://github.com/chenxl124578/CAR-VLA.git --- ### Look Before You Select: Rethinking Vocabulary Sparsification in On-Policy Distillation Authors: Yongliang Miao, Shuang Liu, Yanguang Liu, Yandong Bai, Mengnan Du Published: 2026-09-28 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.34386v1 On-policy distillation (OPD) uses teacher correction on student-generated responses. Full-vocabulary correction can provide important corrections even for tokens that the student assigns low probability, but backpropagating through all token logits becomes memory-intensive for long sequences. Existing memory-saving approaches estimate corrections from sampled tokens or restrict supervision to the student's TopK tokens, introducing sampling noise or changing the full-vocabulary correction. We introduce \textbf{SparseOPD}, which uses full-vocabulary teacher correction to determine which corrections matter before selecting the token logits to differentiate. SparseOPD first constructs the full-vocabulary correction without retaining its backward graph, then selects tokens by correction magnitude rather than student probability. Signed residual compensation preserves the total promoting and suppressing correction mass, while correction-aware budget allocation distributes the sparse support across positions. Finally, the update backpropagates only through the selected token logits. Across six task--scale settings spanning mathematics, chemistry QA, and multimodal reasoning, SparseOPD outperforms Sampled Token and TopK in task-average accuracy and matches or exceeds Full Vocabulary. Gradient cosine similarity reaches 99\% on 4B mathematics, while 8K full-parameter profiling shows 70.5\% lower backward memory. --- ### Just-In-Time Agent Memory with Runtime Agentic Research Authors: Bingyu Yan, Chaofan Li, Hongjin Qian, Shuqi Lu, Chaozhuo Li, Zheng Liu Published: 2026-09-28 Categories: cs.CL, cs.AI, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2609.34385v1 Memory is critical for AI agents. Many existing agent-memory systems follow an Ahead-of-Time (AOT) design, constructing memory before a specific request arrives. While this reduces online serving cost, such request-agnostic memory construction can discard fine-grained information that later becomes important. To address this limitation, we propose Just-In-Time Agent Memory (JAM), a trainable framework for query-conditioned context construction at runtime. A Memorizer preserves complete raw histories in a hierarchical page-store with compact navigational summaries, while a Researcher iteratively retrieves, inspects, and integrates evidence for each request. To train these memory-use behaviors, we introduce Memory-Gym, an evidence-grounded data synthesis pipeline covering nine task types across six domains, and optimize the Researcher through verified-trajectory supervised fine-tuning followed by Hint-guided Group Relative Policy Optimization. We demonstrate the effectiveness of JAM across a variety of benchmarks on agent memory and long-context processing, where it achieves stronger task performance than AOT-style memory systems while remaining substantially more efficient than prior trained agentic memory approaches. To support reproducibility and future research, we release our anonymized source code at https://github.com/VectorSpaceLab/general-agentic-memory. --- ### RoboIRGBench: Benchmarking Implicit Referential Grounding in Vision-Language-Action Models Authors: Aernaer Akelijiang, Jiannan Li, Zhineng Chen, Jingjing Chen, Bin Zhu Published: 2026-09-28 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.34384v1 Vision-Language-Action (VLA) models have shown strong capabilities in robotic manipulation, yet existing benchmarks typically assume that task-relevant information is explicitly specified in the instruction. In practice, however, humans frequently refer to objects, quantities, and relations implicitly, requiring robots to recover the intended target from linguistic and perceptual context. We study this capability as Implicit Referential Grounding (IRG) and introduce RoboIRG-Bench, a manipulation benchmark designed to systematically evaluate it. Built upon RoboMME, RoboIRG-Bench contains 40 variants derived from 11 tasks and covers four challenges, including direct, reasoning-mediated, spatial, and contextual referential grounding. As IRG often requires retaining and retrieving previously established context, we evaluate representative VLAs spanning different memory mechanisms. Our evaluation reveals a noticeable referential robustness gap. Models that perform well under explicit instructions can degrade sharply when the same task-relevant information must be recovered from context. Reasoning-mediated and spatial references are particularly challenging, while models using external VLMs show greater robustness but still exhibit significant failures. Moreover, replacing the external VLM with a stronger model does not eliminate these gaps. We further validate these findings on a Franka Research 3 robot arm, where the gap persists under real-world manipulation and manifests as both incorrect referent grounding and downstream execution failures. These results establish IRG as a distinct and underexplored capability for reliable robotic instruction following and highlight the need for VLAs that can robustly integrate language, perception, reasoning, and action. --- ### Joint and Cross-Modal Video-Audio Generation and Editing: A Unified Formulation and Design Taxonomy Authors: Abhinav Sharma, Sai Karthik Navuluru, Wang Wei, Daksh Dangi, Xiangbo Gao, Li Li, Bo Ni, Vardhan Dongre, Junda Wu, Xiyang Hu, Jiuxiang Gu, Seunghyun Yoon et al. Published: 2026-09-28 Categories: cs.CV, cs.MM, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2609.34381v1 Video and audio are perceived together, yet most generative models treat them in isolation. We examine methods that model the two modalities jointly, generate one from the other, or edit them in a coupled manner, organized around a single question: how is the output kept coherent across modalities in time and semantics? A unified formulation casts joint generation, cross-modal generation, and joint editing as three problems defined on a single distribution over audio-visual pairs, and a taxonomy compares methods along five design axes. To our knowledge, this is the first overview to systematically taxonomize joint audio-visual editing, which we map as nine edit categories spanning 28 edit types. We describe methods, datasets, and metrics for each setting and close with the open problems we view as most consequential. --- ### DPS: Dual-Mode Precision LLM Serving with Semi-Unified Memory Authors: Xuan Truong Nguyen, Tien Son Pham, Tuan Duc Chu, Wookeun Jung, Thanh Tuan Dao Published: 2026-09-28 Categories: cs.DC, cs.AI, cs.ET, cs.PF, cs.SE arXiv: https://arxiv.org/abs/2609.34380v1 Existing LLM serving systems virtualize and optimize KV-cache memory, but treat model-weight memory as fixed throughout execution. Recent work on multi-precision model representations challenges this design by allowing a single stored model to support both full-accuracy and lower-precision execution, making the effective weight footprint runtime-dependent. This creates an opportunity under bursty workloads, where temporary spikes in KV-cache demand often determine throughput and SLO compliance. We present DPS, a dual-precision LLM serving system that turns weight memory into an elastic resource: under normal load, DPS serves the full-accuracy model; under KV pressure, it switches to a nested, lower-precision variant and repurposes unused weight memory for KV cache blocks. DPS is built on Semi-Unified Memory (SUM), which partitions the weight region into a persistent lower-precision sub-region and a shared region that alternates between residual weight tensors and KV-cache blocks, preserving compatibility with paged KV-cache management. We implement DPS on top of vLLM and evaluate it across both dense and MoE models and various production workload traces. Our results show that \sysname improves sustained throughput by $2.1$--$3.3\times$ and effective pass@1 by up to $+41$\,pp over Static FP16, while preserving FP16-class accuracy. --- ### Marathoner: Ultra-Long-Horizon Autonomous Intelligence Authors: Zhang Ruiyang, Ou Jinpeng, Xie Yifan, Zhou Jingang, Pan Lirui, Guo Qingpei, Zheng Zhedong Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34378v1 Humans naturally possess the ability to work persistently toward long-term goals. Given a challenging task, humans can continuously work for months or even years to accomplish a specific objective. In this paper, we propose Marathoner, an autonomous agentic model possessing the ability of ultra-long-horizon execution. Specifically, we propose a comprehensive post-training pipeline to instill this critical capability into base model. For Ultra-Long-Horizon Task Synthesis, we leverage major release PRs containing 1000+ lines of new code from diverse GitHub repositories as the primary source for synthesizing challenging task-level data. Additionally, we introduce Multi-Task Chaining, which chains multiple generated tasks into a single more challenging task, enabling the synthesis of tasks with frontier-level difficulty. For rejection sampling finetuning, we combine strong teacher model with diverse harnesses to generate trajectories on our synthesized tasks and conduct supervised finetuning on base model with rejection sampled trajectories. For reinforcement learning, cold-started model performs real-world execution through harnesses in independent sandboxes during rollout process, effectively facilitating the acquisition of genuine ultra-long-horizon execution capability. We further propose a novel reward strategy, Later Stage Bonus Reward, which explicitly encourages model to perform meaningful maneuvers during later stages of execution. Through extensive evaluation on 5 benchmarks containing ultra-long-horizon tasks, Marathoner achieves consistent and substantial performance improvements over base model and even surpasses performance of strong proprietary model. Further analysis shows that Marathoner can consistently work for 10+ hours and conduct 1000+ tool calls on highly challenging tasks. --- ### Before Agents Act: Assurance-Aware Semantic Scheduling for Evidence Acquisition in Distributed Systems Authors: Jun He, Deying Yu Published: 2026-09-28 Categories: cs.DC, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2609.34376v1 Tool-using agents can initiate consequential infrastructure changes, yet evidence required for admission may expire while other checks run or depend on a shared fault domain. We formulate evidence acquisition as joint witness selection and scheduling under quorum, diversity, freshness, deadline, and resource constraints. Assurance-Aware Semantic Scheduling (AAS) combines integer-program selection, dispatch-aware temporal scheduling, bounded diagnostic expansion, and receipt-aware repair. Formal results state the assumptions needed for dispatch-time freshness and finite diagnostic expansion. In three generated infrastructure workloads, AAS produces 1,075/1,200 valid candidates versus 647/1,200 for constraint-aware forward scheduling; stale candidates fall from 440 to 12. Paired sensitivity studies reuse the same instances and operation latency draws across parameter settings. A corrected timeout intervention finds 18/20 admissions with repair or full resynthesis versus 0/20 for a static plan, with lower committed cost when receipts are reused. On 20 constructed cases requiring a certified decomposition cut, refinement recovers an oracle-matching feasible plan every time. These are controlled simulation results; the bounded oracle shares a temporal search component, and transfer to deployed systems remains untested. --- ### LRC-JEPA: Disentangling Dynamics and Residual Context for Efficient World Models Authors: Luzhe Huang, Lei Chu, Jingyi Liang, Yuhuan Zhao Published: 2026-09-28 Categories: cs.LG, cs.AI, cs.RO arXiv: https://arxiv.org/abs/2609.34375v1 Compact JEPA world models enable efficient latent-space planning, but low-dimensional representation trained under reward-free self-supervision must encode both action-conditioned dynamics and predictable visual context. This competition can entangle controllable state with high-rank nuisance appearance and degrade planning as scenes become more complex. We introduce LRC-JEPA, a lightweight end-to-end world model that routes information into a compact predictive latent $\mathbf{z}$ and learned-query residual-context embeddings $\mathbf{u}$. Only $\mathbf{z}$ is propagated by the dynamics model and used for planning, while $\mathbf{u}$ captures temporally persistent information for cross-attention reconstruction; a differentiable residual connection encourages the latent to retain complementary dynamic content. Under explicit assumptions, we show that the resulting representation is sufficient, minimal, nuisance-invariant, and disentangled. Across four simulated control environments, LRC-JEPA improves average planning success over a parameter-matched JEPA baseline by 9 percentage points and matches or exceeds substantially larger pretrained models. On the real-world Bridge-v2 set, its 5.5M-parameter active encoder outperforms DINO-WM (22.1M) and V-JEPA2 (303.9M) encoders while also enabling faster planning. Physical-state probes, reconstruction interventions, and ablations confirm the effectiveness of LRC-JEPA's representation disentanglement. --- ### ABC-Align: Prediction-Powered Alignment with Adaptive Bias Control Authors: Eric Frankel, Banghua Zhu, Sewoong Oh, Lillian J. Ratliff Published: 2026-09-28 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.34374v1 Language model post-training is often bottlenecked by the need for human-collected preference data, which is expensive and difficult to scale. Reinforcement learning from AI feedback (RLAIF) style approaches that leverage pseudo labels offer an abundant alternative but introduce systematic biases that degrade downstream alignment. Recent general-purpose semi-supervised methods correct for teacher bias using a small set of human-labeled examples, but suffer from high variance especially when human annotations are scarce. To this end, we propose ABC-Align, leveraging abundant pseudo label signal to minimize variance and applying a lightweight, adaptive correction grounded in the human-labeled subset. The correction strength is tuned automatically during training using plug-in estimates of the relevant bias--variance quantities. On LLM alignment with RLHF, DPO, and GRPO where human feedback is scarce, we empirically demonstrate that ABC-Align achieves superior performance over prior semi-supervised baselines in a series of experiments on an increasing scale. Our code is available at https://github.com/SewoongLab/abc-align . --- ### Emergence, Not Bandwidth: Physical Coupling and the Limits of Learned Multi-Agent Communication Authors: Mihir Chauhan, Aniket Bera Published: 2026-09-28 Categories: cs.LG, cs.MA, cs.RO arXiv: https://arxiv.org/abs/2609.34373v2 Rate-limited multi-agent teams raise three questions the emergent-communication literature has answered only empirically: what an optimal message should encode, what compression costs over a horizon, and when a learned protocol is unique enough for a teammate to read. We answer them for rate-limited Dec-POMDPs, then measure how far reinforcement learning falls short of the optimum. Our theorems fix what is achievable independently of any learner, so a gap between an engineered and a learned sender at the same bit budget is an optimization fact, not an information-theoretic one. We instantiate this on three MuJoCo arenas spanning zero, partial and rigid physical coupling, charging every condition exactly 2 bits per decision, and create the discriminating regime by closing a physical side channel within one arena, holding bodies, task and reward fixed. Communication value is governed by coupling: under rigid coupling through a shared object, no channel beats silence (+0.001 +/- 0.001, p = 0.982, n = 25), since proprioception already carries that information; without coupling, every condition solves the task; under partial coupling, the engineered 2-bit sender reaches an interquartile mean of 1.000 but the learned one reaches 0.482, indistinguishable from silence (p = 0.400, n = 25). With a shared alphabet, bandwidth cannot explain the gap. Warm-starting from an engineered receiver localizes the failure: the same channel reaches 0.857 versus 0.562 cold-started (p < 0.001), so it is neither representational nor one of maintenance; reinforcement learning fails to discover the protocol. Cross-play shows learned protocols are individually meaningful but mutually unintelligible: self-play 0.980 collapses to 0.144 across seeds, and our best constructed alignment leaves at least 77% of that gap. All headline results use 25 seeds per arena and seven published baselines at matched rate. --- ### PersMem: Internalizing Personality into Dual-Pathway Memory for LLM Agents Authors: Hanzhong Zhang, Ziwei Xiang, Weicheng Xie, Shizhe Liu, Siyang Song Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34372v1 The profile of a role-playing agent usually depends on the pre-defined personality in a system prompt, whereas its memory processing pipeline, including prioritisation of stored memories and subsequent retrieval, remains independent of this personality. This separation causes the agent's memory processing to be inconsistent with the pre-defined personality, and makes it difficult to validate whether agent behaviours follow this personality. In this paper, we propose Personality-Integrated Memory (PersMem), which integrates personality into the agent's memory processing pipeline, making it consistently personality-dependent. PersMem processes memory using four steps, where the personality is mapped to operation-specific parameters controlling: (i) affective appraisal annotating emotion states of the user input; (ii) retention of previously stored memories along with the current input; (iii) passive affect-driven memory retrieval exploring memories similar to user input in semantics and personality-guided emotions; and (iv) active goal-driven memory retrieval that refines and selects passively retrieved memories for the reply. Consequently, consistency with the pre-defined personality can be examined by inspecting memory-processing traces during human-agent interactions. We evaluate these personality-dependent differences in attachment and Big Five settings. PersMem exceeds the chance baseline for four-way attachment classification by 23.1 percentage points. In Big Five dialogue comparisons, PersMem achieves 67.5% accuracy, 6.7 percentage points above a baseline using uniformly sampled memories. On CoSER, PersMem achieves an average score of 66.13, with scores of 69.33 for Character Fidelity and 84.33 for Storyline Quality. Together, these results show that PersMem produces distinguishable personality-related memory-processing patterns. --- ### From Static to Dynamic: On-Policy Distillation from Image to Video Diffusion Models Authors: Bingqing Jiang, Li Luo, Zichao Yu, Yujin Han, Zhaolong Su, Difan Zou Published: 2026-09-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.34371v1 On-policy distillation (OPD) specializes pretrained video diffusion models through teacher supervision along the student's own generation trajectory. Although large video models are natural teachers, developing specialized video experts can require costly video data and training, while querying them incurs substantially higher latency than querying image experts. More readily available and cheaper to query, image experts offer a cost-effective alternative, particularly for largely temporal-agnostic capabilities such as aesthetics and OCR that admit frame-level supervision. However, heterogeneous image and video latent spaces prevent direct supervision of intermediate student states, while image experts lack cross-frame motion supervision, making temporal consistency vulnerable to frame-level improvements. In this paper, we propose MILD, a Motion-Preserving Image-to-Video Latent Distillation framework that transfers specialized image expertise while preserving pretrained video dynamics. MILD uses a learnable linear connector that aligns student latent states and predicted updates with those of image experts, enabling supervision transfer across heterogeneous latent spaces. We further constrain image-guided corrections around the pretrained student's predictions to preserve video dynamics and incorporate an optical-flow-based motion reward to improve motion quality and temporal consistency. Across specialized image experts and multiple video-student backbones, our method consistently outperforms video-teacher OPD baselines, with further studies demonstrating effective transfer across connector designs and heterogeneous architectures. These results establish image-to-video distillation as an effective route to improving video generation by drawing on the diverse and evolving capabilities of the image-generation ecosystem. --- ### Spexis: Speculative Lookahead Scheduling for LLM Inference Authors: Hyungyu Jung, Jaehyeok Yu, Hoonseo Choi, Sungkyun Kim, Jinho Lee, Jiwon Seo Published: 2026-09-28 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2609.34370v1 Spexis is a multi-GPU LLM inference framework that improves the efficiency of pipeline and tensor parallelism through speculative parallelism. Rather than using speculative decoding only to accelerate token generation, Spexis runs speculation in parallel with normal execution, introducing a new parallelism axis without increasing KV-cache memory usage. This improves memory efficiency and helps mitigate the bottlenecks of multi-GPU inference. Spexis further uses lookahead scheduling to predict speculation quality and future memory pressure, allowing it to reduce wasted speculation, KV-cache eviction, and recomputation. Built on top of vLLM, Spexis largely improves serving performance across a range of GPU configurations, achieving speedups of up to 34% over a baseline that uses the optimal combination of pipeline and tensor parallelism. Spexis's source code is publicly available at https://github.com/mlsys-seo/spexis. --- ### Rate-Distortion Adaptive Primitive Selection for Omnidirectional Gaussian Splatting Authors: Yulong Cheng, Youneng Bao, Junfeng Zhou, Mu Li, Jie Wen Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34367v1 Learned image codecs (LICs) achieve high reconstruction quality, but their decoding speed is often insufficient for immersive virtual reality (VR). Gaussian splatting (GS) codecs render much faster, yet still lag in reconstruction quality and typically decide primitive allocation without considering the coding cost of each primitive. We introduce OIC-GS, an omnidirectional GS codec with a new hierarchical HEALPix primitive grid representation. Gaussian primitives are anchored at predefined spherical locations, eliminating explicit coordinate coding. Finer levels refine their coarser ancestors, naturally supporting coarse-to-fine reconstruction and layered transmission. The predefined grid also enables efficient viewport decoding by selecting only view-relevant primitives. We further introduce a lightweight entropy model for quantized primitives and optimize the codec under a spherical rate-distortion objective. Primitives with insufficient rate-distortion benefit are automatically removed when their quantized opacity becomes zero, allowing OIC-GS to adapt both primitive density and level of detail without a fixed primitive budget. A single bitstream supports full-sphere, viewport-dependent, and progressive decoding. The first viewport reaches final quality after decoding only 52% of the bitstream, and is then rendered at 1,270 FPS. On a 100-image omnidirectional benchmark, OIC-GS outperforms all evaluated GS codecs, reducing WS-PSNR BD-rate by 49.6% over GaussianImage++ and 68.6% over SGI, which uses a learned entropy model. --- ### When Harness Beats Scale, and When Reading Beats Both Authors: Ivan Bondarenko, Nikolay O. Nikitin Published: 2026-09-28 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2609.34366v1 We describe our system for DocSem, the document-grounded quantitative reasoning shared task at DocInsights 2026, and analyze why it succeeded on labeled data and failed on the test set. The pipeline pairs hybrid block retrieval with Program-of-Thoughts (PoT) generation executed in a sandboxed interpreter, self-consistency sampling, and entity enrichment from chunk-level knowledge graphs. On our held-out split, application architecture moved the metrics far more than model scale did: PoT added 0.282 joint accuracy to a compact 7B model but at most 0.005 to a 72B model, and a 27B model with the full harness matched the 72B (0.884 vs.\ 0.873) at roughly 2.7$\times$ fewer parameters and a quarter of the CO$_2$. We read this through a distinction between world knowledge, which scales steeply with parameters, and language knowledge, which scales gently, and show that structured-output training makes a compact model harness-ready rather than merely small. On the raster, watermarked test PDFs the same system collapsed to 13.58\% joint (rank 149 of 163); a controlled re-rendering of the validation set reproduces the OCR half of the collapse while bounding what the simulation misses. Auditing the physical nature of evaluation inputs precedes architecture, and the leaderboard's bimodality is consistent with reading quality, not reasoning, having separated the field. --- ### SyncRA: Learning Temporal Correspondence in Omni-Modal Models Authors: Zelong Xu, Yan Li, Wenhe Hu, Xiyang Hu Published: 2026-09-28 Categories: cs.CV, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2609.34363v1 Recent omni-modal models demonstrate strong perception of audio and visual inputs, yet often struggle to connect what they hear with what they see at the same moment. This weakness in temporal correspondence can cause models to associate spoken cues with the wrong visual scenes, producing plausible answers grounded in incorrect audio-visual pairings. We diagnose this problem through controlled temporal swaps, revealing that model answers do not reliably follow changes in these pairings. To address it, we propose Synchrony-Guided Representation Alignment (SyncRA), a lightweight method for strengthening temporal correspondence between audio and vision. Specifically, SyncRA contrasts intermediate audio-visual representations within each video, aligning matching moments while separating mismatched ones to capture local temporal correspondence within a shared global context. The objective derives supervision directly from existing input timing, requiring no additional annotations and leaving inference unchanged. We evaluate SyncRA across four open omni-modal models spanning different sizes and architectures on five public video benchmarks. SyncRA consistently outperforms answer-only fine-tuning across all model-benchmark combinations, while substantially improving the ability to track changing audio-visual pairings in controlled evaluations. These results demonstrate that lightweight, targeted supervision can effectively strengthen temporal correspondence and translate into broad improvements in audio-visual question answering. --- ### CoeF-SFL: Preserving Collaborative Server-Client Learning with Enhanced Communication Efficiency Authors: Junwoo Bae, Jin-Hyun Ahn Published: 2026-09-28 Categories: cs.AI, stat.ML arXiv: https://arxiv.org/abs/2609.34360v1 Split Federated Learning (SFL) enables resource-constrained clients to participate in collaborative training, but vanilla SFL exchanges smashed data and gradients at every batch, which incurs significant communication overhead. Recent methods reduce this overhead with an auxiliary network at the client-side cut layer. However, we identify that this approach makes the client optimize a local objective that differs from the end-to-end objective, which fundamentally limits the collaborative training between the client and the server. We propose Compensated Feedback based SFL (CoeF-SFL), a communication-efficient framework that retains the end-to-end objective without any auxiliary network. In CoeF-SFL, the client and the server exchange the smashed data and the gradients once per round and reuse them during local training. Since this reuse makes the gradients stale on the client side, we compensate them with a curvature-based correction in the activation space and develop two variants. CoeF-D approximates the Hessian with a diagonal gradient outer product, while CoeF-J exploits the tractable Jacobian-based Hessian of a surrogate loss that upper-bounds the true loss. We provide the theoretical background of each method, characterizing its compensation. Across vision and language tasks, model capacities, cut layers, and data distributions, CoeF-SFL significantly outperforms auxiliary-network-based methods under the same communication frequency, and the improvement is most substantial on vision tasks. Code is available at https://anonymous.4open.science/r/CoeF-SFL-2686/README.md --- ### Improving Large Language Models for Code through Runtime Program-State Reasoning Authors: Hongwei Li, Spandan Garg, Yufan Huang Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34359v1 Large language models receive limited explicit training in reasoning about runtime program states. We study whether training models to reason about runtime program states improves downstream software-engineering capabilities. We introduce two complementary program-state reasoning tasks. Buggy input-output reasoning requires a model to generate a concrete input that exposes a behavioral difference between a buggy program and a hidden correct implementation and to predict the resulting execution behavior. Precondition-postcondition reasoning requires an agent to symbolically characterize a bug-triggering precondition, predict the expected postcondition, explain their causal connection, and instantiate this reasoning as an executable regression test. By incorporating these two tasks into a staged post-training pipeline, we develop Comet-9B, a 9B language model based on Qwen3.5-9B Base. We evaluate the resulting checkpoints on repository-level patch generation, regression-test generation, and security PoC generation. Adding both program-state reasoning tasks to supervised fine-tuning (SFT) on issue resolution improves success rates by 7.25 percentage points on SWE-bench Pro and 9.70 points on SWT-Bench Verified. Sequential reinforcement learning on the two tasks yields further gains of 7.25, 26.79, and 4.67 percentage points on SWE-bench Pro, SWT-Bench Verified, and CyberGym, respectively. Despite having only 9B parameters, Comet-9B achieves a score comparable to the reported GPT-5.2 result on SWE-bench Pro and matches the reported success rate of a GPT-4o-based agent on SWT-Bench Verified. --- ### FORGE: Form-Optimal Routing of Grounded Evidence for Frozen LLM Agents Authors: Xi Xiao, Yunbei Zhang, Chen Liu, Lin Zhao, Jialin Chen, Tianchen Zhao, Xiang Xu, Youngeun Kim, Tianyang Wang, Min Xu Published: 2026-09-28 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.34358v1 In agentic AI systems, frozen foundation models are increasingly deployed as closed-weight API endpoints, making downstream adaptation possible only through the inputs and inference procedures surrounding the model. As a result, for each input query, two coupled decisions largely determine both answer quality and token cost: what evidence to provide and how much reasoning budget to allocate. Fixed defaults along these axes are often suboptimal, misallocating support form or reasoning depth on roughly 80% of queries in our analysis. To address this challenge, we propose FORGE, a unified framework for adapting frozen models through per-query routing over a joint action space that spans both support form and thinking depth. Under an entropy-regularized, cost-aware utility objective, we derive a closed-form Boltzmann routing target and instantiate the policy as a lightweight 269K-parameter factorized router. The routing policy is trained around the frozen host, without any weight access, through a three-stage pipeline: offline arm enumeration, supervised Kullback-Leibler (KL) distillation from the Boltzmann target, and Group Relative Policy Optimization (GRPO) refinement with host feedback. Across 5 knowledge-intensive benchmarks and 8 frozen backbones ranging from 7B to 671B parameters, FORGE improves accuracy at 42-45% lower token cost on both main hosts, transfers zero-shot across hosts at lower token cost, and composes with intrinsic thinking budgets where available. --- ### Predictive Semantic Safety: From Visual Physical Reasoning to Safety-Critical Control Authors: Taekyung Kim, Salem Fradi, Yanning Dai, Mateusz Ostaszewski, Jürgen Schmidhuber Published: 2026-09-28 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.34356v1 Physical interactions can create future hazards that are not apparent from the robot's current geometric surroundings. We present a framework termed Predictive Semantic Safety (PSS), which connects visual physical reasoning to backup-based safety filtering. A vision-language model (VLM) predicts physical events and their timing or directly predicts object displacements. An explicit motion model converts event hypotheses into object trajectories. Split conformal prediction calibrates position errors jointly across specified objects, observation times, and future times; geometric shape bounds convert the resulting position regions into predicted object occupancy. PSS evaluates a prescribed backup maneuver against this occupancy and derives input-affine constraints for minimally modifying the nominal input while preserving backup feasibility under the robot dynamics and input limits. MuJoCo experiments with a Unitree Go1 consider falling fixtures, impact-driven support loss, and contact propagation. PSS achieves a safe episode rate of 99.3%, compared with 43.3% for a Backup Control Barrier Function baseline that only uses current obstacle geometry. --- ### FAST-Brain: A Flow-Aligned Spatio-Temporal Surrogate Brain Model Authors: Shucheng Liu, Changchun Shi, Kai Zhang, Hongtu Zhu Published: 2026-09-28 Categories: cs.LG, q-bio.NC arXiv: https://arxiv.org/abs/2609.34354v1 Modeling resting-state functional magnetic resonance imaging (rs-fMRI) data is crucial for understanding brain-wide neural activity. However, traditional methods struggle to capture complex temporal dynamics over long horizons, to account for the brain's anatomical spatial structure, and to model high-dimensional ambient signals that lie on a low-dimensional intrinsic subspace. We propose FAST-Brain, a unified flow-aligned spatio-temporal surrogate brain model that addresses all three challenges. At its core is a flow-aligned generative framework that directly predicts the clean blood-oxygen-level-dependent (BOLD) signal, paired with a graph convolutional network that captures spatial structural constraints and a Transformer that models long-range temporal dependencies. Theoretically, we show that under a low-dimensional subspace assumption, the approximation error of our model scales with the intrinsic dimension rather than the ambient dimension, which justifies our direct modeling of the BOLD signal. Extensive experiments on synthetic and Human Connectome Project datasets demonstrate that FAST-Brain achieves state-of-the-art performance in recovering functional connectivity, effective connectivity, and the implicit low-dimensional signal subspace. --- ### SemRD-V2X: Closure-Guided Communication with Bounded Inference for Cooperative Perception Authors: Hu Xu, Chun Li, Siyuan Qiu, Zeyan Li, Jianfeng Xu Published: 2026-09-28 Categories: cs.AI, cs.IT arXiv: https://arxiv.org/abs/2609.34353v1 Vehicle-to-Everything (V2X) cooperative perception improves 3-D detection by sharing intermediate features, but dense remote features may repeat context that the ego agent can infer locally. Most communication-efficient designs optimize masks or codes empirically, leaving a more basic question open: which remote evidence is indispensable given the receiver's own observation? We introduce a closure-fidelity perspective on ego conditioned remote perception. Under a finite deductive abstraction and explicit conditions, its rate--distortion function decomposes over an irredundant core, and the exact zero-distortion rate becomes $P_A H(π_A)$. This analysis suggests a concrete design principle: transmit compact evidence and recover derivable context with bounded receiver-side inference. Guided by this principle, SemRD-V2X is an operational neural proxy that combines exact-budget BEV support selection, pointwise channel compression, and masked shared-weight reconstruction before standard fusion. Experiments on simulated V2XSet and real-world DAIR-V2X validate the resulting design. In a controlled five-run V2XSet comparison against a locally reproduced V2X-ViT-v1 baseline on one Tesla V100, SemRD-V2X reduces the analytical feature payload by $26.6\times$ while improving AP@0.5/AP@0.7 by 4.13/8.57 points, with 3.81\% additional mean compute latency. These results position closure fidelity as both an analytical lens and an actionable design principle for communication-efficient cooperative perception. --- ### Fuzzy Distribution Modeling for Synthetic Tabular Data Generation with Causality Preservation Authors: Michael Vasilakakis, Dimitris K. Iakovidis Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34349v1 Synthetic tabular data generation provides an effective alternative for the training of machine learning models when real-world data is limited or inaccessible. However, the heterogeneous, non-smooth, and incomplete nature of tabular data poses fundamental challenges to conventional probabilistic and deep generative models, where their interpretability remains limited. This paper proposes a novel fuzzy distribution modeling methodology for synthetic tabular data generation based on fuzzy sets theory. Feature distributions are represented using fuzzy sets and feature dependencies are modeled through Fuzzy Cognitive Maps, resulting in a low-parameter, and an interpretable data representation. Synthetic samples are generated by sampling fuzzy concepts rather than raw values, enabling native support for mixed data types, missing values, and domain constraints. The methodology further supports linguistic queries and IF-THEN reasoning, facilitating transparent simulation of decision-making processes. Experimental results on benchmark datasets demonstrate competitive performance with respect to utility, fidelity and privacy compared to state-of-the-art methods, while offering substantially improved interpretability. These results establish fuzzy distribution modeling as a principled and effective approach for synthetic tabular data generation in fuzzy systems and decision support applications. --- ### Commutator Memory: Sparse, Path-Local Reading and Steering in Language Models Authors: John Sweeney Published: 2026-09-28 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.34348v1 Gradient updates on different data generally do not commute: training a language model on two data sources in opposite orders gives different weights, even with the same data and total exposure. Loss or benchmark deltas show that the models differ, not where. We ask whether this path dependence leaves a parametric training-history memory: a weight component that flips sign when the two sources are swapped, is localized in output space, changes the held-out loss gap between the two orders under targeted interventions, and reveals which trained model came from which order. For one small SGD step of size $η$ on each of sources $A$ and $B$, the weight difference $θ_{AB}-θ_{BA}$ is, to leading order, $η^2 b_{AB}$, where $b_{AB}=H_Bg_A-H_Ag_B$ is the Lie bracket of the two gradient fields at the base model. We define commutator memory by projecting the bracket through the logits into one score per vocabulary token; the scores sum to the bracket's prediction of the gap. The scores are localized: on three models, the same readout of the measured $θ_{AB}-θ_{BA}$, or of a bracket from disjoint batches, shares 82-99% of the original top-20 tokens, versus 35-49% for norm-matched random directions. They are causally actionable: in Qwen-3-4B SFT, downweighting the ten tokens with the largest predicted share of the gap closes a median 32% of the measured gap, while frequency-matched tokens with near-zero scores have almost no effect. The weights themselves carry the component: projecting the difference between the two trained models onto $b_{AB}$ identifies which came from which order in 92% of cases across four LLMs (chance 50%). Controlled tests also cover matched-batch DPO, a frozen-rollout GRPO-style objective, and an AdamW endpoint check. The memory is defined per source pair, not per example, and its projection on $b_{AB}$ decays with further training. --- ### SAIL: Spatial Audio Intelligence with Large Language Models via Disentangled Acoustic-Spatial Encoding and Dual-Stream Q-Former Authors: Zhengding Luo, Jinyang Wu, Haozhe Ma, Yanghao Zhou, Woon-Seng Gan, Wenwu Wang Published: 2026-09-28 Categories: cs.SD, cs.AI, eess.AS arXiv: https://arxiv.org/abs/2609.34347v1 Spatial audio large language models (LLMs) enable embodied agents, wearable assistants, and immersive systems to recognize sound events, localize sources, and reason about their spatial relationships. However, existing spatial audio LLMs often rely on early fusion of acoustic and spatial features and source-agnostic token representations. These designs make it difficult to preserve the correspondence between individual sound events and their spatial attributes, particularly in multi-source scenes. To address this limitation, we propose SAIL, a Spatial Audio Intelligence framework with LLMs that preserves acoustic-spatial structure and source-level correspondence from audio encoding to LLM alignment. SAIL introduces a Disentangled Spatial Audio Transformer that represents Mel-spectrogram and interaural phase difference features as separate acoustic and spatial streams. Source-discriminative task queries further learn event, direction, and distance information for each source. A Dual-Stream Q-Former then aligns the two streams with the LLM using acoustic and spatial queries organized by source slots. Compared with the early-fusion baseline, SAIL achieves consistent improvements in dual-source sound event detection, direction and distance estimation, and spatial reasoning. These results demonstrate the importance of structured, source-discriminative audio representations for multi-source spatial understanding and reasoning. --- ### E-WAVE: Event-based Continuous Optical Flow via Warping-Aligned Visual Encoding Authors: Jiale Wu, Xiaoyang Bai, Haoming Yu, Yiwei Chen, Yifan Peng, Weiwei Xu Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34346v1 Temporally dense optical flow is essential for dynamic perception in immersive VR/AR systems, where rapid head, hand, and object motion must be continuously captured and tracked. Existing frame-based optical flow estimation methods are constrained by the tradeoff between temporal resolution and computational cost; while event cameras, with their high temporal resolution and energy efficiency, serve as a natural solution to the dilemma. However, event-based approaches commonly rely on correlation volumes to capture pairwise voxel correspondences, which incur substantial memory and computation overhead. We present E-WAVE, a correlation-free framework for high-temporal-resolution (HTR) optical flow estimation from event streams. Instead of constructing all-pairs correlation volumes, E-WAVE employs global attention mechanism to model long-range feature dependencies and performs trajectory guided feature warping using Bézier curve. Through iterative updates, it predicts trajectories that allow for querying at arbitrary timestamps without repeated inference. Experiments on MultiFlow and DSEC-Flow demonstrate a 25% lower trajectory error and comparable endpoint flow estimation accuracy relative to state-of-the art baselines. Additional evaluations on self-captured data using a head-mounted prototype validate that E-WAVE remains robust under challenging real-world conditions. --- ### CRISP: Cultural Reward Modeling for Implicit Situated Propriety Authors: Zekun Yuan, Yangfan Ye, Baohang Li, Shuaibo Zhao, Zekun Zhou, Ziming Li, Qichen Hong, Kun Chen, Xiaocheng Feng Published: 2026-09-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.34345v1 As large language models (LLMs) are increasingly deployed across countries and regions, the ability to recognize and respond appropriately to diverse cultural contexts becomes increasingly important. However, existing research has largely focused on cultural knowledge or tasks with predefined response spaces, while open-ended culturally situated behavior remains comparatively underexplored. In this work, we introduce CRISP-RM, a culturally situated reward model that assigns rewards according to cultural appropriateness in open-ended social scenarios. During policy optimization, we further introduce Norm Grounding Supervision (NGS), providing guidance that enhances the policy's sensitivity to relevant cultural norms. To construct culturally situated data, we employ a collaborative multi-agent framework that instantiates implicit cultural norms into diverse social scenarios and further curate NormCompass as a dedicated testbed. We conduct comprehensive experiments to evaluate the effectiveness of CRISP-RM in both reward modeling and policy optimization. Best-of-\(N\) experiments show that CRISP-RM consistently outperforms strong general reward models. During GRPO policy optimization, CRISP-RM generally improves culturally situated behavior, while incorporating NGS yields further gains. Further analyses demonstrate the advantages of CRISP-RM in distinguishing culturally appropriate behavior beyond superficial fluency and politeness, while NGS provides complementary gains during policy optimization by improving norm grounding. --- ### Learning to Steer, Steering to See: Unveiling the Geometry of RLVR in Large Language Models via Trainable Vectors Authors: Yuchen Cai, Ding Cao, Qixiang Yin, Xin Xu, Kai Yang, Siye Wu, Pengyuan Wang, Jiaxuan Wang, Weijie Liu, Saiyong Yang, Guangzhong Sun, Guiquan Liu et al. Published: 2026-09-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.34344v1 Reinforcement learning (RL) has become a key paradigm for enhancing the reasoning of large language models, yet the high dimensionality of parameter updates makes its training dynamics hard to analyze. We study reinforcement learning with verifiable rewards (RLVR) and use vector steering to identify a low-dimensional effective manifold in activation space associated with RL-induced gains. We uncover two geometric properties. (1) Effective Manifold Capacity: the capacity needed to reproduce RL gains can be very small but is not infinitely compressible; at extremely low capacity, intervention dimensionality and input-dependent expressiveness become key constraints, and this requirement varies with injection depth. (2) Control Manifold Separation: effective control directions lie mainly in the low-variance complement of the activation principal subspace. Within a task and base model, the learned geometry stays largely consistent across training configurations, and across tasks geometric alignment correlates with capability transfer. Experiments on 5 LLMs and 6 verifiable-reward tasks support these findings. We then propose Alpha-Stabler, a plug-and-play framework with a Predictor that monitors principal-subspace intrusion for early collapse warnings, and a Controller that removes the principal-subspace component of activation gradients during backpropagation while preserving the orthogonal complement. Alpha-Stabler stabilizes training for 2,000 steps and consistently improves RL gains, offering practical insights for robust post-training. Code: https://github.com/caiyuchen-ustc/On_Policy_Vector_Training --- ### The Composition Gap in Dataset Distillation Authors: Guang Li, Takahiro Ogawa, Miki Haseyama Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34343v1 Dataset distillation compresses a training set into a small synthetic set, usually evaluated one at a time. In federated and data-governance settings, several parties distill their own data and a user trains on their union. We ask whether the union of separately distilled sets reproduces training on the union of the real data composability and show that it can fail even when every source is distilled exactly and the total budget admits an exact joint distillate. Compressing a training trajectory into fewer steps transforms the source statistics nonlinearly, so averaging compressed sources differs from compressing their average. For quadratic objectives we derive the exact composition error for two-to-one step compression in terms of the source-Hessian variance and the linear terms of the losses, and on a smooth network at small step sizes this prediction captures the local endpoint discrepancy in magnitude and direction. For learned synthetic sets, however, the composed error decomposes exactly into this local discrepancy and an aggregate source residual. Under endpoint matching the residual exceeds the structural term by more than an order of magnitude, and under distribution matching the two terms partly cancel. Joint distillation also retains an accuracy advantage when both sets are distilled from the same dataset, where the local discrepancy is exactly zero. Training fidelity and downstream accuracy are therefore distinct requirements, neither established by evaluating each set on its own. --- ### SAGE: Structured Strategic Reasoning for Efficient LLM Game Playing Authors: Zhiwei Chen, Tianchun Wang, Zhongtao Rao, Haiming Zhu, Ding Cao, Tianxiang Zhao Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34342v1 A strong LLM strategic agent should reason prospectively over uncertain futures, adapt its strategy to opponents' behavioral tendencies, and continuously recalibrate its decision process from interaction experience. However, incorporating these sources in free-form reasoning could lead to unsupported strategic assumptions, inconsistent opponent estimates, and harmful interference from irrelevant historical interactions. To address these issues, we propose SAGE, a training-free inference-time framework that structures LLM strategic reasoning around three coordinated operations: anchor, adapt, and recalibrate. SAGE first anchors reasoning to an equilibrium policy that provides a strategically valid prior. It then conditions deviations from this anchor on a soft belief over opponent behavioral tendencies, enabling opponent-specific exploitation. Finally, SAGE distills strategically related interactions into counterfactual hypotheses about previously missing considerations, allowing past experience to recalibrate the model's reasoning. We evaluate SAGE on three repeated imperfect-information games: Leduc Hold'em, Liar's Dice, and Goofspiel, against various opponent types in each game. Compared with reasoning-intensive LLM agents, including Suspicion-Agent, ReTA, Agent-Pro, EMO, and Hypothetical Minds, SAGE achieves up to a 127.6% payoff improvement in Liar's Dice while reducing input and output token usage by up to 80% and 90%, respectively. In direct match-up play, it attains non-negative mean payoff against 5/10, 8/10, and 8/10 evaluated opponents in Leduc Hold'em, Liar's Dice, and Goofspiel, respectively, while using relatively fewer tokens. Code is available at https://github.com/chenzhwsysu57/SAGE. --- ### SkillPE: Creativity-Oriented Cinematic Skill Evolution for Text-to-Video Prompt Engineering Authors: Yanwei Huang, Mingxuan Zhu, Shujie Li, Shiyuan Liu, Yuanxing Zhang, Arpit Narechania Published: 2026-09-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.34335v1 Achieving high-quality, cinematic results in text-to-video generation remains challenging for non-experts, whose prompts often lack professional narrative and creative design. We propose SkillPE, a prompt engineering (PE) framework that evolves reusable cinematic skills from expert-authored seeds. SkillPE represents shot logic, composition, lighting, sound design, and other filmmaking cues in a fine-grained format, and retrieves movie references categorized as resonators (good matches), dissonants (weak matches), and divergents (creatively useful near-misses). The first two refine when and how a skill should be applied, while divergents inspire alternative cinematic realizations at different degrees of modification while preserving the user intent. Candidate skills are assessed through generated videos along prompt fidelity, cinematic quality, narrative appeal, and creativity to construct the final skill libraries. Experiments on StoryEval and VBench show improvements of up to 1.40 points over the strongest external baseline and 0.51 points over seed skills on 7-point four-dimensional evaluation, while remaining competitive on benchmark-native metrics. Overall, SkillPE offers a practical approach to balancing fidelity and creativity in cinematic text-to-video generation. Code is available at https://github.com/Ais0n/SkillPE . --- ### MiCo: Mutual Information Coverage Optimization through Semantic Erasure Modeling for Efficient MLLM Inference Authors: Tinghao Wang, Yichen Guo, Qizhe Zhang, Yuan Zhang, Weimin Ouyang, Rui Huang, Jiajun Cao, Sixiang Chen, Hao Jiang, Jixian Wu, Zheng Lu, Bofan Zhu et al. Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34330v1 Multimodal large language models (MLLMs) have demonstrated impressive performance in multimodal understanding, but processing large numbers of visual tokens results in high computational costs. While many methods have been proposed to reduce the number of visual tokens, most of them rely on heuristics and are prone to discarding substantial visual information during pruning, leading to degradation in model performance. In this work, by using a semantic erasure model, we derive a general mutual information coverage objective from task log-loss and propose MiCo, a training-free two-stage pruning method. MiCo first uses visual signals to select a representative candidate pool before visual tokens enter the language model, then performs task-aware subset selection within it. At each stage, suitable observable proxies instantiate the derived objective as a monotone submodular coverage function, which MiCo greedily optimizes under the token budget. MiCo is evaluated on diverse MLLMs ranging from 7B to 13B parameters across a broad range of image and video benchmarks spanning general visual reasoning, fine-grained OCR and grounding, hallucination detection, and long-video understanding. MiCo consistently achieves the best performance across nearly all evaluated models under all pruning ratios. On LLaVA-NEXT-13B, MiCo uses only 5.6% visual tokens, retains 97.5% of baseline performance, and achieves a 3.8-fold inference speedup. Our experiments demonstrate the effectiveness of MiCo and our mutual information coverage objective for visual token pruning. --- ### Knowing When Thinking Is Not Enough: Teaching Small Reasoning Models to Reason Beyond Their Parametric Knowledge Authors: Chanuk Lee, Minki Kang, Sangwoo Park, Woongyeong Yeo, Jinheon Baek, Sung Ju Hwang Published: 2026-09-28 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.34327v1 Scaling test-time computation is a powerful way to improve language-model reasoning, and is particularly appealing for small reasoning models (sRMs) that are cheap to serve. However, is additional thinking always the right operation? By intervening at intermediate reasoning states across two model families and multiple scales, we find that self-refinement largely consolidates probability mass onto solutions already reachable from the current state, rather than making new ones reachable. These interventions reveal two failure regimes: execution bottlenecks, where the correct path is reachable and reflection can recover it, and knowledge bottlenecks, where relevant external information makes it reachable. Motivated by this distinction, we introduce FlyBy, a selective querying framework, and train 4B and 8B variants to reason first, diagnose what remains unresolved, and, at a knowledge bottleneck, query stronger models whose parametric knowledge extends beyond its own. Supervised fine-tuning bootstraps a multi-depth query action, and cost-aware reinforcement learning calibrates whether to query, what to ask, and how much to spend. On 1,158 hard problems across six benchmarks, FlyBy-4B achieves 45.96% pass@8, surpassing Qwen3-14B (41.64%) at 2.7 times lower serving cost, while also exceeding Qwen3-8B in pass@1 (16.85% vs. 15.31%). Scaling to FlyBy-8B further improves pass@8 to 51.81%. --- ### Routing Without Embeddings: Fast And Interpretable Routing With Regular Expressions Authors: Yifan Lu, Qiyue Zhang, Haotian Shan, Hanjie Chen, Jiarong Xing Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34326v1 Large Language Model (LLM) routers commonly rely on neural query embeddings, with larger encoders expected to better capture query intent and difficulty. Yet scaling Qwen2.5 encoders from 0.5B to 72B parameters brings little improvement in routing accuracy (Figure 1b), suggesting that small encoders may already capture the query properties needed for routing. We therefore investigate which properties matter and whether they can be extracted directly from text without a neural encoder. We introduce REGEXROUTE, a pipeline that uses sparse autoencoders (SAEs) to discover interpretable regular-expression (regex) features. Using unlabeled text, an LLM turns descriptions of grouped SAE latents into regex extractors and refines them to match latent activation patterns. These extractors supply numerical features to a lightweight routing head, eliminating neural encoding at inference (Figure 1a). Across four benchmarks, one fixed set of 128 features achieves 76.43% average routing accuracy, comparable to 76.41% for the strongest neural text encoder baseline, with much smaller latency and strong robustness. These findings establish explicit, interpretable text features as a practical basis for designing and understanding LLM routers. --- ### DORA: Dynamic Online Reinforcement Agent for Token Pruning in Vision Transformers Authors: Kaixuan He, Song Chen, Yi Kang Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34325v1 Vision Transformers (ViTs) incur quadratic self-attention cost in the number of tokens. Most token-reduction methods adapt token identities within a prescribed layer-wise compression schedule, or search a static mask offline, and thus limit online adaptation of when and how much to prune. We propose DORA (Dynamic Online Reinforcement Agent), which learns an input-adaptive pruning policy itself for frozen ViTs. At each eligible block, a hierarchical actor decides whether to prune, how many tokens to remove, and which tokens to remove from each image's evolving representation. Because early deletions change the states observed by later decisions, DORA formulates pruning as a finite-horizon Markov decision process. Complete-prefix shadow evaluations convert final-prediction fidelity into localized per-step credit, while closed-loop accuracy feedback adjusts the fidelity penalty toward a shared accuracy-drop target. A privileged critic and all shadow computations are training-only. Deployment retains the frozen backbone and a lightweight actor that applies hard deletion and packed variable-length FlashAttention, converting token reduction into measured speedups. On ImageNet-1K with DeiT-Base, DORA reduces FLOPs by 38.4% relative to the uncompressed backbone within one percentage point of accuracy loss. Averaged across four ViT-type backbones at matched accuracy, DORA uses 13.2% fewer FLOPs and achieves 32.4% higher throughput than the corresponding per-backbone baseline means. Under zero-shot transfer to ImageNet-A, these gains widen to 20.3% and 45.6%, respectively. --- ### Test-Time Scaling via Budgeted Multi-Attribute Verification Authors: Bo Xue, Ji Cheng, Shen-Huan Lyu, Yuanyu Wan, Shuang Qiu Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34322v1 Verifying LLM-generated answers under a shared computational budget requires jointly deciding which candidates to inspect and which verification attributes to evaluate. We formulate this problem as multi-attribute good-arm identification under a global budget: each candidate is an arm evaluated along several costly attributes, and the goal is to certify as many candidates as possible whose mean scores exceed the prescribed thresholds on all attributes. We propose \textsc{BMA-GAI}, an algorithm that combines cost-aware arm selection with adaptive sampling of attributes. Every observation serves both to guide adaptive allocation and to support anytime-valid certification, which removes the need for a separate confirmation stage. We establish an asymptotic coverage guarantee for \textsc{BMA-GAI} and derive a matching information-theoretic converse that characterizes the intrinsic complexity of the problem, thereby proving that \textsc{BMA-GAI} is first-order optimal away from critical budget levels. Experiments on synthetic benchmarks and an LLM answer-verification task show that \textsc{BMA-GAI} allocates the verification budget more efficiently and certifies more high-quality candidates than competing methods. --- ### One Rollout Is All You Get: Fully Test-Time Adaptation for GUI Agents Authors: Ziqiang Wang, Li Gu, Zhixiang Chi, Linlian Jiang, Zihuan Jiang, Linqiang Guo, Siobhan Reid, Zhi Liu, Yang Wang Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34321v1 GUI agents are deployed with frozen weights and discard everything they experience on the job. Existing ways to update an agent's weights assume something deployment withholds: ground truth, rollouts beyond the single attempt (retries, samples, practice runs), or a learning phase other than deployment. Because GUI actions can be irreversible, a deployed agent gets one attempt per task occurrence, in arrival order, and every attempt counts. No ground truth is available at any point. We define fully test-time adaptation for GUI agents by these constraints and pair it with a minimal weight-space method, SOLO. Auxiliary models read each episode: a judge selects the episodes it deems successful, and a proposer-verifier pair relabels a failed episode's prefix with the subtask that prefix completed. Admitted episodes enter a short sliding window, and each admission updates a small adapter by top-K self-distillation on the agent's own predictions, provided the window holds a judged success. On recurring task streams built from WebArena, VisualWebArena and MobileWorld, SOLO improves on the frozen agent with both UI-TARS-7B and Qwen3-VL-8B, by three to six points of success rate, and exceeds two in-setting memory methods on the web streams. --- ### Certified Selective Automation of LLM Agent Evaluation Authors: Chengguang Gan, Yunhao Liang, Qinghao Zhang, Shiwen Ni Published: 2026-09-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.34320v1 Evaluating LLM agents still ends with a human reading trajectories, because automatic judges carry no guarantee on how often they are wrong. We ask the operational question: what fraction of agent evaluation can a judge take over, with a certificate that the error rate among auto-decided trajectories stays below a budget alpha? Agent corpora resist the standard answer: many agents attempt the same tasks, so trajectories arrive in correlated clusters, and the i.i.d. certificates of existing selective-judging methods can overstate what is safe: a naive certificate can claim 98% automation while its realized error exceeds the budget in 17.5% of task resamples. We introduce a task-level bootstrap certificate that is valid in every regime we test while matching the naive certificate's coverage; finite-sample cluster-valid alternatives certify nothing at realistic task counts. Under this certificate, a 4B logprob judge trained with SFT and reject-weighted GRPO certifies 0.30-0.59 of evaluation on tool-use and web corpora at alpha=0.1, the only judge, among strongly elicited frontier models, certifying on both headline corpora. Certified coverage is predictable before training from base rate and discrimination alone (leave-one-corpus-out R^2=0.96). Finally, the certificate doubles as a self-training filter: pseudo-labels harvested inside certified regions have contamination bounded by alpha by construction (realized 0.000-0.041 across six harvests), letting a judge enter an unseen domain at in-domain strength with zero target training labels. ---