# Research Papers — 2026-09 · bundle page 134 of 266 > Full-text markdown of papers #6651–6700 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/134 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/133.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/135.md) ### OpenSAL360: Open-Source Crowdsourcing Platform for Omnidirectional Video Saliency Collection Authors: Alexey Bryncev, Andrey Moskalenko, Kira Shilovskaya, Ivan Kosmynin, Dmitriy Vatolin Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21480v1 Omnidirectional video saliency prediction plays an important role in many immersive multimedia applications, including viewport-adaptive streaming and compression, foveated rendering, mesh simplification, perceptual quality assessment. Yet progress in this area remains constrained by the cost and complexity of collecting eye-tracking data with VR headsets, which makes large-scale dataset creation difficult to extend. We present OpenSAL360, the first open-source platform for scalable, low-cost 360° video saliency collection. Unlike conventional VR-based protocols, it requires only a standard screen, mouse, and internet connection, enabling parallel saliency data collection from common crowdsourcing assessors without specialized hardware. We validate our collection protocol against seven well-established VR eye-tracking datasets and conduct ablation studies on key interface, pre-, and post-processing parameters. To demonstrate the effectiveness and scalability of the proposed methodology, we collect and publicly release a saliency dataset covering 500 omnidirectional videos annotated by 2,000+ crowdsourcing assessors, making it, to the best of our knowledge, the largest dataset in this field. We make OpenSAL360 publicly available at https://github.com/msu-video-group/OpenSAL360. --- ### MT-WAM: Reorienting the One-Pass Predictive Representation Toward Action Generation Authors: Yiguang Yang, Jiankun Peng, Xiaoming Wang, Yiran Zhang, Zhibo Fang Published: 2026-09-18 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.21474v1 Fast-WAM shows that video-action co-training improves control without generating future video at inference, making the representation from a single video diffusion Transformer forward central to action generation. However, future-observation prediction does not explicitly prioritize the future dynamics and visual structure needed for control. We present MT-WAM, which retains the original training objectives and adds complementary supervision for future two-dimensional point trajectories and visual features. A lightweight dual-stream branch copied from the video backbone's final blocks provides target-specific processing, while a structured attention mask prevents cross-stream attention. Motion-stream tokens supply additional dynamics conditions to the action expert. Future visual-feature prediction provides supervision in a feature space that captures object and spatial structure. This supervision trains the video backbone to provide more informative visual context for action generation under changing visual conditions, without adding visual-feature-stream tokens to action conditioning. At inference, MT-WAM uses video and motion caches computed once per replan and skips future-video prediction. Without additional embodied policy pretraining, MT-WAM achieves 98.2% success on LIBERO and 73.7% on LIBERO-Plus, exceeding Fast-WAM by 23.8 percentage points on the latter. On RoboTwin 2.0 Clean2Rand, Random success increases from 6.30% to 19.40%; across four real-world tasks, average success increases from 67.0% to 77.8%. --- ### Risk-Aware Occupancy for Safety-Oriented End-to-End Autonomous Driving Authors: Jiaxing Chen, Hengduo Zou, Yiren Zhao, Bolin Gao Published: 2026-09-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.21470v2 Conventional end-to-end driving systems model the environment with sparse objects and lane elements. While efficient, this paradigm discards planning-critical information in crowded and occluded scenarios, particularly for unstructured obstacles, ambiguous free space, and complex interactions. We propose risk-aware occupancy, a dense BEV representation that explicitly fuses geometric occupancy, map-derived traffic constraints, and future dynamic-agent occupancy as complementary risk signals. Built upon this representation, we develop ROIDrive, an instance-centric end-to-end framework with a dedicated risk-aware occupancy branch. The predicted occupancy is tokenized via sliding-window sampling and injected into planning queries via cross-attention, while temporal query consistency mitigates unreliable flickering queries. We also contribute RiskOcc4D-nuScenes, a benchmark derived from nuScenes and Occ3D-nuScenes with four automated annotation pipelines for multi-dimensional risk supervision. Experiments on representative occupancy architectures verify the learnability and transferability of our representation. Integrated with GenAD, it reduces collision rates by 35.0% (UniAD metric) and 52.9% (ST-P3 metric), confirming the efficacy of the proposed representation modality. --- ### SkillIR: Evolving Scene-Aware Skills for Agentic Image Restoration Authors: Jie Shao, Shengkai Hu, Xu Zhang, Beihang Song, Yongcheng Jing, Xu Wu, Jun Wan Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21468v1 This paper studies agentic image restoration, in which multimodal agents coordinate specialized restoration tools to recover images affected by complex degradations. Existing restoration agents often derive complete tool-use plans from the original degraded image or retrieve previously successful trajectories, providing limited support for adapting individual actions to evolving intermediate restoration states. We find that accepted tool executions can change the residual degradation state and, consequently, the applicability of subsequent tools. To address this issue, we propose SkillIR, a skill-guided framework that represents restoration experience as degradation-centered action evidence rather than complete tool-use trajectories. SkillIR consolidates context-dependent action outcomes into scene-aware restoration skills that characterize applicable conditions, expected effects, and attributable failure cases. Instead of prescribing a complete restoration plan, the retrieved skills guide one bounded action at a time within a verified residual-state loop: each tool output is treated as a candidate, committed only after transition verification, and followed by reassessment of the active residual degradations. After each rollout, the resulting evidence is used to create, refine, or patch dynamic skills, enabling accumulated restoration experience to improve decision-making for subsequent inputs. Experiments on synthetic and real-world multi-degradation datasets demonstrate that SkillIR improves restoration quality and enables more reliable and effective tool use. --- ### OmniVChat: Synthesizing, Benchmarking, and Training for Native Audio-Visual Dialogue Authors: Haolin He, Yunfei Chu, Qi Chen, Wen Huang, Yuan Feng, Muzhi Zhu, Zheqi Dai, Haoning Xu, Dongchao Yang, Chunyat Wu, Zining Liang, Zhengxi Liu et al. Published: 2026-09-18 Categories: eess.AS, cs.AI, eess.IV arXiv: https://arxiv.org/abs/2609.21465v2 We define OmniVChat (Omni Video Chat) as the task of native audio-visual dialogue between a user and an omni model. In OmniVChat, omni models directly and simultaneously receive audio and video from a user and return text. The user's query is embedded in the audio and video, without a separate text question, external captioning, or speech recognition. Direct audio-visual input reduces external latency and computation while preserving perceptual cues. However, research on OmniVChat faces two constraints: data availability and evaluation. Recordings of people using their own devices are scarce. Furthermore, good replies often depend on multimodal context and can be phrased in many ways, making keyword matching unreliable for evaluation. Recent progress in agent systems and video generation makes generation for comprehension viable, which means using synthesized dialogues for training and evaluation. Therefore, we present OmniVChat-Studio, a multi-agent data engine for synthesizing single- and multi-turn audio-visual dialogues. We use synthesized dialogues to build OmniVChat-Bench, an evaluation benchmark that evaluates omni models' basic dialogue abilities across five ability categories. Replies are judged by a large language model based on explicit scoring criteria. We also present OmniVChat-RL, a reinforcement learning reward design that jointly targets reply correctness, efficiency, and style in OmniVChat. Training Qwen3-Omni-Instruct with OmniVChat-RL on synthesized dialogues improves its performance on both OmniVChat-Bench and the human-recorded OmniVChat-Bench-Human. These gains validate the reward design and show transfer to real-world dialogues in training and evaluation. --- ### PSEE: Progressive Sensor Event Expansion for Point-Supervised Temporal Action Localization Authors: Jiaxi Yin, Ge Wang, Han Ding, Fei Wang Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21462v1 Temporal action localization (TAL) in wearable sensor streams identifies action classes and temporal boundaries, enabling finer-grained activity understanding than conventional action recognition. However, training typically requires costly start--end annotations for every action instance. To reduce this burden, we study point-supervised TAL, where each instance is labeled with only one timestamp and its class. We propose Progressive Sensor Event Expansion (PSEE), which combines semantic activations, sensor-specific transition evidence, and adaptive temporal ownership to recover point-supervised pseudo segments. These segments supervise standard TAL detectors without modifying their inference procedures. Cross-subject experiments on four inertial-sensing benchmarks demonstrate improved pseudo-boundary quality over adapted point-supervised baselines, compatibility with different TAL detectors, and robustness to point sampling. Code is available at https://github.com/joeeeeyin/PSEE. --- ### AtomEgo: Exploring Ego-Robot Integration for Embodied Foundation Model Pretraining Authors: Di Wu, Dongchen Zheng, Junhe Sheng, Zhongxing Wei, Songxin Zhang, Zejian Xie, Xiaoquan Sun, Junyang Zheng, Zhuoyang Song, Jiaxing Zhang, Jiayu Chen Published: 2026-09-18 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.21461v1 Embodied foundation models are constrained by the limited scale and diversity of robot demonstrations, motivating the use of large-scale egocentric human interaction data. However, how to effectively incorporate such data into embodied-model pre-training remains unclear because of substantial embodiment and action-space gaps between humans and robots. We present AtomEgo, a systematic study of ego--robot co-training supported by a curated corpus of approximately 2,659 hours and a scalable data processing pipeline. Across vision--language--action and world--action model architectures, we investigate three representative paradigms: joint co-training with domain-specific action heads, progressive ego-to-robot transfer through embodiment alignment, and joint video--action modeling. We evaluate these paradigms through multi-task real-robot experiments and language-conditioned cross-embodiment representation analysis. Our results reveal a simple principle: Data Scale * Alignment Quality --> Capability Gain; egocentric data can improve generalization, but their value depends on how effectively they are aligned and utilized. This principle can provide practical guidance for scalable ego--robot pre-training. --- ### Efficient Architecture Search under Leave-One-Subject-Out Evaluation Authors: Heinke Hihn, Friedhelm Schwenker Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21457v2 Deep neural architectures are widely used for signal processing in automated pain assessment systems. However, architecture design has remained largely a manual task despite the potential efficiency benefits of Neural Architecture Search (NAS). Embedding NAS in a Leave-One-Subject-Out (LOSO) evaluation is computationally demanding because a fully nested implementation requires $N$ independent architecture searches and, assuming approximately linear training cost, scales as $\mathcal{O}(N^2)$. We propose a block-based, leakage-controlled approach that shares NAS runs between subjects, reducing the number of searches from $N$ to $B$, where $B \ll N$, dubbed PainNAS. On the BioVid Heat Pain dataset, PainNAS yields comparable subject-level accuracy with substantially fewer parameters and FLOPs. --- ### CompAdapt: Adaptable Composite Motion Modeling for Physics-Consistent Text-to-Video Generation Authors: Haoran Qin, Renlong Wu, Tianyu Huang, Yukang Ding, Hui Li, Wangmeng Zuo Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21455v3 While diffusion-based text-to-video (T2V) models have demonstrated impressive capability in generating realistic and temporally coherent videos, they often fail to respect fundamental physical dynamics. Although recent physics-constrained methods incorporate explicit dynamics priors to improve physical plausibility, they remain limited to simple single-type motions, depend on manually specified parameters, and struggle to generalize to unseen physical laws. In this work, we propose CompAdapt, a physics-consistent T2V framework for adaptable generation across complex real-world scenarios. It extends neural dynamics modeling beyond single-type motions to encompass composite physical behaviors, including coupled motions, multi-stage transitions, and multi-object collisions. Furthermore, CompAdapt translates natural language prompts into structured physical semantics, enabling end-to-end specification of motion types, temporal relations, and initial physical parameters. To generalize to novel physical environments, CompAdapt introduces dynamics-aware prior matching, achieving one-shot adaptation without retraining the core dynamics module. In addition, a physics-aware latent feature fusion module improves visual fidelity under fast and complex motion. Experiments on physics-focused T2V benchmarks demonstrate that CompAdapt improves physical consistency over both general T2V models and physics-constrained baselines, while preserving high visual quality and adaptability to unseen dynamics. The project page is available at https://makapic.github.io/CompAdapt/ . --- ### Improving the Predictive Performance of Bootstrap Aggregating by Dirichlet Resampling Authors: Quoc Viet Le, Joonha Park Published: 2026-09-18 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2609.21454v1 We revisit Breiman's observation that reducing inter-tree correlation without weakening individual trees can improve random forests. Building on this principle, we introduce two variants: Dirichlet-Multinomial Bagging Random Forest (DM) and Dirichlet-Weighted Random Forest (DW). Both modulate sample reweighting via a concentration parameter $α>0$. We provide a simple theoretical criterion that clarifies when these variants behave indistinguishably from standard random forests, and we use it to guide a lightweight tuning strategy. In a controlled evaluation on public classification benchmarks, DM and DW are consistently competitive and often stronger than other random-forest (RF) baselines, with negligible additional runtime. --- ### Understanding LLM Quantization through Activation-Guided Compensation and Orthogonal Residuals Authors: Yamato Narita, Issei Sato Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21450v1 Post-training weight-activation quantization reduces the memory and inference costs of large language models, but aggressive W4A4 quantization remains difficult because activation outliers degrade effective quantization resolution. Although weight optimization, channel-wise scaling, and orthogonal rotation mitigate this problem, the error components they address and their relationship remain unclear. Using an exact decomposition of local weight-activation quantization error into an activation-guided weight compensation term and an orthogonal residual, we bound the residual using persistent channel-wise outlier and regular activation quantities. This decomposition clarifies which error components can be addressed by weight compensation and which require transformation design. We then use the residual bounds to derive practical guidelines for applying randomized Hadamard rotation, sign selection, and channel scaling. In particular, the analysis explains how random signs suppress constructive interference among persistent outlier channels, how sampling multiple sign patterns can improve transformation selection, and how second-moment balancing leads to an $L_2$ scaling rule while a further relaxation recovers SmoothQuant-style $L_\infty$ scaling. We evaluate these guidelines through backpropagation-free configurations across eight Llama and Mistral models, obtaining performance competitive with gradient-trained SpinQuant. --- ### ME-Dex 1.0: Bringing Heterogeneous Tactile Sensing into World Action Modeling Authors: Xuancheng Zhang, Xuetao Liu, Qianying Tang, Jizhe Wang, Runsheng Wang, Zhijing Cheng, Bochen Lin, Haoran Wen, Ming Li, Kun Zhan, Yu Liu Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21449v2 World Action Models bring the predictive capabilities of video models into robot action generation, providing a rich foundation for modeling future visual states. Tactile sensing complements this foundation with direct measurements of physical interaction. Some existing methods use tactile features as conditioning inputs without jointly predicting future tactile states, visual observations, and actions. Our key insight is that tactile signals, like video, provide observations of the evolving world state and should be modeled as future observations alongside video. We present ME-Dex-1.0 (MachEmbodied-Dex-1.0), a unified World Action Tactile Model for joint visual, tactile, and action learning. ME-Dex-1.0 adopts a Mixture-of-Transformers architecture comprising a Video Expert, a Tactile Expert, and an Action Expert, all trained with flow matching. We use shared attention connects the experts in intermediate layers, allowing action generation to draw on learned representations of visual and tactile dynamics during joint denoising. To support multi-source heterogeneous tactile inputs, a Canonical Hand Model and a Unified Tactile Autoencoder map tactile observations from different embodiments and sensing layouts into shared spatial and latent spaces. To address the limited availability of paired visual, tactile, and action data, we develop the Agentic Tactile Data Engine, an agent-based data production platform. It supplements RoboTwin and DexJoCo with tactile data recorded directly from force sensors during trajectory replay in simulation. Experiments on the RoboTwin, DexJoCo, and ManiFeel simulation platforms, together with real robot evaluations, demonstrate improved manipulation performance using both grippers and dexterous hands equipped with tactile sensing. --- ### FootQuery: Future-Touchdown-Guided Retrieval from Depth History for Perceptive Humanoid Locomotion Authors: Tao Dong, Jia Yu, Yuxuan Fan, Linna Zhao, Jiaqi Gong, Andong Yang, Chao Gao, Guyue Zhou Published: 2026-09-18 Categories: cs.RO, cs.LG, eess.SY arXiv: https://arxiv.org/abs/2609.21447v1 Humanoid locomotion over complex terrain requires anticipating footholds that may no longer be visible at touchdown. Limited camera coverage and self-occlusion make it necessary to retrieve relevant terrain information from earlier observations. We present FootQuery, a perceptive locomotion framework that queries depth history using each foot's predicted next touchdown. The policy predicts touchdown locations and uncertainty from proprioception and uses these distributions, together with per-foot features, to query sparsely sampled historical depth frames. During training, realized contacts are projected into historical images to supervise retrieval at the regions where those contacts were visible. The retrieved per-foot features are fused with global visual memory to generate control actions. A progressive force-assistance curriculum supports early exploration, while event-consistent tread-midline shaping encourages coordinated stair contacts. Deployment requires only proprioception and onboard depth images. In simulation, the complete framework outperforms its component ablations on the most challenging tested stairs, gaps, and platforms. Real-world experiments on a Unitree G1 demonstrate continuous traversal with a single policy across outdoor stairs and indoor routes combining stair ascent and descent, platforms, and gaps. These results support organizing visual history around anticipated contacts for perceptive humanoid locomotion. --- ### Optimal Randomized Proper Online Learning Authors: Zachary Chase, Idan Mehalel Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21445v1 We prove that the optimal expected mistake bound of online learning a function class $\mathcal{H}$ by a randomized proper learning algorithm is $O(\mathtt{L}(\mathcal{H}) \log T)$, where $\mathtt{L}(\mathcal{H})$ is the Littlestone dimension of $\mathcal{H}$ and $T$ is the time horizon. Our result improves upon the previously best known bound of $O(\mathtt{L}(\mathcal{H}) \log^6 T)$ given by Daskalakis and Golowich (STOC 2022), and is optimal up to a universal constant for worst-case classes. --- ### Interference-Driven Clustered Optimisation for FM Spectrum Coordination Authors: Federica Mangiatordi, Emiliano Pallotti Published: 2026-09-18 Categories: cs.IT, cs.AI, cs.CE arXiv: https://arxiv.org/abs/2609.21441v1 Cross-border FM spectrum coordination involves protecting foreign broadcasting services while preserving domestic coverage, amid increasingly large radio-planning datasets containing thousands of transmitters and millions of transmitter-pixel relationships. In such scenarios, conventional optimisation approaches become computationally demanding due to the high dimensionality of the associated power-control problem. This paper proposes an interference-driven clustered optimisation framework for large-scale FM spectrum coordination. The proposed method exploits the observation that violations of foreign-service protection are typically dominated by a limited subset of transmitters. Protected services are therefore analysed to identify dominant interferers and quantify their impact on interference. These relationships are represented through an interference graph from which optimisation-oriented transmitter clusters are extracted. The clusters decompose the global power-control problem into smaller optimisation tasks solved with clustered simulated annealing, followed by a global refinement that captures residual inter-cluster interactions. Coverage and interference are evaluated using frequency-dependent protection criteria and a dynamic strongest-service assignment model. To enable operational-scale planning, the framework uses sparse matrices and GPU-accelerated computations. Tests on realistic cross-border FM coordination scenarios show that the clustering strategy greatly reduces optimisation complexity and runtime while maintaining foreign-service protection and domestic coverage. The method also yields an interpretable ranking of transmitters that contribute most to harmful interference, supporting optimisation and spectrum planning. --- ### Think Locally, Refine Globally for Memory-Efficient 3D Reconstruction Authors: Jingke Zhou, Chenhang Ma, Zhizhou Zhong, Mingkai Liu, Zhuang Zhou, Yicheng ji, Binghua Su, Bo Cai, Xianliang Huang Published: 2026-09-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.21437v1 We propose LoG-VGGT, a memory-efficient framework for long-sequence 3D reconstruction that balances local temporal modeling with global camera consistency. Instead of relying on full global attention, our method introduces cross-window attention at a small subset of transformer blocks, enabling effective information propagation across adjacent temporal windows while keeping memory usage bounded. To mitigate long-term pose drift, we further design a global camera consistency refinement module, where camera tokens interact with compact register tokens via cross-attention to enforce scene-level constraints across the entire sequence. This design enables joint optimization of camera representations and significantly improves long-horizon pose stability without incurring the high cost of sequence-wide attention. Extensive experiments demonstrate that LoG-VGGT achieves improved depth accuracy and robust camera pose estimation across multiple long-sequence benchmarks, while delivering competitive streaming reconstruction performance. --- ### GVPO++: Group Variance Policy Optimization for LLM Post-Training and On-Policy Distillation Authors: Kaichen Zhang, Yuzhong Hong, Junwei Bao, Hongfei Jiang, Yang Song, Dingqian Hong, Hui Xiong Published: 2026-09-18 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.21432v1 Post-training plays a pivotal role in enhancing the reasoning capabilities and task-specific expertise of large language models (LLMs). Despite recent advances in post-training methods, such as Group Relative Policy Optimization (GRPO), their practical deployment remains impeded by training instability arising from the reliance on importance sampling. We introduce Group Variance Policy Optimization (GVPO), a novel post-training method that integrates the analytical solution of KL-constrained reward maximization into its gradient weighting scheme. This formulation provides an intuitive interpretation: GVPO's gradient corresponds to the mean squared error between the central distance of implicit rewards and that of actual rewards. GVPO offers two key advantages: (1) it guarantees a unique optimal solution, exactly to the KL-constrained reward maximization objective, and (2) it enables flexible sampling distributions without requiring importance sampling. Beyond general post-training, we show that GVPO naturally extends to on-policy distillation (OPD). Furthermore, GVPO enables the optimization of a broad family of extended OPD objectives, providing a principled foundation for diverse objective design. By unifying theoretical guarantees with practical adaptability, GVPO establishes a new paradigm for reliable and versatile LLM post-training and on-policy distillation. --- ### Decision-Focused Learning for Mean-Variance Portfolio Optimization via KKT-Based Reformulation Authors: Kensei Nosaka, Shunnosuke Ikeda, Yuichi Takano Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21427v1 Mean-variance portfolio optimization (MVO) is a central framework in data-driven asset management. A widely adopted approach is a two-stage framework that first predicts expected returns and then solves the optimization problem based on these predictions, with the predictive models trained by minimizing prediction errors. However, this objective of prediction is not aligned with the quality of the downstream portfolio decision. Decision-focused learning (DFL), which directly minimizes the downstream decision loss within the learning process, has thus emerged as a promising direction. However, existing DFL approaches to MVO rely on surrogate losses or constraint relaxations for tractability, creating a structural mismatch between predictive model training and the constrained MVO solved at evaluation. We propose a single-level optimization formulation that incorporates the Karush-Kuhn-Tucker (KKT) optimality conditions of the lower-level MVO into the upper-level learning problem. This formulation explicitly preserves the budget and short-sale constraints while remaining tractable for standard nonlinear optimization solvers. Rolling-window experiments on real-world ETF (Exchange Traded Funds) data across two asset universes with different correlation structures show that our method achieved the best performance on multiple investment metrics and also demonstrated performance improvement due to the proposed regularization. --- ### Tracing the Evidence Behind Zero-Shot Time-Series Forecasting: A Source-First Taxonomy and Audit Framework Authors: Delun Kong, Wanyun Ling, Chenxi Liu, Ziyue Li Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21425v1 Zero-shot time-series forecasting (TSF) is often described as forecasting without target-specific parameter updates, but that training-status condition does not specify what evidence the system may use. A frozen language model prompted with serialized values, a time-series model pretrained on broad forecasting corpora, and a retrieval-augmented forecaster may all satisfy the no-update condition while drawing on different transferable evidence. This paper argues that zero-shot TSF should therefore be governed as an evidence-access claim. We propose a source-first taxonomy that separates three primary evidence sources---frozen LLM prior reuse, parametric time-series pretraining, and retrieval-augmented external memory---from the architectures that implement them. After the source is identified, four additional audit questions remain: task interface, forecast object and scoring, prediction-time context, and resource budget. The resulting agenda is to make zero-shot leaderboards auditable by reporting evidence boundaries and interface assumptions alongside scores, so that benchmark progress reflects transferable forecasting capability rather than undisclosed changes in context, memory, or budget. --- ### P$^3$-SAM: SAM with Perceptual Parallel Prompt for Few-Shot Strip Steel Surface Defect Segmentation Authors: Qian Xu, Hang Xiong, Anpeng Wang, Sam Kwong, Cong Zhang, Runmin Cong Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21424v1 Few-shot semantic segmentation (FSS) of strip steel surface defects (S$^3$D) has posed significant challenges distinct from natural scenes. Unlike natural images, S$^3$D task exhibits unique characteristics including low local contrast, uneven illumination, and complex fine-grained texture patterns. Although recent methods based on Segment Anything Model (SAM) have shown promise in FSS on natural images by leveraging SAM's powerful pre-trained representations, these unique industrial characteristics of S$^3$D images lead to performance drop when directly applying SAM to industrial defect scenarios. In this paper, we propose a novel Perceptual Parallel Prompt (P$^3$) framework that empowers SAM, creating the P$^3$-SAM model to address these challenges through two core strategies. First, we develop a Perceptual-Optimized Encoding (POE) strategy that enhances local contrast and preserves critical texture details for S$^3$D segmentation. Second, we introduce the Parallel Prompt Generator (PPG) strategy that simultaneously generates both semantic and spatial prompts, enabling comprehensive guidance for SAM's decoder across varying images. Extensive experiments on three few-shot S$^3$D benchmarks demonstrate that P$^3$-SAM achieves state-of-the-art performance, with particularly notable improvements of 12.00% in mIoU on Surface Defects-4i dataset. --- ### DENSE: Distilling Agent Trajectories into Evidence-Grounded Shortcut Trees for Self-Refinement Authors: Siyuan Liu, Fan Yu, Dongyu Ru, Yizhu Liu, Yifan Yang, Xuezhi Cao, Xunliang Cai, Yixin Cao Published: 2026-09-18 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.21423v4 Online agent deployments accumulate execution trajectories at massive scale and behavioral diversity, for which predefined annotation criteria hardly exist. Extracting useful evidence therefore demands costly manual annotation or verifier signals that fail to scale, leaving valuable evidence buried among redundant, incomplete, and failed executions. This raises a question: without post-execution rewards or correctness labels, how can reusable experience be distilled from the trajectories themselves? To address this challenge, we introduce DENSE (Distilling Evidence from Nested Subtask Executions), which organizes trajectory-derived evidence into nested shortcut trees. By consolidating redundant attempts, identifying resolved subtasks, and retaining useful steps alongside outstanding requirements, DENSE transforms noisy execution traces into structured and reusable task-solving feedback. To evaluate whether such feedback helps agents retry the same task, we design REFIT, which measures success-rate changes between the initial attempt and feedback-guided retries. Among feedback methods without external outcome supervision, DENSE achieves the highest strict pass rate across four agent models on Terminal-Bench 2.1, improving over initial attempts by 7.12-21.81 percentage points with 19.0-43.6% fewer agent tokens on retries. In addition, on hard tasks DENSE consistently outperforms self-reflection in cumulative pass rate across multiple feedback iterations on all four models, demonstrating its strong potential for continual agent self-improvement. --- ### Brownian Heads for Deep ReLU Representations: Activation Mass and the Cost of Same-Sample Selection Authors: Mahdi Mohammadigohari, Nicole Mücke Published: 2026-09-18 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.21422v1 Deep representation learning often selects hidden features and fits the final predictor on the same sample, so fixed-feature analysis performed after selection can omit selection cost. We study the conditional empirical Rademacher complexity of deep ReLU representations followed by bounded-norm predictors in additive or Lévy-Brownian RKHSs, termed Brownian heads. For a fixed representation, we derive an exact dual identity and sharp bounds in terms of activation mass, the average norm of the observed hidden vectors. Under same-sample selection, the representation supremum induces a quadratic Rademacher process. Brownian layer-cake and Gaussian-projection identities reduce it to coordinatewise or signed projected threshold traces, separating realized scale from selection complexity. For samples with pairwise-distinct inputs, explicit scalar ReLU families match the finite-trace and VC rates up to universal constants at the realized trace-and-envelope level. Induced-norm contraction also yields architecture-level bounds for rectangular, rank-deficient ReLU networks. Experiments verify the sharp bounds and rates, exhibit a selection gap at fixed activation mass, and assess the predictive feasibility of Brownian heads. --- ### When Online Adaptation Hurts: Parameter-Frozen Test-Time Ensembling for Continual Medical Image Segmentation Authors: Ruijie Huang Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21412v1 Medical image segmenters often get worse when sites, scanner vendors, or protocols change. Continual test-time adaptation (CTTA) addresses this problem without target labels, but it can be impossible to update a model on a non-stationary stream and can lead to a lot of errors. We examine a more reasonable and meaningful alternative: parameter-frozen inference enhancement(PIE). We use a source-trained segmenter that learns about anatomy-preserving scale and flip views, maps their predictions back to the native location, and averages the probabilities. We do not modify the weights of the model or the normalization statistics. On a cardiac MRI stream from M\&Ms, which is trained on vendor A and evaluated sequentially on vendors B, C, and D, PIE has 0.7786 mean Dice, compared to 0.7680 for source-only inference and 0.7388--0.7416 for five other online-adaptation baselines. The controlled ablations show that performance saturates at 28 views, and confidence weighting, class-prior correction, connected-component filtering, morphological refinement, and inter-slice smoothing have no effect or cause negative transfer. Qualitative results on cardiac MRI and fundus images are also consistent with the frozen ensemble keeping thinner and nested anatomical structures. These results provide a strong, stable baseline for medical CTTA and expose an important failure mode: adaptation and handcrafted refinement can be less reliable than carefully designed inference. --- ### Quantization-Aware Kalman Estimation for Diffusion Sampling Authors: Qitan Shi, Cheng Jin, Jiawei Zhang, Yuantao Gu Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21407v1 Quantization offers a practical path to deploying diffusion models with reduced memory and computation, but aggressive compression can cause quantized outputs to deviate substantially from their full-precision counterparts. Sampling-stage correction methods seek to compensate for such deviations during sampling, but existing approaches rely primarily on local information and underexploit trajectory history, limiting their ability to correct errors that propagate across timesteps. In this work, we formulate sampling with a quantized denoiser as an online estimation problem, using the history of quantized denoiser outputs to recover the underlying full-precision outputs required by the sampler. We propose QuAKE, a Quantization-Aware Kalman Estimator that combines a smooth trajectory prior with a conditional Gaussian observation model. At each sampling step, QuAKE recursively updates the posterior over the output window in closed form and feeds its posterior mean to the sampler. QuAKE is a lightweight plug-and-play corrector that requires no modification to the quantized network and naturally supports arbitrary high-order multistep ODE samplers. Experiments across W4A4-quantized text-to-image diffusion models show that QuAKE consistently outperforms existing methods in reducing the distributional discrepancy from full-precision sampling. --- ### SIRA: Reasoning-Aware Surgical Instrument Segmentation via Query-Anchored Alignment Authors: Zhibo Zhang, Qijie Wang, Zengqiang Yan Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21402v1 Surgical instrument segmentation (SIS) plays a critical role in robotic assistance and surgical workflow analysis. However, most existing SIS methods formulate segmentation as a category-driven localization problem, limiting their ability to capture procedural context and task-dependent semantics in surgical workflows. We introduce Reasoning-Aware Surgical Instrument Segmentation (RA-SIS), a task formulation that frames segmentation as query-conditioned inference under surgical context. To benchmark this setting, we construct SurgRS, a surgical reasoning segmentation dataset consisting of 41,000 image-text pairs, which aligns instance-level masks with structured query-answer supervision to enable semantic grounding at the pixel level. Based on SurgRS, we propose Surgical Instrument Reasoning and Segmentation Assistant (SIRA), a multimodal framework that disentangles target-level and query-level semantics and integrates them with visual features through query-anchored dual alignment. By aligning query semantics with spatial features and segmentation prompts, SIRA enhances semantic-visual consistency in mask prediction. Extensive experiments on SurgRS demonstrate improvements over existing reasoning-aware baselines. Code is available at https://github.com/linxir226/SIRA. --- ### Talking Past the Machine: Morality, Politeness, and Alignment in Human-AI Dialogue Authors: Marina Mitiaeva, Lu Xiao Published: 2026-09-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.21401v1 Conversational AI systems produce fluent, socially appropriate responses, yet whether they participate in cooperative communication or merely simulate its surface forms remains unclear - a question central to how these systems are evaluated, trusted, and designed. This study investigates how morality, politeness, and alignment - three dimensions central to cooperative dialogue - function in human-AI interaction compared to human-human conversation. We analyze 15,881 human-ChatGPT and 10,784 human-human multi-turn dialogues, using mixed-effects models to identify which features predict turn-to-turn alignment. We observe a consistent dissociation: AI produces the surface features of cooperative communication without the underlying social architecture. Moral output appears preconfigured rather than negotiated; warmth is generated without face sensitivity; linguistic convergence declines persistently. Most strikingly, the cooperative mechanisms themselves reverse direction: hedging and softening associated with greater accommodation between humans are associated with reduced alignment when produced by AI, and purity framing associated with human divergence coincides with users converging toward the AI. Agency - giving users room to shape the exchange - is the most consistent predictor of alignment across both interaction types, while lower moral assertiveness in more recent models is not accompanied by better cooperation. Together these patterns suggest that AI reproduces the surface of cooperation without the mutual adaptation that grounds it between humans - and, more surprisingly, that mechanisms sustaining human accommodation can run in reverse with AI, suggesting a turn-level view may be insufficient for interaction-level success. --- ### A Scene Language Model for Open-Vocabulary Scene Mapping Authors: Adam Lilja, Fabio Hübel, Siming He, Junsheng Fu, Claire Tomlin, Lars Hammarstrand, Jitendra Malik, Jonas Frey, Marco Pavone Published: 2026-09-18 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.21400v1 Open-vocabulary 3D scene mapping aims to build a persistent representation of the objects in an environment. Existing systems typically rely on engineered mapping pipelines to associate observations, merge information across views, and maintain a consistent scene representation over time. Many additionally store feature-rich object representations, such as embeddings or image crops, increasing the size and complexity of the persistent memory. We introduce SceneLM, a Scene-Language Model that directly maintains a textual scene map. The full scene is represented as a structured text list of objects, which serves as the model's only persistent memory. For each input image, the model reads the current scene state and updates the map by adding, editing, and removing objects. To learn this behavior, we introduce supervision tasks for iterative scene map maintenance together with an automatic annotation pipeline that generates training data from images without human labels. We evaluate SceneLM on both a language-grounded retrieval benchmark and a localization benchmark. Across both benchmarks, the model produces a scene map that achieves competitive performance with complete mapping systems built from dedicated perception and geometric modules while producing a scene representation that is 6-12x more compact. We further show that SceneLM can be run online on an edge device through experiments on a quadruped. These results show that a persistent open-vocabulary 3D scene map can be maintained directly by a single vision-language model using only a lightweight text representation. Training and inference code is available on https://goldengait.github.io/scenelm/. --- ### Omni Demand Understanding: A Benchmark for Contextual User-Intent Inference in Multimodal Interaction Authors: Qi Chen, Yunfei Chu, Haolin He, Yifan Yang, Zihan Liu, Yuxuan Wang, Ziyang Ma, Ruiyang Xu, Meng Gao, Yinsong Yan, Ling Wang, Hui Wang et al. Published: 2026-09-18 Categories: cs.CL, cs.CV, cs.MM, cs.SD arXiv: https://arxiv.org/abs/2609.21392v1 Natural audio-visual interaction is emerging as an important interface for AI assistants, allowing users to communicate through speech and vision rather than carefully composed text prompts. However, existing benchmarks of interactive capabilities still focus primarily on response quality, leaving a more fundamental question underexplored: can a model correctly infer the user's underlying demand from complex multimodal interaction? Real-world user demands are often underspecified in speech and must be inferred from multimodal cues and dialogue history. This inference is further complicated by ambiguous or disfluent expression and noisy acoustic environments. Conversely, request-like speech may not constitute a demand to the assistant, leading to false triggers. We establish Omni Demand Understanding (ODU) as a distinct multimodal contextual inference problem: given an interaction stream, a model must detect whether a user demand is present and infer intent from multimodal and conversational context. ODU evaluates this capability along five dimensions, covering both single-turn and multi-turn interactions. We construct ODU-Bench using a challenge-driven taxonomy, taxonomy-guided agentic video generation, and human-recorded interactions, followed by media-grounded annotation and human verification. We evaluate 14 native MLLMs. Even the strongest, Gemini 3.1 Pro, recovers only 44.7% of key information that must be inferred from visual, acoustic, or conversational context. Moreover, 11 of the 14 models exhibit false-trigger rates above 50% on non-demand scenarios. These results reveal a systematic capability gap in current MLLMs' ability to infer contextual user demands. We hope ODU can establish the evaluation of a previously underexplored yet essential capability in multimodal interaction: correctly understanding user demands before generating an appropriate response. --- ### WS-NeRF: A Mamba-Driven World-State-Aware Adaptive Deblurring Neural Radiance Field Authors: Hang Jiang, Jinghao Wang, Yiming Zhang, Xinhong Wang, Luwei Ran, Yinfeng Yu Published: 2026-09-18 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2609.21391v1 Neural Radiance Fields (NeRF) have attracted extensive attention in recent years due to their strong capability for high-quality 3D reconstruction and novel view synthesis from multi-view images. Existing methods usually rely on high-quality sharp inputs, while real-world image acquisition is highly susceptible to blur degradation, which severely affects the reconstruction quality of NeRF. In this paper, we propose a novel Mamba-driven world-state-aware adaptive deblurring neural radiance field, termed WS-NeRF, to address image degradation and 3D inconsistency. We formulate the alternating optimization of radiance fields as a dynamic evolution process with temporal memory, and jointly exploit comprehensive multi-dimensional world states and a mixture-of-experts mechanism to dynamically adjust the confidence of deblurring priors. Experimental results show that WS-NeRF significantly improves blurry radiance field reconstruction quality, achieving better performance on PSNR, SSIM, and LPIPS, while exhibiting more stable iterative recovery behavior. --- ### Offline Multimodal Large Language Models for Decision Support in Air Operations Authors: Joao P. A. Dantas, Jelton A. Cunha, Gabriel Dietzsch Published: 2026-09-18 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.21390v1 Air operations rely on complex rules, established procedures, and time-critical analysis under limited connectivity and strict security constraints. In such environments, analysts must combine written doctrine with images, often without access to external computing resources. This paper studies offline large language models as decision support tools, deployed in isolated and restricted environments to give analysts access to doctrinal knowledge that remains traceable to its original sources through natural language interaction. We describe a modular retrieval-augmented architecture suitable for operation without Internet connectivity, supporting both text and image input from technical manuals. As a first step toward evaluating this architecture, we report a pilot study with four image analysts of the Brazilian Air Force, combining (i) a doctrinal knowledge assessment based on their electronic-target identification doctrine, comparing human and proposed system performance on the same test, and (ii) a measurement of the cognitive workload involved in manually producing a reconnaissance target report (Relatório de Missão de Reconhecimento - REMIR) without AI assistance. The results show a demanding manual task, especially in terms of mental demand (6.0/7) and effort (5.0/7), while the proposed system matches the human score (8/10) and completes the assessment in 7.1 minutes (compared to a human average of 26.5 minutes), establishing a baseline for future AI-assisted evaluation. Finally, we describe a future evaluation protocol to systematically compare manual and AI-assisted workflows. --- ### Consistent Relexicalization of Clinical Documents using Graph-Based Approach Authors: Dipankar Das, Atri Mandal, Sandeep Singh, Tushar Shandhilya Published: 2026-09-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.21387v1 Relexicalization is a pivotal technique in clinical NLP, as it facilitates robust masking of sensitive information while synthesizing datasets that retain high-fidelity, real-world characteristics. However, preserving structural integrity, relational coherence, and temporal consistency during transformation remains a significant challenge. Existing approaches frequently rely on independent entity replacement, which results in clinical inconsistencies across longitudinal records. This reduces the value of such relexicalized datasets for downstream scientific analysis. To address these limitations, we introduce G-RELIC (Graph Based Contextual Relexicalization with Improved Consistency) which combines the power of LLMs with graphs. G-RELIC implements a graph-based mapping mechanism which optimizes for one-to-one correspondence between original and surrogate entities. It also introduces a deterministic temporal repositioning algorithm to preserve temporal consistency. Empirical evaluations on diverse, real-world clinical datasets validate that G-RELIC significantly outperforms state-of-the-art baselines. G-RELIC yields a 30.4 percentage point improvement in relational integrity (62.1% to 92.5%) and 45.9 percentage point improvement in temporal coherence (46% to 91.9%) without compromising on the recognized privacy benchmarks for clinical datasets. This maximizes the analytical utility of relexicalized datasets while minimizing re-identification risk. --- ### AgentVidBench: A Multi-Hop Video Question Answering Benchmark for Evaluating MLLM Agents Authors: Seoyeon An, Hyeonseo Jang, Minsu Kim, Chanho Lee, Younghan Park, Kangwook Lee Published: 2026-09-18 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.21386v1 Comprehensive video understanding is crucial for advancing artificial intelligence toward the intricate dynamics of the physical world. While recent advances in Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities in video understanding, existing benchmarks remain confined to simple scene-level queries or global summaries that require only single-step inference. Real-world video understanding involves more challenging tasks that require multi-hop multimodal reasoning, and there is a critical absence of video benchmarks equipped to rigorously evaluate these agentic capabilities. To bridge this gap, we introduce AgentVidBench, a multi-hop video question answering benchmark focused on evaluating the spatial, temporal, and causal reasoning capabilities of MLLM agents. Beyond standard question-answer pairs, AgentVidBench provides step-by-step solution traces to support trajectory evaluation that assesses whether agents explicitly acquire the evidence needed to justify their answers. Experiments with 12 proprietary and open-source MLLMs show that single-turn performance remains limited on AgentVidBench, while integrating these models into state-of-the-art agentic workflows generally improves performance with respect to both accuracy and trajectory scores. We further present a simple yet effective agentic strategy that serves as a competitive baseline on AgentVidBench, establishing our benchmark as a holistic testbed for future research on agentic video understanding. Code and datasets are available at https://github.com/krafton-ai/agentvidbench and https://huggingface.co/datasets/agentvidbench/agentvidbench. --- ### Prediction Dynamics in Depth-Recurrent Language Models Authors: Xinyue Luo, Fei Yu Published: 2026-09-18 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.21383v1 Depth-recurrent language models refine predictions through repeated latent updates. Why can intermediate answers agree with the endpoint while their scores continue to change? We derive a sharp margin characterization that decomposes the conservatism of a magnitude bound into common translation, direction relative to the winner, and the pairing of each competitor's update with its score gap. Across Huginn-3.5B and Ouro-1.4B, accounting for update direction and competitor pairing reduces the mean earliest qualifying depth by a further 22.5-34.4% of the total depth beyond translation removal under full answer-text scoring. This retrospective comparison uses completed trajectories. Substantial contributions also occur under label scoring. For shared predictive distributions, we separate common and contrast motion orthogonally and express the common component through candidate-set mass and within-set concentration. Common and contrast energies can attenuate at different rates, allowing a growing preference-change share to coexist with shrinking absolute updates. These findings explain finite-depth answer preservation through the geometry and composition of observed score changes. --- ### Probabilistic Forecasting of Business Process Executions with Neural Temporal Point Processes Authors: Jiaxin Yuan, Daniela Grigori, Han van der Aa Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21382v1 Operators of service-based systems act on forecasts of how a running execution will continue, and such a forecast is actionable only if its reliability is known. Mainstream deep-learning models for this task are discriminative and deterministic: they emit a single next activity and a single remaining-time estimate, without a distribution to reason over. We instead cast the problem as generative sequence modelling with marked temporal point processes, which define a joint density over the next mark and its inter-event time and therefore deliver predictive distributions by construction. Real event logs violate the simple-point-process assumption these models rest on, since consecutive events frequently carry identical timestamps; we handle such ties explicitly and combine a transformer encoder with a mixture decoder over inter-event times, trained by exact log-likelihood. On ten public logs, the resulting model matches discriminative baselines on point accuracy, dominates them on the calibration and sharpness of remaining-time distributions, and is the cheapest at inference, since a full predictive distribution is obtained in a single forward pass without sampling. --- ### Knowledge-Graph-Augmented Chronos-2 for HEC-RAS Surrogate Forecasting Authors: Edward Holmberg, Elias Ioup, Mahdi Abdelguerfi Published: 2026-09-18 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.21381v1 We investigate whether coupling a time-series foundation model to hydraulic project knowledge improves surrogate forecasting of HEC-RAS water-surface elevation (WSE). We present KG-Chronos-2, which combines a frozen Chronos-2 predictor with exact-state residual decoding, graph-conditioned historical retrieval, and input-aligned correction. We compare the method with persistence, a residual LSTM, project-conditioned recurrent GeoFNO, a hydraulic DCRNN-style model, and frozen Chronos-2. Task-specific fitting uses the 2008 simulation. Evaluation covers 64 fixed 24-hour windows from the 2011 and 2002 simulations at 4,675 cross sections in 71 reaches on a shared geometry. KG-Chronos-2 achieves event-balanced root-mean-square error 0.246970 in native WSE units. It reduces RMSE by 14.13% relative to frozen Chronos-2, 29.38% relative to the hydraulic DCRNN-style model, and 39.54% relative to recurrent GeoFNO. The 95% hierarchical-bootstrap interval for its event-balanced RMSE difference from frozen Chronos-2 is (-0.075177, -0.016317). KG-Chronos-2 also achieves the lowest active-window and final-lead RMSE among the six completed systems. These results support coupling a frozen temporal predictor to project knowledge for warm-start HEC-RAS forecasting on the fixed benchmark. --- ### JEPA Guided Diffusion: Predictive Vision-Language Conditioning for Generative Traffic Forecasting Authors: Trinh Tra Giang Nguyen, Thanh Nguyen Vo, Nguyen Hoai Thuong Bui, Ha Duc Bui Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21379v1 Accurate traffic forecasting requires both understanding scene dynamics and synthesizing realistic future observations. Recent diffusion-based video generation models produce visually plausible predictions but require expensive end-to-end training and often entangle scene understanding with image synthesis. In this work, we propose a decoupled forecasting framework that separates future representation learning from video generation. A frozen V-JEPA encoder first extracts predictive latent representations from the observed traffic videos, capturing the underlying scene dynamics in a semantic latent space. A lightweight latent alignment module then projects these representations into the conditioning space of a frozen Cosmos diffusion module, enabling future video synthesis without retraining the large generative model. By freezing all foundation models and training only the lightweight alignment module, the proposed framework substantially reduces optimization complexity while preserving forecasting capability. Experimental results on the AI City Challenge 2026 Track 5 benchmark demonstrate that the proposed method achieved a score of 75.1297, ranking third in the competition. These results suggest that predictive world representations learned by V-JEPA can effectively guide downstream video generation, providing a practical and efficient alternative to end-to-end diffusion-based forecasting. --- ### ArenaFlow: From Trajectory Ranking to Hierarchical Credit Propagation for Open-Ended Agent RL Authors: Qiang Zhang, Ruixue Ding, Fanrui Zhang, Xi Chen, Boli Chen, Shihang Wang, Yinfeng Huang, Yi Zheng, Pengjun Xie, Kaipeng Zhang, Jiawei Liu, Zheng-Jun Zha Published: 2026-09-18 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.21378v1 Reinforcement learning has substantially improved large language model (LLM) agents in verifiable domains, but remains difficult to apply to open-ended agent tasks, where solutions are diverse and reliable scalar rewards are hard to obtain. Recent pairwise evaluation methods alleviate reward discrimination collapse by replacing pointwise scoring with relative preferences. However, they still compress rich comparative feedback into a single trajectory-level reward, obscuring decisive intermediate steps and preventing successful behaviors from being consolidated into reusable skills. We propose ArenaFlow, a hierarchical credit propagation framework for open-ended agent reinforcement learning. ArenaFlow leverages tournament-based relative ranking to derive trajectory-level reward signals. Each comparison is further equipped with structured reflective evaluation, which reveals three types of supervision: pivotal success steps, reusable strategy skills, and usage attribution of retrieved skills. At the step level, ArenaFlow propagates trajectory-level advantages to high-confidence pivotal steps according to tournament survival depth, enabling more targeted optimization of local reasoning behaviors. At the skill level, ArenaFlow estimates skill utility from group-level usage attribution and maintains a global skill memory through utility-aware updating, pruning, and retrieval. The resulting high-utility skills further serve as policy priors for future exploration. Extensive experiments validate ArenaFlow's effectiveness on open-ended agent tasks. --- ### RobotEQ-Video: A Video-Centric Benchmark for Social Proactive Intelligence with World-State Taxonomy Authors: Xinyi Che, Zheng Lian, Kuofei Fang, Xuehao Wang, Xinghai Gao, Junqing Wu, Chuyu Wu, Liyi Liu, Yanhan Huang, Keyi Xie, Haomin Ouyang, Jinyang Wu et al. Published: 2026-09-18 Categories: cs.CV, cs.HC arXiv: https://arxiv.org/abs/2609.21371v1 Social Proactive Intelligence (SPI) extends proactive assistance beyond task completeness to consider social appropriateness in diverse embodied scenarios. However, prior SPI research faces two key limitations. First, existing work focuses on static images, whereas dynamic videos provide crucial cues for inferring human states and needs, offering richer information than isolated images. Second, prior work often relies on free-form data collection pipelines, which fail to guarantee comprehensive coverage of diverse scenarios. To address these gaps, we introduce RobotEQ-Video, shifting the focus from image-centric to video-centric analysis. To ensure comprehensive video coverage, we construct a hierarchical world-state taxonomy organized into a four-level coarse-to-fine structure, comprising 6 domains, 20 dimensions, 142 level-1 attributes, and 816 level-2 attributes. The resulting benchmark comprises 2K+ videos with 100K+ human annotations and 16K+ labels for assessing behavior properness. Benchmark evaluation reveals that current systems remain unreliable and fall short of human performance. We further explore how world models can help tackle this task. This work advances SPI research from static images to dynamic videos and ensures more comprehensive scenario coverage during benchmarking. --- ### ProTracer: Proprioception-Guided Failure Diagnosis in Robot Manipulation Authors: Chang Dong, Mehdi Hosseinzadeh, King Hang Wong, Lingqiao Liu, Francois Fraysse, Feras Dayoub, Minh Hoai Nguyen Published: 2026-09-18 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.21369v1 This paper presents a comprehensive framework for robot manipulation failure analysis that includes binary failure detection, failure categorization, explanation generation, and the additional capability of failure onset localization, which aims to identify the earliest moment at which a robot execution deviates from a valid task-completion trajectory and is ultimately followed by task failure. To address these tasks, we propose ProTracer, a training-free framework that leverages existing Vision-Language Models (VLMs) together with proprioceptive signals for failure analysis. Our method uses proprioceptive dynamics to identify temporally informative action boundaries and converts richer robot-state signals into structured natural-language descriptions that can be jointly analyzed together with visual observations by the VLM. This design combines the temporal precision of proprioceptive signals with the multimodal reasoning capabilities of modern VLMs without requiring additional model training. We further introduce FailTime, a benchmark with synchronized visual and proprioceptive observations for evaluating conventional failure diagnosis tasks as well as failure onset localization. Experiments demonstrate that ProTracer achieves strong performance across both conventional failure diagnosis tasks and the newly introduced failure onset localization task, highlighting the importance of proprioceptive reasoning for fine-grained temporal failure analysis. --- ### Hiding in Plain Sight: A Diffusion-based Mitigation of Geolocation Privacy Leakage in Vision-Language Models Authors: Yining Wang, Xi Li, Mi Zhang, Xiaohan Zhang, Xiaoyu You, Zhenxing Qian, Mi Wen Published: 2026-09-18 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.21363v1 Multimodal large reasoning models (MLRMs) have demonstrated remarkable capabilities in complex visual understanding. However, this very power introduces a critical yet underexplored privacy threat: adversaries can exploit MLRMs to precisely infer users' geographic locations from casually shared photographs, by performing structured reasoning over subtle visual cues such as architectural styles, vegetation, and lighting conditions. In this work, we present a systematic study of MLRM-driven geolocation privacy leakage. We first reveal that refusal-based safeguards are critically insufficient, as carefully crafted jailbreak prompts can raise model response rates to 100%. We further identify that existing defenses, which inject imperceptible perturbations into shared images, suffer from structural limitations intrinsic to their pixel-space optimization, resulting in degraded black-box transferability and pronounced visual artifacts. Motivated by these findings, we propose a diffusion-based framework that provides targeted, proactive defense against geolocation privacy leakage. By injecting perturbations into the latent space of a diffusion model during reverse sampling, our method operates directly on high-level semantic representations, thereby resolving the effectiveness-utility bottlenecks by construction. We further ground our optimization with GeoCLIP, a model explicitly aligned with GPS coordinates, as a surrogate to pinpoint and disrupt the geographic signals that MLRMs exploit for location inference. This targeted semantic disruption yields significantly stronger black-box transferability while preserving perceptual image quality, offering a seamless integration on social media platforms. --- ### Beyond Atomic Tokens: Factorizing Syllables for Language Model Pretraining Authors: Nghia Hieu Nguyen, Thai Bao Huynh, Binh-An Dinh-Le, Phu Gia Hoang, Dat Tien Nguyen, Kiet Van Nguyen, Ngan Luu-Thuy Nguyen Published: 2026-09-18 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.21362v2 Conventional tokenizers represent text as characters or statistically derived subwords, overlooking the internal phonological structure of syllables and often requiring large vocabularies. We introduce \textbf{Phonemic Tokenizer}, a linguistically motivated tokenizer for Vietnamese and Chinese that converts each syllable into IPA and factorizes it into three phonological components: onset, rime, and tone. The three components jointly occupy one contextual position, preserving syllable-level sequence length while enabling representation sharing across phonologically related syllables. Non-phonological and unsupported units are handled through character-level fallback. This deterministic design requires no corpus-dependent vocabulary learning and yields vocabularies of only 112 entries for Chinese and 256 for Vietnamese. Intrinsic evaluation shows that the tokenizer achieves substantially higher Rényi efficiency in both languages, represents every entry in a standard Vietnamese syllable dictionary with a Fertility of exactly one, and generally produces shorter Vietnamese sequences than existing pretrained tokenizers. We further instantiate the tokenizer in \textbf{PhonemicBERT}, which combines factorized component embeddings and reconstructs complete masked syllables using three prediction heads. Under a controlled Chinese pretraining setup, PhonemicBERT-Zh is competitive with or outperforms character, subword, and SubChar alternatives across diverse language-understanding tasks. PhonemicBERT-Vi also achieves competitive or superior results to established Vietnamese and multilingual pretrained models. These results establish phonemic factorization as a compact, efficient, and interpretable alternative to atomic and statistically segmented text representations. --- ### Field Tracking of Insects Using a Stereoscopic Event-Based Camera Setup Authors: Pratham G. Shenwai, Martin J. Lankheet, John T. Hrynuk, Mandiyam Y. Mahadeeswara, Mandyam V. Srinivasan, Sridhar Ravi Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21354v1 High-speed tracking of small, fast-moving organisms in their natural environments is important to better understand their behavior and ecology. Traditional frame-based imaging suffers from motion blur due to low temporal resolution, and data storage limitations, propelling a search for more adaptive solutions. Event cameras, which capture changes in brightness at pixel level instead of entire frames, have emerged as a promising solution by increasing temporal resolution and data efficiency. Here, we demonstrate the use of event-based imaging with standard video-based processing methods by converting the asynchronous events into conventional video formats, allowing us to leverage the event camera's enhanced temporal detail to capture intricate insect flight movements and apply established image analysis techniques. Coupling this conversion process with a stereoscopic configuration provides continuous, low-latency, three-dimensional tracking of fast-moving subjects in field conditions. As a result, we substantially mitigate motion artifacts and achieve more accurate representations of animal movements. By making event-based imaging more readily applicable in natural field settings, our method support broader applications across animal behavior and ecological research, agricultural management, and other fields requiring high-fidelity object tracking in the wild. --- ### PrismAlign: Prior-Steered Multi-View VLM Alignment for Hallucination-Robust Table OCR Authors: Guangyi Liu, Qianjun Huang, Boyu Hou Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21351v1 Table extraction suffers from frequent structural errors and semantic hallucinations. We propose PrismAlign, a multi-VLM framework aligning diverse visual perspectives to resolve ambiguity. It integrates priors of table logic to assess output plausibility, decoupling structural alignment from cell content alignment. A Bayesian decision strategy maximizes alignment accuracy by exploiting the correlation between extraction errors and computable rule violations. Evaluated on open-source and custom VLMs, PrismAlign reduces hallucinations and achieves state-of-the-art performance on OmniDocBench 1.5, as well as on the table category of CC-OCR and PureDocBench. --- ### From Memory to Behavior: A Behavior-Aware Role-Playing Framework for Social Media Influencers Authors: Ji-Lun Peng, Yi-Zhen Zhang, Chun-Nan Chou, Yun-Nung Chen Published: 2026-09-18 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.21349v1 Large language models have shown strong potential as role-playing agents for real individuals, yet faithful impersonating remains challenging. Existing in-context learning-based methods fail to capture how individuals react under different situations. In addition, LLM-based evaluation is difficult for obscure individuals. To address these challenges, we propose Situation--Internal state--Behavior Persona method to incorporate situation-dependent behavioral strategies. We further design an evaluation protocol that provides LLM evaluators with references about the impersonated individual. We evaluate our approach on a newly constructed dataset for the task of generating replies on social media. Experimental results show that our proposed method outperforms state-of-the-art ICL-based baselines, while our evaluation protocol achieves moderate correlation with human judgment. Besides, experiments on fictional-character benchmarks demonstrate that our proposed method is applicable beyond the social media setting. These findings suggest that incorporating behavioral information broadly improves the fidelity of role-playing for real individuals on social media or fictional characters. --- ### Cube-Splat: High-Fidelity 360° Gaussian Splatting SLAM via Cubemap Factorization and Adjoint-Consistent Optimization Authors: Xiangfei Guo, Hao Shi, Yufan Zhang, Zhonghua Yi, Yongqi Mao, Xiaoting Yin, Kaiwei Wang Published: 2026-09-18 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.21347v1 Recent progress in 3D Gaussian Splatting (3DGS) has enabled dense visual SLAM with pinhole cameras, yet most pipelines are not designed for panoramic imagery. We present Cube-Splat, the first panoramic GS-SLAM framework that factorizes each 360° frame into a cubemap of four fixed-orientation virtual pinhole views sharing a single optical center. By designating the front face as the primary pose state, we accumulate gradients from all faces via an adjoint mapping, thereby enabling multi-face observations to coherently update a single state while strictly preserving cross-view geometric consistency. Concurrently, our mapping module densifies and optimizes anisotropic Gaussians using aggregated cubemap rays for high-fidelity, dense reconstruction. Furthermore, to rigorously evaluate panoramic SLAM under diverse and challenging conditions, we introduce SynPano, a highly scalable, photorealistic synthetic dataset featuring parameterized complex trajectories and multi-modal ground truth. Extensive evaluations on two public benchmarks (PALVIO and OmniBlender) and our SynPano dataset, collectively encompassing both indoor and outdoor scenes, demonstrate that Cube-Splat achieves state-of-the-art (SOTA) performance in tracking accuracy and reconstruction fidelity. Both the source code and the SynPano dataset are available at https://github.com/guoxf304/CubeSplat. --- ### IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation Mixture-of-Experts Authors: Ran Cheng, Longfei Xu, Zheng Liu, Kaikui Liu, Xiangxiang Chu Published: 2026-09-18 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.21346v1 Mixture-of-Experts (MoE) scales capacity, but existing designs cannot set three quantities independently. For a single token, participation is how many experts contribute knowledge to its output, execution is how many are actually computed (compute cost), and materialization is how many expert-sized parameter sets must be built and stored (memory cost). Sparse routing keeps execution and materialization low, but shrinks participation: for each token, only a few experts contribute. Dense output-mixing restores full participation, but its execution grows with the number of experts. Parameter-merging keeps execution at one expert, but its materialization grows with the number of routing decisions. We propose IntBMoE, a block-conditioned MoE that decouples all three by pairing dense expert composition with sparse block execution. Its blocks come from a small learned codebook, one per entry. At each internal layer, a lightweight hypernetwork merges all expert bases in that layer's pool into one composed expert. Participation is full, because every composed expert draws on the entire pool. Execution stays sparse, because a router sends each token to only a few blocks. Materialization is bounded, because the codebook, not the input, fixes how many blocks exist. Dual-Path Residual Gating (DPRG) further couples two independently composed paths through multiplicative gating. Experiments on image classification show consistent gains over representative sparse and dense MoE baselines. Additional experiments on language modeling and sequential recommendation validate its generalization beyond vision. IntBMoE is fully deployed in AMap's generative recommendation system, serving hundreds of millions of users under a 60ms latency budget, with a 2.4% relative UVCTR gain in online A/B testing. Our code is available at https://github.com/AMAP-ML/DreamX-Rec/. --- ### CESBench: Benchmarking Large Language Models on Cryptographic Engineering Security for IoT Devices Authors: Wenquan Zhou, An Wang, Jing Liang, Peien Feng, Jingqi Zhang, Yaoling Ding, Liehuang Zhu Published: 2026-09-18 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.21344v1 For Internet of Things (IoT) devices, a secure algorithm alone is not enough: an attacker with physical access can attack the implementation directly, and its flaws are hard to fix once deployed. Large language models (LLMs) are now used to build and analyze such implementations. LLM benchmarks exist for cryptography and general cybersecurity, but none covers cryptographic engineering. In this paper, we present CESBench, 380 expert-written items across six sub-domains of cryptographic engineering security for IoT devices: side-channel, fault injection, implementation, countermeasures, evaluation, and integration. Four task types target different competences: 209 multiple-choice items test recall, 67 judgment items require a security verdict and its justification, 63 scenario items require an engineering diagnosis, and 41 code tasks are graded by 572 test cases. To validate the benchmark, 11 open-weight and proprietary LLMs answer every item. Multiple-choice and code responses are scored automatically, and judgment and scenario responses by an LLM judge, whose scores are checked against a second judge from another model family and human re-scoring. Composite scores range from 54.4% to 83.6%. The top score on each task type is 98.6% for multiple choice, 95.1% for code, and 88.4% for scenario diagnosis, but only 58.8% for judgment. Across models, 88.5% of verdicts are correct, yet their justifications earn only 53.4% of the rubric marks. Multiple choice is near its ceiling for the strongest models and most code tasks are solved, whereas justifying a security verdict remains the weakest competence. The benchmark, prompts, and per-item results are public. --- ### Robust Dual-Regularized Variable Selection under Outlier Contamination Authors: Abdul-Nasah Soale, Adewale F. Lukman, Essoham Ali Published: 2026-09-18 Categories: stat.ME, stat.CO, stat.ML arXiv: https://arxiv.org/abs/2609.21342v1 Real data often contain unusual observations that can exert disproportionate effects on variable selection, especially in complex predictor settings. We propose a two-stage {\it sparse median outer product of gradients (smOPG)} method for variable selection in single index models with outlier contamination. We first estimate sparse local gradients via \(\ell_1\)-penalized local median regression and then recover the active predictor set from a rank-one sparse approximation of the resulting gradient matrix using regularized singular value decomposition. The combination of median regression and local weighting provides robustness to both response outliers and leverage points. Extensive simulations across varying dimensions and contamination mechanisms demonstrate the favorable variable selection performance of smOPG relative to existing methods. Applications to air pollution and genomic data demonstrate practical utility, while theory establishes active-set recovery without requiring selection consistency of individual local regressions. --- ### Conformal Privacy Auditing: Calibrated Re-identification Attacks with Statistical Guarantees Authors: Shuo Huang, Gholamreza Haffari, Xingliang Yuan, Ting Yu, Lizhen Qu Published: 2026-09-18 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2609.21340v1 Empirical identity leakage from released text is increasingly driven by attackers that combine large language models (LLMs) with auxiliary knowledge to link documents to individuals. Existing audits typically report success rates for specific attack pipelines but lack finite-sample statistical guarantees, while training-time protections such as differential privacy are difficult to translate into release-time decisions for individual natural-language documents. We introduce Conformal Privacy Auditing(CPA), a distribution-free calibration framework that provides a statistical certificate of re-identification risk for each released document against LLM-empowered adversaries. CPA outputs a conformal ambiguity set of candidate identities that is guaranteed to contain the true identity with user-chosen confidence under exchangeability, together with an interpretable leakage proxy derived from set size. CPA supports both logit-access and sampling-only attackers, enabling audits of open-source models and proprietary API models in a unified framework. Across multiple release benchmarks and attacker configurations, CPA achieves calibrated coverage and reveals sharp shifts in certified identifiability as auxiliary knowledge, LLM augmentation, and release mechanisms vary, providing a statistically grounded basis for reporting and comparing release-time linkage risk across attacker configurations, datasets, and release mechanisms alike. --- ### Co-Evolving Zero-Day Jamming: Adaptive Attack Synthesis and Graph Attention-Based Online Detection Authors: Ghilas Aissou, Rémi A. Chou, Taejoon Kim Published: 2026-09-18 Categories: cs.IT, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2609.21334v1 Effective evaluation of zero-day jamming detectors requires robust adversarial models. However, existing attack models often assume prior knowledge of the target receiver, limiting their utility as evaluation benchmarks. On the detection side, existing detectors fail to capture the global temporal-spectral structure of jamming behavior and cannot differentiate zero-day strategies as they emerge. This paper addresses these limitations through a two-pronged framework. First, an online detection framework is introduced that combines a graph attention network (GAT) for temporal-spectral representation learning with Dirichlet process (DP)-means clustering. This framework jointly classifies known and discovers zero-day strategies within a unified learning objective. Second, an inference-driven reinforcement learning (RL) jammer is proposed as an adversarial benchmark. The jammer treats the target receiver as a black-box, infers the detector state via hypothesis testing, and optimizes the trade-off between attack impact and stealth. Simulation results show that the proposed RL jammer outperforms benchmarks, achieving 33% higher attack efficacy and 67% higher stealth. The proposed detection framework against the proposed RL jammer is shown to achieve 20% higher detection accuracy than the benchmarks. ---