Attention
-
CW-WAM: A Structured Contact-Wrench World Action Model for Force-Sensitive Manipulation
2026-10-21 17:55:53Houzhu Ding, Xinda Qi, Luqi Wang, Ming YangAbstract
Force-sensitive manipulation requires contact and load information that is not provided explicitly by visual observations. We present CW-WAM, a world action model that jointly predicts video, motion, and contact wrench. Inspired by hybrid motion-force control, it contains a unified interaction module that treats both motion and contact wrench as modeling targets and predicts time-aligned motion-wrench chunks as control references. A pair-causal attention mechanism structures the motion-wrench dependencies, while a long fixed-length wrench memory retains past interaction information. The predicted wrench chunk acts as a force reference for a 20-Hz outer-loop admittance controller within a slow-fast control system that maintains force feedback and smooth motion between model queries. CW-WAM improves contact prediction and increases the insertion success rate from 25% to 70% for square pegs with 1-mm clearance per side and from 40% to 95% for power adapters, compared with WAM without wrench input. It also improves lifting-height accuracy across different loads. Our framework with an admittance controller reduces force-reference RMSE from 4.8 to 2.7 N for peg insertion and from 2.0 to 1.3 N for wiping, respectively, with smaller force fluctuations during successful execution. Project website: this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.30117
PDF
-
Attn-Forcing: Learning Full-Context Attention Proxies for Streaming Video Generation
2026-10-21 16:59:51Junyao Gao, Sibo Liu, Jiaxing Li, Yanan Sun, Hanxin Zhu, Weidong Zhang, Jun Zhang, Cairong ZhaoAbstract
Increasing the attention context length significantly improves streaming video generation quality, while incurring prohibitive memory and computational costs as the context grows. Existing methods attempt to capture this contextual information through attention selection or compression, but remain limited by bounded representations and lack explicit attention-level supervision to retain historical contributions to current queries. To address this, we introduce Attn-Forcing, a novel framework that preserves full-context attention contributions for streaming video generation under bounded memory and computation. We achieve this by introducing attention proxies that continuously approximate the attention responses induced by the historical context. Specifically, we implement the proxies as learnable visual latents and employ a Proxy Update Module to incorporate visual context from historical latents. Moreover, we propose an Attention Matching Loss to explicitly constrain the attention behavior of the learned proxies using full-context attention supervision. Extensive qualitative and quantitative experiments demonstrate that Attn-Forcing consistently improves streaming video generation quality and achieves state-of-the-art performance over existing baseline methods.
Abstract (translated)
URL
https://arxiv.org/abs/2610.30036
PDF
-
RealMe: Immersive Volumetric Video from Minimal Captures
2026-10-21 16:49:14Chenxu Zhou, Zhiyuan Yu, Yudong Jin, Yinji Shentu, Yu Zhang, Jiaming Sun, Xiaowei ZhouAbstract
This paper addresses the problem of creating immersive volumetric videos from sparse and low-resolution capture resources. Conventional dense multi-view reconstruction methods can produce high-quality volumetric videos, but their reliance on costly capture setups makes them impractical in real-world scenarios. Recent approaches use generative models to relax capture requirements, but still suffer from limited resolution and 4D inconsistency. To overcome these constraints, we propose RealMe, a unified framework built around a high-resolution video diffusion model that enhances limited captures into high-resolution (2K), 4D-consistent videos for volumetric video reconstruction. Specifically, RealMe constructs auxiliary data from capture videos and encodes them into latents along with layout-aware positional embeddings. The video diffusion model adapts factorized spatio-temporal attention to the multi-view capture layout, denoises these latents across viewpoints and time, and decodes them into enhanced videos to refine the 4D representation. This adaptation helps the model capture temporal dynamics and multi-view relationships, leading to more 4D-consistent generation. To reduce the data gap between generated videos and real captures, we apply a latent frequency loss that provides frequency-domain supervision in latent space. Extensive experiments on the public DNA-Rendering benchmark and our self-captured RealVV dataset demonstrate that RealMe is able to create immersive volumetric videos and outperforms state-of-the-art methods in visual fidelity and 4D consistency.
Abstract (translated)
URL
https://arxiv.org/abs/2610.30023
PDF
-
HiddenLens: A Multi-Layer LLM Framework for Detecting and Resolving Hidden Assumptions in Software Requirements
2026-10-21 16:35:35Tanu Singh, Aashna Suman, Vinod PatidarAbstract
Natural-language software requirements often omit assumptions about performance, operating conditions, user behaviour, and domain constraints, affecting architecture, testing, and deployment. If undiscovered until implementation, resolving them is far more costly than addressing them early. Although ambiguity and vagueness are widely studied, less attention has been given to assumptions embedded in requirements that appear clear and reasonable. We propose HiddenLens, a novel six-layer BERT-LLM pipeline and, to our knowledge, the first framework engineered to detect, classify, question, and resolve hidden assumptions rather than merely flag ambiguity. Layer 1 uses BERT attention entropy to localise unresolved spans; Layer 2 uses CoT-prompted LLM reasoning to articulate each assumption and its failure risk; Layer 3 uses fine-tuned BERT to classify assumptions into six non-functional requirement categories: performance, security, scalability, usability, reliability, and domain; Layer 4 generates ranked clarification options using FS LLM prompting; Layer 5 enables human validation through confirmation, editing, or rephrasing; and Layer 6 rewrites the requirement using the confirmed resolution, producing an explicit, testable, and traceable requirement. We also introduce the HiddenLens Corpus (HLC), comprising 312 requirements annotated with 847 hidden assumptions across six NFR categories. Inspired by PROMISE and PURE, HLC addresses the lack of annotated data for this task and is, to our knowledge, the first dataset developed specifically for it. HiddenLens outperforms keyword-based and QuARS-Adapted baselines by 71% and 61%, respectively, achieving an end-to-end F1-score of 0.79; its BERT-based NFR classifier achieves an F1-score of 0.847. These results indicate that HiddenLens can help analysts resolve hidden assumptions before they propagate through software development.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29996
PDF
-
Training-free Acceleration for Blind Compressed Image Restoration via Activation-Calibrated Singular Value Decomposition
2026-10-21 16:16:41Suhang Yao, Xin Li, Xijun Wang, Yufei Wang, Yeying Jin, Zhibo ChenAbstract
Transformer-based models achieve strong performance on blind compressed image restoration (CIR) under unknown compression settings, but their densely stacked projection operators make deployment costly. We study training-free acceleration for such models through an empirical, operator-level analysis. Specifically, we profile where computation is concentrated, characterize channel-wise activation distributions under compressed inputs, and examine the singular-value decay of representative projection weights. These measurements identify the recurring projections in the attention, feed-forward, convolutional, and prompt-conditioning pathways as effective low-rank replacement targets that preserve the original inference graph. They further establish truncated SVD (TSVD) as a principled training-free baseline for blind CIR and reveal why a uniform low-rank objective is mismatched to the channel-imbalanced feature distributions of compressed images. Accordingly, we formulate Activation-Calibrated Singular Value Decomposition (ASVD) for this restoration setting, deriving low-rank factors entirely offline from pretrained weights and lightweight activation statistics gathered from only a few compressed calibration images. By weighting the decomposition with channel-wise activation magnitudes, our ASVD assigns larger penalties to errors on more active channels, requiring neither retraining nor online calibration and leaving the restoration pipeline intact. When instantiated on PromptCIR at rank 64, ASVD consistently improves PSNR-B over TSVD across four standard benchmarks under identical parameter and MAC budgets. At this operating point, the factorized model cuts parameters by 59% (34.75M to 14.34M) and computation by 16.1% (258.79 to 217.16 GMACs for a 256 x 256 input), enabling acceleration without retraining.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29959
PDF
-
Acharya 2.0: An AI Tutor for Self-Learning Subjects in Hindi
2026-10-21 15:37:24Shantipriya Parida, Debasish Dhal, Sonal Khosla, Sambit Sekhar, Muhammed Abdur Rahmaan KamaldeenAbstract
Text processing is crucial for low-resource languages as it enables the development of tools and technologies that bridge the digital divide and provide equitable access to information and services. By empowering low-resource languages with robust text processing, we can preserve linguistic diversity, enhance education, improve communi- cation, and foster inclusivity in a rapidly digitizing world. Due to the lack of sufficient annotated datasets and lin- guistic resources, it is challenging to develop accurate and reliable language technologies for these low-resource languages. Large Language Models (LLMs) have attracted sig- nificant attention with their performance and the value they produce. However, with the ever-evolving language, adapting LLMs for a specific task is time-consuming and resource-intensive. Thus, Retrieval-Augmented Genera- tion (RAG) has overcome this limitation by augmenting LLMs with external knowledge, thereby improving their efficiency and extensive task-specific adaptation. Most applications of the RAG technique have been limited to English and has seen limited application in In- dic languages. This paper introduces the RAG-powered AI-Tutor application, Acharya 2.0, designed to produce academic assistance to Hindi-speaking students. The client-server-based web application integrates many fea- tures to support students in accessing and interacting with educational content.
Abstract (translated)
URL
https://arxiv.org/abs/2610.25192
PDF
-
How Do DiTs Denoise? An Aggregator-Predictor View of Attention and MLPs
2026-10-21 15:14:22Gyeonghun Kang, William Xingxu Chen, Xiang ChengAbstract
How do diffusion transformers (DiTs) represent image content and learn rules for reusing it to generalize? We develop an aggregator--predictor view of DiTs as input-adaptive local denoisers, in which context aggregation and patch prediction provide functional roles for interpreting attention and MLPs. We first establish that, for Gaussian Markov random field (GMRF) priors, the global Bayes denoiser can be approximated using only a subset of context patches, with error decaying exponentially in context radius. We further show that patchwise linear denoisers, often invoked to explain diffusion generalization, recover the empirical local denoiser with sufficiently informative aggregation. Within this view, AdaLN-modulated attention naturally parameterizes an input- and noise-dependent aggregator, while the MLP admits a learned key--value representation for tokenwise patch prediction. Our experiments support this view. DiTs trained on real images exhibit noise-dependent denoising contexts resembling those in the analytic GMRF setting, while attention mass exhibits a qualitatively corresponding broadening pattern. In a controlled CelebA64 study, shared-memory patch predictors used in place of DiT MLPs achieve competitive denoising and generative performance. Finally, cross-dataset component transplantation reveals compatibility patterns consistent with the proposed roles of QK in spatial weighting, VO in feature transport, and MLPs in patch prediction. Together, these results provide a component-level account of DiT denoising and motivate architecture-specific fine-tuning and controlled intervention.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29854
PDF
-
Spatially Gated Demographic Steering for Debiasing Multi-Person Text-to-Image Generation
2026-10-21 12:32:16Venkatesh Thirugnana Sambandham, Torsten SchonAbstract
Text-to-Image models reproduce societal stereotypes, and existing mitigation applies one demographic correction to the whole scene. In multi-person scene generation this creates a trade-off: global steering improves aggregate balance only by homogenizing the people within an image. Steering-based methods like Embedding Arithmetic drive the intra-image diversity below the unsteered baseline. Text- and latent-projection approaches shift the aggregate skew with only marginal gains in intra-image diversity. We present Spatially Gated Steering (SGS), which steers the demographics of each person independently by extracting per-person masks from the Diffusion Transformer's (DiT) internal attention maps at a critical denoising timestep and applying a distinct demographic vector within each mask for the remaining steps. SGS needs no fine-tuning, inversion or layout specifications, and uses a lightweight segmentor over frozen DiT attention features to localize subjects at early timesteps. On skewed occupational prompts, SGS reaches high intra-image racial and gender diversity while keeping aggregate balance competitive and image quality near baseline, a result we validate on two architecturally distinct diffusion transformers, FLUX.1-Dev and SD3.5-Large. Codebase and model weights are available at this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2610.29651
PDF
-
MS2CAST: A Video Anomaly Detection Framework via Multi-Scale Spatial Cross-Attention and Efficient Spatiotemporal Networks
2026-10-21 12:29:24Ali Safaei, Mohammad KazemiAbstract
Video anomaly detection (VAD) is a challenging task for surveillance systems due to the scarcity and diversity of anomalous events. While semi-supervised VAD approaches have advanced the field, many rely on computationally intensive optical flow networks to capture motion dynamics. Convolutional neural networks (CNNs) are also widely used for spatial encoding, but their effective receptive field is highly local. To address these limitations, we introduce MS2CAST, an end-to-end semi-supervised predictive VAD framework. For spatial encoding, we employ a multi-scale cross-attention encoder that fuses the hierarchical feature representations from a vision transformer (MViTv2-S) and a CNN (ConvNeXt-T) to exploit their distinct receptive fields. To model spatiotemporal dynamics without optical flow estimation, we integrate two interchangeable spatiotemporal modules in the latent space. One module (SwinLSTM) incorporates a Swin Transformer, while the other (VMRNN) integrates Vision Mamba blocks. Our objective function combines L1 loss, Gradient Difference Loss (GDL), and Multi-Scale Structural Similarity Index Measure (MS-SSIM) to ensure pixel-wise accuracy, sharpness, and structural integrity. Evaluations across the UCSD Ped2, CUHK Avenue, and ShanghaiTech datasets demonstrate frame-level Receiver Operating Characteristic (ROC) Area Under the Curve (AUC) scores that are competitive with top-tier techniques while maintaining a peak GPU memory usage of under 1 GB.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29649
PDF
-
Watch Your Step: Anticipated Perception for Humanoid Locomotion
2026-10-21 11:52:29Matthias Heyrman, Chenhao Li, Ioannis Dadiotis, Marco HutterAbstract
Perceptive locomotion policies typically observe a dense robot-centric elevation map, although only the terrain beneath and around the next footsteps informs the control decision. Learned encoders compress the whole visual input, leaving the policy to work out which inputs matter while learning the baseline locomotion task. Learning both at once slows training and, without computationally expensive attention mechanisms, makes generalization to new terrain harder. We instead remove redundant inputs from the elevation map before the policy sees it. A lightweight foothold prediction model (FPM) predicts future footholds from proprioception and the velocity command alone, and we sample the elevation map only in small patches around these predicted footholds. We train the prediction model alongside the locomotion policy, periodically retraining it on new policy rollouts as the robot's gait matures. This allows the perception to anticipate where the policy will step, while the policy learns to leverage terrain information around those steps. With the proposed lightweight preprocessing, the policy reaches a given terrain difficulty in roughly a third of the training iterations required with the complete map. It outperforms policies that observe the complete map where footholds are sparse, including in narrow walled corridors, and on unseen terrain, and it matches an attention-based encoder trained at several times the cost. The policy transfers zero-shot to a Unitree G1 humanoid traversing various obstacles, stairs, gaps, and step-downs beyond the training distribution in both indoor and outdoor environments.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29612
PDF
-
No Frames, Just Frequencies: Bandwidth-Efficient Deployment-Time Adaptation for Smart Glasses
2026-10-21 11:46:59Francesco Corti, Moritz Becker, Olga SaukhAbstract
Deep neural networks deployed on smart glasses must adapt to changes in the distribution of incoming data while adhering to strict privacy and device constraints. We present No Frames, Just Frequencies, a bandwidth-efficient source-assisted adaptation framework that characterizes distribution shifts using compact frequency descriptors computed on-device. Rather than transmitting camera frames, the glasses compute and send frequency descriptors to a host device that contains a replica of the model. The host device maps the received frequency descriptor to a distribution shift operator that synthesizes out-of-distribution data and adapts a model replica using a pre-defined adaptation protocol. We evaluate the framework on mixtures of ImageNet-C corruptions using convolution-based (ConvNeXtV2-Femto), self-attention-based (DeiT-Tiny), and hybrid (TinyViT-5M) vision models, on person classification (Visual Wake Words), and on object detection (MS-COCO). Our results reveal that adaptation driven solely by the transmitted descriptors recovers $93-95\%$ of the supervised oracle adaptation gain on all three architectures, $82\%$ on Visual Wake Words and $94$--$98\%$ on MS-COCO. Notably, adaptation quality degrades gracefully as the descriptor resolution decreases from $224 $ to $28$ pixels, consistently across architectures, enabling bandwidth reductions from $213\times$ to $784\times$ across the frequency descriptors.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29607
PDF
-
A Diagnostic Study of Faithfulness in Diffusion Language Models for Table-to-Text Generation
2026-10-21 11:42:19Zaynab Merimi, Noussayba El Marrakchi, Arthur SatoufAbstract
Diffusion language models generate by iteratively denoising a masked sequence under bidirectional attention, without the strict left-to-right commitment of autoregressive decoding. Whether this difference affects how reliably they remain grounded in structured inputs is largely unexplored. We study this question in table-to-text generation using ToTTo, where highlighted source cells make factual grounding directly testable. We compare the masked diffusion models MDLM and LLaDA with GPT-2 and GPT-Neo under an identical retrieval-based few-shot protocol, together with a QLoRA-adapted LLaDA, and we score reference similarity and source grounding as separate axes. The two pull apart. Within LLaDA, additional demonstrations raise BLEU from 11.3 to 22.0 while NLI entailment against the source table falls from 0.78 to 0.52; QLoRA reaches the best BLEU of the study, 24.15, and grounds no better for it. Manual annotation shows the failure modes differ in kind: the autoregressive baselines tend to wander off the table, while LLaDA-based models more often stay on topic and swap out individual entities. Reference-based metrics alone are insufficient to assess faithfulness in table-to-text generation.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29602
PDF
-
PACE: Periodic and Channel-Aligned Encoding of Four-Bit Recurrent States
2026-10-21 10:31:52Patrick Haller, Filipe Laitenberger, Alan AkbikAbstract
Linear-attention language models keep their context in a recurrent state of fixed size, where an inference system stores one copy of this state for every sequence it decodes at the same time. Storing the state at lower precision therefore lets more sequences share an accelerator. Existing state quantizers round the state after every generated token and report that four bits lose accuracy unless the bit width is raised or the model is fine-tuned.\par We introduce PACE, a post-hoc four-bit state quantizer. PACE rounds the state only once every few generated tokens, as chunked kernels already do once per chunk during prefill, and keeps the tokens generated in between in a small buffer. Rounding less often reduces the error that accumulates in the state. Additionally, we find that the large entries of the gated-delta state gather in columns, because some channels of the update written at each token remain large across many tokens. PACE therefore shares each quantization scale within one column. It also keeps the heads that retain rounding error longest at eight bits. On four hybrid models from 4B to 35B parameters, PACE with mixed four- and eight-bit heads is competitive with eight-bit DAMP on the three larger models at about two thirds of its memory, PACE with four-bit heads throughout scores higher than Q-Mamba at equal memory, and rounding every few tokens preserves the decoding speedup of the buffer.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29545
PDF
-
BlazeVGGT: 13x Faster VGGT at 2,000 Frames via Camera Token without Training
2026-10-21 09:11:19Shujun Yao, Junhui Lin, Congying Han, Tiande Guo, Ziwen Liu, Bonan LiAbstract
Feed-forward visual geometry models, exemplified by VGGT, have demonstrated strong multi-view reconstruction capabilities through dense global attention. However, the resulting all-to-all cross-view computation scales quadratically with the number of input views, making long-sequence inference particularly expensive due to substantial frame-level redundancy among highly overlapping observations. We uncover a topology-first, geometry-later evolution in VGGT camera tokens: shallow-layer tokens reliably capture local view neighborhoods, whereas deeper-layer tokens progressively encode globally organized inter-view geometry. Motivated by this observation, we introduce BlazeVGGT, a training-free framework for structure-aware frame-level key--value sparsification. At each global-attention block, BlazeVGGT constructs a view-level similarity kernel from camera-token relations and employs greedy determinantal point process inference to select a compact set of mutually complementary views. The selected views constitute the key--value context, while all input frames are retained as queries. This asymmetric attention design preserves dense per-frame predictions while eliminating redundant cross-view computation, allowing our method to be directly integrated into pretrained VGGT as a plug-and-play module without parameter updates. Its advantages become increasingly pronounced with sequence length: at 2,000 frames, BlazeVGGT achieves approximately 13$\times$ faster inference than VGGT while reducing the Chamfer Distance from 0.41 to 0.32. Extensive experiments demonstrate a favorable accuracy--efficiency trade-off in point-cloud reconstruction and camera-pose estimation, compatibility with existing VGGT accelerators such as FastVGGT and LiteVGGT, and transferability to VGGT-based models including StreamVGGT and PAGE-4D. Codes are publicly available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29454
PDF
-
Sparsity Should Be ALLOTted by Output: Training-Free Allocation of Block-Sparse Attention for Video Diffusion Transformers
2026-10-21 09:01:48Tianze Xia, Xiaoling Zhou, Ruya Lan, Yiqiao Liao, Chengru Song, Shiming XiangAbstract
Full 3D attention is costly in video Diffusion Transformers. Block-sparse attention reduces this cost by computing only selected attention blocks. However, existing methods either overlook the potential to optimize sparsity allocation or rely on heuristic rules or local metrics to guide it. In this work, we observe that sparsifying attention in different layers and at different denoising steps affects final-video fidelity to substantially different degrees. Yet the identities of the most sensitive layer-step pairs remain nearly unchanged across inputs. Building on these findings, we propose Allocating Layer-step Latency by Output Tests (ALLOT), a training-free sparsity allocation method. Treating each layer-step pair as a cell, ALLOT estimates each cell's contribution to final-video error through (i) cell probing and (ii) trajectory calibration, then jointly optimizes cell densities under a measured-latency budget. It preserves existing selection strategies and kernels, replacing only the allocation component. At equal or shorter wall-clock time, ALLOT outperforms every compared allocation strategy with XAttention, with an 11.8 dB PSNR gain on HunyuanVideo. Extensive experiments across four additional published pipelines on Wan2.1, Wan2.2, and HunyuanVideo show that ALLOT outperforms or ties the default in all nine settings.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29441
PDF
-
CCG Diffusion: Combinatory Categorial Grammar-Driven Compositional and Syntactically Disambiguated Text-to-Image Generation
2026-10-21 08:38:18Ryosuke Yamaki, Daichi Mochihashi, Nobutaka Shimada, Seitaro Shinagawa, Tadahiro TaniguchiAbstract
Text-to-image generation models do not explicitly receive the syntactic structure of the prompt, i.e., which word modifies which and where each phrase attaches. Consequently, they often struggle to bind attributes to the correct objects, to depict spatial relations and counts correctly, and to render syntactically ambiguous prompts in the intended interpretation. Existing syntax-based control methods mainly address attribute binding and have limited ability to control spatial relations and counts or to specify the interpretation of syntactically ambiguous prompts. They rely on phrase-structure or dependency grammars and require additional rules or constraints on the tree structure to extract modification relations between words or to specify a syntactic interpretation. We propose CCG Diffusion, a training-free method that controls image generation using Combinatory Categorial Grammar (CCG). In CCG, the category of each word carries rich syntactic information: it specifies what the word combines with and what phrase it forms. Modification relations between words can therefore be read directly from a CCG parse. Moreover, prepositional phrase (PP) attachment ambiguity, a representative syntactic ambiguity, can be resolved simply by fixing the preposition's category. CCG Diffusion applies the same two CCG-based mechanisms, prompt rewriting and attention guidance, both to controlling attribute binding, spatial relations, and counts and to resolving PP ambiguity. On two architecturally different backbones, Stable Diffusion and Sana, CCG Diffusion achieves the highest scores among the compared methods in attribute binding and 2D spatial relations and significantly outperforms all of them in PP ambiguity resolution.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29415
PDF
-
LiveAct: Towards Real-Time Multi-Agent World Models with Heterogeneous Action Interfaces
2026-10-21 08:34:03Xinyu Wang, Huafeng Shi, Yue Ma, Zijie Meng, Teng Xi, Yuzhuo Chen, Yucheng Wang, Ruicheng Zhang, Guoxin Zhang, Pengfei Wan, Xiu Li, Fangneng ZhanAbstract
Interactive video world models predict how an environment evolves under user actions, but most existing systems assume one controlled agent and one task-specific action interface. We study multi-agent video world modeling, where one model generates a joint visual future for multiple controlled agents while interpreting heterogeneous controls ranging from binary game buttons to continuous robot commands. This task presents two complementary challenges. First, every agent-indexed action stream must control its intended visual entity independently. Second, the model must adapt the meaning of each control variable across different tasks, where the same tensor dimension may denote an attack, steering command, or robot command. We introduce LiveAct, centered on a Heterogeneous Action Interface (HAI) whose two components mirror this decomposition. To support independent control, Agent-Aware Spatial Modulation uses language-driven spatial gates and agent-specific attention to bind each action stream to its intended visual agent. To accommodate heterogeneous controls, Schema-Guided Action Adaptation describes each control variable in language and modulates its semantic token with the synchronized numerical value. We further transfer the complete interface to a blockwise causal generator through teacher forcing and rollout-aware distribution matching, enabling efficient streaming generation while retaining agent-specific control and action semantics. We instantiate the formulation on six multi-agent sources spanning games and multi-robot manipulation, and evaluate visual quality, action following, agent attribution, and latency. Across games and robot manipulation, LiveAct reaches 0.898 and 0.942 agent-attribution accuracy without agent locations or identity crops, and the causal generator attains 17.70 FPS at $512\times 448$.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29408
PDF
-
Heard But Not Said: Prompting Understates Temporal Information of Musical Moments in Frozen Large Audio-Language Models
Abstract
Large audio-language models (LALMs) understand music, yet struggle to locate a described moment in time. We ask whether this reflects missing temporal information or a bottleneck in how the time span is extracted. Using a paired-query probe built from Slakh2100 stem activations, we attach a state query (e.g., ``only piano and bass are playing'') and a change query (e.g., ``the guitar drops out, leaving piano and bass'') to the same span and vary how much preceding audio the model is shown. Across frozen LALMs, three ways of extracting the span ask different questions: full-track prompting asking whether the model can say when; windowed yes/no scoring asking whether it verifies the moment when handed the segment, and whether it needs what came before; attention asking whether it can find the moment on its own. State queries score above baseline without context, while change queries improve with preceding context in most short-segment models; 30s padded encoders show no comparable recovery. Prompting and attention expose almost none of this. Prompting failures thus understate what LALMs can verify, though not what they can locate, by an amount set by query type and encoder.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29256
PDF
-
Cache Locally, Recompute Selectively for Streaming Video Understanding
2026-10-21 05:39:35Junhyeok Lee, Sangyoon Lee, Jaeho LeeAbstract
Streaming video understanding (SVU) requires video-LLMs to process incoming video continuously while answering user queries with low latency. Recent methods reduce query-time cost by prefilling frames into KV caches during idle time and reusing the cached KVs at query time. To bound per-frame computation over unbounded video streams, these methods restrict the context available during prefill. This creates a fundamental context mismatch: the restricted context may omit information needed to properly contextualize video tokens, yet, still contain information irrelevant to the eventual query, which is unknown during prefill. We propose FrameBlend, a simple yet effective plug-and-play method to address this mismatch through selective KV recomputation, without requiring any additional training. At query time, FrameBlend re-prefills selected video tokens using query-relevant context and replaces their cached KVs with recomputed ones. Crucially, this recomputation enables shorter attention windows during idle-time prefill, substantially reducing prefill cost while improving accuracy. Experiments across three SVU frameworks and five streaming and offline benchmarks demonstrate accuracy gains of up to 3.9 pp. On RVS-Ego, FrameBlend reduces per-frame prefill FLOPs by over 20% and total processing time by 14-59% for retrieval-based frameworks, including the cost of query-time recomputation. These results demonstrate that a simple change in how contextualization is distributed between idle time and query time can improve the accuracy-efficiency trade-off for SVU.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29229
PDF
-
NavBind: Diagnosing Instruction-Action Binding at Route Forks in Navigation VLAs
2026-10-21 04:50:26Josh Qixuan Sun, Xiaoying Xing, Yurun Chen, Huaiyuan Weng, Wentao Zhu, Mark CrowleyAbstract
Vision-language-action (VLA) navigation models translate visual observations and instructions into actions. Navigation success assesses whether an agent reaches the goal. However, reaching the goal successfully does not establish whether the model genuinely follows the instructions or relies on visual cues or navigation history. We introduce NavBind to diagnose instruction-action binding: whether changing the instruction influences the probabilistic distribution of the actions, with the navigation context fixed. We construct 3,252 route-fork contexts by recombining route-aligned instruction segments from R2R-CE and RxR-CE. Each context pairs full instructions that share the description of the route already traveled but specify different branches. Both are evaluated under the same agent state and navigation history. We compare generated local action sequences, the likelihoods of fixed reference action sequences under both instructions, and changes in visual attention and attribution. Across the evaluated models, higher reported navigation success does not imply that the model more faithfully follows the instructions.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29203
PDF
