Video_Caption
-
Benchmarking Visual State Tracking in Multimodal Video Understanding
2026-06-02 17:12:05Sihyun Yu, Nanye Ma, Pinzhi Huang, Hyunseok Lee, Shusheng Yang, June Suk Choi, Ellis Brown, Oscar Michel, Boyang Zheng, Jinwoo Shin, Saining XieAbstract
Understanding a video requires more than recognizing isolated moments, as humans continuously track entities, states, and events over time. This capacity for visual state tracking is fundamental to video understanding, yet remains underexplored in current evaluations of Multimodal Large Language Models (MLLMs). We introduce Visual STAte Tracking benchmark (VSTAT), a video-based benchmark designed to diagnose visual state tracking in MLLMs. VSTAT consists of 834 clips drawn from both synthetic and real-world videos, paired with 1,500 questions that cannot be answered from any single frame or short segment, requiring continuous perception and integration of events across the entire video stream. Despite their strong performance on existing video benchmarks, we find that state-of-the-art MLLMs perform far below humans and only modestly above answer-prior baselines. To analyze this gap, we compare MLLMs' thinking traces with the underlying video stream to understand why and when MLLMs fail on VSTAT. We find that MLLMs reason and track correctly in text, but fail at visually perceiving the events they need to track. Finally, our preliminary evaluation suggests that recent agentic approaches, including MLLM-based video agents and coding agents, do not readily resolve these failures, still falling short on VSTAT.
Abstract (translated)
URL
https://arxiv.org/abs/2606.03920
PDF
-
Where Do We Need Temporal Context in Low-Resource Video Task Adaptation?
Abstract
Parameter-efficient fine-tuning (PEFT) and probing enable adaptation of foundation models using only a small number of trainable parameters, making it attractive for video understanding where annotation and computation are expensive. However, video PEFT has focused on adapting image-pretrained models, while standard PEFT methods can also be applied to video representations. These settings are rarely compared and both confine temporal reasoning to a single component of the model, leaving open how temporal context should be distributed across backbone, PEFT and probe. In this work we provide a systematic study of model adaptation strategies for video understanding. We evaluate methods across appearance-focused, motion-focused and spatially dense settings, with a particular focus on scenarios with limited data where parameter-efficiency is most beneficial. Our results provide new insights into PEFT and probing across settings and demonstrate the importance of temporal context allocation for effective video adaptation
Abstract (translated)
URL
https://arxiv.org/abs/2606.03837
PDF
-
VidMsg: A Benchmark for Implicit Message Inference in Short Videos
2026-06-02 13:31:57Issar Tzachor, Michael Green, Rami Ben-AriAbstract
Understanding short online videos involves more than identifying visible objects and actions; video makers often include an underlying message or purpose in the clip. We introduce VidMsg, a benchmark for evaluating implicit message understanding in short, internet-native video clips. VidMsg contains 400 YouTube-derived clips across 9 practical topic areas and 52 fine-grained target messages, covering domains such as career and finance, education, health and well-being, culture, safety, sustainability, and lifestyle. VidMsg is constructed through a message-first pipeline: an LLM first translates target messages into indirect search scenarios, which are used to retrieve candidate clips. Human annotators then retain clips that convey the intended message without being overly explicit. VidMsg is designed primarily for bidirectional message-clip retrieval for scalable applications such as video search and recommendation, where systems must capture holistic video understanding. In addition to retrieval, VidMsg includes a diagnostic multiple-choice QA benchmark, where models select the intended message of a clip from semantically related alternatives. Experiments with contemporary video-language and retrieval models show that strong models often fail on VidMsg, because the task requires pragmatic inference, integration of contextual cues, and discrimination among semantically close messages. We also introduce VidVec-Msg, a baseline method that improves message-oriented retrieval while leaving substantial headroom for future work.
Abstract (translated)
URL
https://arxiv.org/abs/2606.03635
PDF
-
Moment-Video: Diagnosing Temporal Fidelity of Video MLLMs on Momentary Visual Events
2026-06-01 17:32:20Xiaolin Liu, Yilun Zhu, Xiangyu Zhao, Xuehui Wang, Yan Li, Xin Li, Haoyu Cao, Xing Sun, Shaofeng Zhang, Xu Yang, Zhihang Zhong, Xue YangAbstract
Video multimodal large language models (MLLMs) have made rapid progress on general and long-form video understanding, yet their ability to preserve brief answer-critical visual evidence remains underexplored. Many practical questions are determined by momentary visual events: localized actions or state transitions that may last only a few frames. Such evidence can be skipped by sparse frame sampling, suppressed by visual-token compression, or diluted by coarse temporal aggregation, causing failures that language-side reasoning cannot reliably recover. We introduce Moment-Video, a benchmark for diagnosing the temporal fidelity of video MLLMs through momentary visual event understanding. Each question is grounded in a localized, visually observable, and sampling-sensitive event, requiring models to notice, count, describe, or reason about transient evidence rather than rely on persistent objects, global scene context, or language priors. Moment-Video contains 1,000 human-verified video-QA pairs across 7 domains and 25 fine-grained subcategories, covering four task types: Temporal Occurrence, Temporal Counting, Action Description, and Temporal Reasoning. We evaluate 33 proprietary and open-source MLLMs on Moment-Video. The best-performing model, Seed-2.0-Pro, achieves only 39.6% overall accuracy, while most open-source models remain below 25%, revealing a substantial gap in momentary visual event understanding. Diagnostic analyses show that denser frame sampling improves some models but does not eliminate the bottleneck, and longer videos introduce stronger temporal-localization challenges. These findings suggest that current video MLLMs still lack temporally faithful representations for capturing, preserving, and using brief but decisive visual evidence.
Abstract (translated)
URL
https://arxiv.org/abs/2606.02522
PDF
-
Multi-modal Video Representation Alignment for Robust Self-supervised Driver Distraction Detection
2026-06-01 15:01:17David J. Lerch, Livien Majer, Zeyun Zhong, Manuel Martin, Frederik Diederichs, Rainer StiefelhagenAbstract
Robust self-supervised learning of multi-modal video representations is critical for real-world applications such as driver distraction detection, where multiple sensors provide complementary but noisy signals. Conventional contrastive objectives, such as InfoNCE, assume all negatives are equally informative and all positives are reliable. However, this assumption is frequently violated in multi-modal data due to viewpoint changes, occlusions, or semantic overlap across modalities. In this work, we propose a novel framework for multi-modal global alignment that addresses these challenges by jointly modeling faulty negatives and unreliable or faulty positives. We introduce soft targets derived from cycle-consistency scores to relax the hard-negative assumption, and a weighting mechanism based on similarity distributions to mitigate the impact of noisy or faulty positives. Our approach extends traditional pairwise alignment to a principled global multi-modal setting, aggregating alignment information across all modality pairs. We evaluate our method on the Drive&Act dataset, demonstrating that it consistently outperforms both pairwise and existing global alignment baselines across RGB, IR, Depth, and Skeleton modalities. Cross-view ablation studies further show strong generalization to unseen camera perspectives, highlighting the robustness of our representations. Overall, our framework provides a scalable and effective solution for self-supervised global multi-modal representation learning, enabling reliable driver distraction detection and pioneering in real-world multi-modal video understanding. Our code will be published on GitHub.
Abstract (translated)
URL
https://arxiv.org/abs/2606.02352
PDF
-
InfoMerge: Information-aware Token Compression for Efficient Video Large Language Models
2026-06-01 12:24:24Xinxin Liu, Shiwei Gan, Xiao Liu, Yafeng Yin, Lei Xie, Sanglu LuAbstract
Video Large Language Models (Video-LLMs) achieve strong performance in video understanding, but their excessive visual tokens bring substantial computational overhead. Existing training-free compression methods improve inference efficiency by reducing visual tokens, yet they often rely on local adjacent-frame similarity for temporal redundancy estimation or allocate token budgets mainly according to segment length. Such designs are sensitive to frame-level noise and fail to capture the non-uniform information distribution of real-world videos. To address these challenges, we propose InfoMerge, a training-free visual token compression method that improves token utilization through robust redundancy estimation and content-aware budget allocation. Specifically, we propose the Temporal Fingerprint Difference: a segment-level second-order temporal redundancy estimation strategy, which models the temporal similarity structure of tokens at the same spatial positions within each segment. We further introduce Content-Aware Budget Allocation (CABA), which dynamically allocates segment-level token budgets based on segment uniqueness and spectral-entropy-based representational richness. By reducing repeated preservation of redundant static regions and allocating more tokens to informative segments, InfoMerge makes better use of the limited token budget while maintaining strong performance. Extensive experiments show that InfoMerge achieves strong efficiency--accuracy trade-offs across multiple benchmarks and backbones, with more pronounced advantages under aggressive compression. On LLaVA-OneVision-7B, InfoMerge retains 98.8\% of the original average performance while reducing 85\% of visual tokens and achieving a 4.24-fold speedup in the prefill stage.
Abstract (translated)
URL
https://arxiv.org/abs/2606.02161
PDF
-
Understanding-Enhanced Model Collaboration for Long-Tailed Egocentric Mistake Detection
2026-06-01 11:50:16Boyu Han, Qianqian Xu, Shilong Bao, Zhiyong Yang, Ruochen Cui, Qingming HuangAbstract
In this report, we address the problem of determining whether a user performs an action incorrectly from egocentric video data. To this end, we propose an Understanding-Enhanced Model Collaboration Method (UE-MCM) that combines efficient coarse-grained video understanding with accurate fine-grained action reasoning. Specifically, UE-MCM contains a small model branch and a large model branch. The large model branch focuses on whether the fine-grained action itself is executed incorrectly, while the small model branch jointly takes the coarse-grained video and fine-grained segment as input to identify actions that may be locally correct but inconsistent with the overall workflow. The small model branch is built on a CLIP4CLIP video encoder initialized from a CLIP model enhanced by Diffusion Contrastive Reconstruction, and the large model branch uses the Qwen3-VL Embedding model to extract high-capacity representations from fine-grained action segments. The small-branch prediction and the large-branch prediction are then adaptively fused by a lightweight collaboration gate. To handle the long-tailed distribution of mistake instances, we optimize the classifiers with complementary objectives, including reweighted cross-entropy, AUC-oriented learning, and label-aware adjustment. The resulting system balances speed and accuracy, making it effective for detecting subtle, rare, and ambiguous mistakes in egocentric instructional videos.
Abstract (translated)
URL
https://arxiv.org/abs/2606.02120
PDF
-
MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention
2026-06-01 09:07:15Pengyu Wang, Chenkun Tan, Shaojun Zhou, Wei Huang, Qirui Zhou, Zhan Huang, Zhen Ye, Jijun Cheng, Xiaomeng Qian, Yanxin Chen, Xingyang He, Huazheng Zeng, Chenghao Wang, Pengfei Wang, Hongkai Wang, Shanqing Gao, Yixian Tian, Chenghao Liu, Xinghao Wang, Botian Jiang, Xipeng QiuAbstract
Video understanding is shifting from the offline paradigm -- taking a fully recorded video as input and producing a single answer after it ends -- toward real-time interaction, in which the model perceives new frames while still replying, revises its answer as new evidence appears, and remains silent when there is nothing to say. We present MOSS-Video-Preview to validate this paradigm. Our central claim is that perception must not be blocked by generation; its natural realization is a two-channel architecture. We argue that a cross-attention backbone is better suited to real-time vision-language fusion than the prevailing decoder-only design: visual features enter through a side channel rather than joining the autoregressive sequence, so perception and generation run on separate, non-blocking pathways -- reducing the frequency of visual processing and exposing a clean channel-wise interface for independent compression. We complement this with a data synthesis pipeline that converts dense captions into real-time understanding QA whose answers are revised to match what the model has perceived so far, and we specialize an offline model on these data to elicit real-time behavior. Our model trails the strong Qwen2.5-VL-7B baseline overall -- a gap we attribute primarily to data and scale rather than the architecture -- yet attains competitive offline video and multimodal understanding, remains robust on the spatial and fine-grained temporal reasoning central to real-time use, and acquires behaviors that offline models lack: continuous perception, answer revision, and timely silence. On a single H200 with 256 frames per video, it achieves about a 5x speedup in time to first token and 2.7x higher decoding throughput, with negligible degradation in offline ability. Our study of paradigm, architecture, and data outlines a viable path toward real-time video understanding.
Abstract (translated)
URL
https://arxiv.org/abs/2606.07639
PDF
-
3rd Place at CVPR 2026 CASTLE Challenge: Agentic Multi-View Long-Context Video Understanding via Hierarchical Knowledge Graph Retrieval
2026-06-01 09:01:32Raghad Albusayes, Munirah AlyahyaAbstract
This paper presents our winning methodology for the CASTLE 2026 Challenge at the CVPR 2026 EgoVis Workshop, where our team secured third place globally. The challenge tasks participants with answering highly complex visual, spatiotemporal, and verbal questions, including visual counting, action localization, multi-view tracking and speaker temporal reasoning, within massive, multimodal video streams. The underlying dataset consists of over 600 hours synchronized footage captured by 15 ego and exo camera sources. To tackle the extreme scale and long-context demands of this environment, we introduce a training-free agentic framework optimized for long-form video understanding. Our framework introduces two core architectural components: i) a Video Knowledge Graph that maps static and dynamic entities, their temporal relationships, and intersecting events to enable multi-hop relational reasoning, and ii) an adaptive agentic workflow that resolves complex queries through a hierarchical retrieval and indexing. Empirical results demonstrate that our framework achieves high zero-shot reasoning accuracy on long-context multi-view streams. Our code will be released at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2606.01933
PDF
-
EvoCut: Multi-Layer Evolution-Aware Visual Token Compression for Efficient Large Vision-Language Models
2026-06-01 06:21:28Hongyu Lu, Feng Zhang, Wenwei Jin, Huanling Hu, Pengfei Zhang, Yao Hu, Jiawei Li, Shikai JiangAbstract
Large vision-language models (LVLMs) achieve strong performance on image and video understanding tasks, but their inference efficiency is constrained by the large number of visual tokens produced by vision encoders. Most existing visual token compression methods estimate token importance from attention scores or representation properties at specific layers, overlooking how visual tokens evolve across the vision encoder. Such layer-specific criteria may provide incomplete importance estimates and limit performance preservation after compression. To address this issue, we analyze layer-wise visual token evolution directions and observe that tokens form multiple group evolution directions across vision-encoder layers. Our analysis further shows that informative tokens tend to exhibit persistent deviations from common group evolution directions. Based on this observation, we propose EvoCut, a training-free and attention-free visual token compression method that estimates token importance from multi-layer evolution deviation. Experimental results show that EvoCut can retain only 11.1\% of the visual tokens on LLaVA-1.5-7B while preserving 94.4\% of the average performance, demonstrating its effectiveness in balancing efficiency and accuracy.
Abstract (translated)
URL
https://arxiv.org/abs/2606.01756
PDF
-
V-LynX: Token Interface Alignment for Video+X LLMs
2026-05-30 03:54:44Jungin Park, Jiyoung Lee, Kwanghoon SohnAbstract
This study introduces an intriguing phenomenon in Video LLMs: rather than merely translating frames into textual embeddings, Video LLMs establish a continuous manifold, token interface, allowing visual tokens to operate as standalone entities within the architecture. Exploiting this discovery, we propose V-LynX, a scalable framework that integrates novel modalities into Video LLMs by repurposing the internalized interface. Departing from conventional paradigms that necessitate heavy modality-specific encoders or paired supervision, V-LynX employs a lightweight auxiliary pathway in parallel with the frozen vision encoder. Our method integrates new sensory inputs with intrinsic video priors by aligning both attention responses and statistical distributions using unpaired unimodal data sets. This ensures manifold compatibility while preserving the integrity of the Video LLMs. Extensive benchmarks demonstrate that V-LynX achieves SOTA and efficiency across audio-visual QA, 3D reasoning, high-frame-rate, and multi-view video understanding. The code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2606.00508
PDF
-
Linear Scaling Video VLMs for Long Video Understanding
2026-05-29 17:59:02Cristobal Eyzaguirre, Jiajun Wu, Juan Carlos NieblesAbstract
Video vision-language models (VLMs) are increasingly used in long-horizon and streaming settings, yet most video encoders still rely on spatiotemporal self-attention, causing compute and latency to grow quadratically with the number of frames. Existing efficiency methods improve scalability but often lose accuracy relative to full self-attention, for example through aggressive frame/token dropping or coarse attention approximations. We introduce StateKV, an inference-time method that adapts pretrained long-video VLMs to linear-time video prefill by carrying cross-frame context in a fixed-capacity, importance-based recurrent state, paired with a second full per-frame cache used for decoding. Across three long-video benchmarks and seven models spanning three families and multiple scales, StateKV remains close to full self-attention and consistently outperforms dominant sliding-window / recency-based streaming approximations, without fine-tuning or architectural changes. StateKV also reduces video-prefill cost measured FLOPs, enabling stronger accuracy at a fixed compute budget by running larger models. These results suggest a practical step toward scalable long-video understanding.
Abstract (translated)
URL
https://arxiv.org/abs/2605.31598
PDF
-
PEEK: Picking Essential frames via Efficient Knowledge distillation
2026-05-29 09:01:56Killian Steunou, Anas Filali Razzouki, Khalil Guetari, Moun\^im A. El-Yacoubi, Yannis TevissenAbstract
Video-language models can process only a limited number of frames, making frame selection a key bottleneck for efficient video captioning. Most captioning pipelines still rely on uniform sampling, which is computationally cheap but agnostic to visual content. Adaptive frame sampling has recently emerged as a promising approach for selecting the most informative frames from a video; however, existing methods remain computationally expensive. We introduce PEEK, an efficient dynamic frame sampling method that distills caption-conditioned frame relevance rankings from a stronger teacher model into a lightweight temporal model that operates only on visual content. We find that, overall, on ActivityNet Captions and MSR-VTT, our method outperforms state-of-the-art methods across all evaluated downstream vision language models, especially when only one or two frames are selected for captioning, obtaining the best CIDEr for most frame budgets. On ActivityNet Captions, PEEK is particularly strong, winning 14 out of 16 configurations. Zero-shot evaluation on MSR-VTT shows that our model transfers best at low frame budgets, while results at four and eight frames are more mixed as temporal coverage and visual diversity become increasingly competitive. Compared with recent adaptive baselines, PEEK is both more accurate in the low-budget regime and more efficient: it adds only $5.2\%$ to the captioning time, compared with $65.4\%$ for CSTA and $211.9\%$ for MaxInfo. We release our code and pre-trained checkpoint at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2605.31029
PDF
-
EarlyTom: Early Token Compression Completes Fast Video Understanding
2026-05-28 14:36:44Hesong Wang, Xin Jin, Lu Lu, Chenhaowen Li, Jian Chen, Qiang Liu, Huan WangAbstract
Video large language models (Video-LLMs) have demonstrated strong capabilities in video understanding tasks. However, their practical deployment is still hindered by the inefficiency introduced by processing massive amounts of visual tokens. Although recent approaches achieve extremely low token retention ratios while maintaining accuracy comparable to full-token baselines, most of them perform compression only at the late stage of prefilling, leaving the efficiency of the vision encoder unoptimized. In this paper, we first show that vision encoding contributes a large portion to the time-to-first-token (TTFT). Therefore, instead of compressing visual tokens only after the vision encoder, performing compression inside the encoder still leaves substantial room for exploration. Based on this insight, we propose EarlyTom, a training-free token compression framework that performs early-stage visual token compression inside the vision encoder, enabling significantly better TTFT reduction and higher throughput. In addition, we introduce a decoupled spatial token selection strategy that improves the overall compression effectiveness. EarlyTom reduces TTFT by up to 2.65x and FLOPs by up to 61% on a single NVIDIA A100 GPU for the LLaVA-OneVision-7B model, while maintaining accuracy comparable to the full-token baseline. These improvements substantially enhance the practicality of deploying Video-LLMs in real-world production scenarios.
Abstract (translated)
URL
https://arxiv.org/abs/2605.30010
PDF
-
Semantic and Visual Evidence for Efficient Long-Video Reasoning: A Solution for the HD-EPIC VQA Challenge
2026-05-28 05:53:34Yinsong Xu, Wei Jing, Liuxin Zhang, Wanjun Lv, Hui LiAbstract
Understanding long-form egocentric videos remains challenging for multimodal large language models (MLLMs) due to limited context length and insufficient grounding of fine-grained visual details. The recently proposed HD-EPIC benchmark highlights these limitations: even strong long-context models achieve relatively low performance across diverse video question answering tasks. In this paper, we propose a unified framework that decouples long-video reasoning into two complementary forms of evidence: semantic evidence and visual evidence. Semantic evidence captures global procedural structure through a coarse-to-fine extraction pipeline, while object-centric visual evidence preserves fine-grained grounding through bounding boxes and visual embeddings. During inference, we formulate reasoning as a query-conditioned evidence retrieval and integration process, dynamically selecting relevant information from both sources. Our approach achieves competitive performance in the HD-EPIC-VQA Challenge across multiple task categories. More broadly, our results demonstrate that explicitly structuring, retrieving, and integrating semantic and visual evidence is critical for effective long-video understanding with MLLMs.
Abstract (translated)
URL
https://arxiv.org/abs/2605.29402
PDF
-
From Pixels to Words -- Towards Native One-Vision Models at Scale
2026-05-27 17:59:55Haiwen Diao, Jiahao Wang, Penghao Wu, Yuhao Dong, Yuwei Niu, Yue Zhu, Zhongang Cai, Weichen Fan, Linjun Dai, Silei Wu, Xuanyu Zheng, Mingxuan Li, Yuanhan Zhang, Bo Li, Hanming Deng, Huchuan Lu, Quan Wang, Lei Yang, Lewei Lu, Dahua Lin, Ziwei LiuAbstract
Current vision-language models (VLMs) typically stitch together separate image encoders and language decoders via multi-stage alignment, a modular framework that inevitably fragments pixel-level signals across frames and scatters early pixel-word interactions. In parallel, native VLMs, despite impressive performance on single images, remain largely unexplored in multi-image, video understanding, and spatial intelligence. Hence, we introduce NEO-ov, a native foundation model that learns cross-frame and pixel-word correspondence end-to-end, without any external encoders, auxiliary adapters, or post-hoc fusion. By eliminating module boundaries entirely, NEO-ov enables fine-grained and unified spatiotemporal modeling to emerge natively inside the model. Notably, NEO-ov largely narrows the gap to modular counterparts while excelling at fine-grained visual perception, validating that native "one-vision" architectures are not only feasible but competitive at scale. Beyond empirical performance, we unveil systematic architectural analyses and detailed training recipes to facilitate subsequent native multimodal modeling. Our code and models are publicly available at: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2605.28820
PDF
-
VidPrism: Heterogeneous Mixture of Experts for Image-to-Video Transfer
2026-05-27 09:43:06Rui Lin, Chuanming Wang, Huadong MaAbstract
With the rapid development of pre-training technologies, adapting large-scale Vision-Language Models (VLMs) for video understanding \emph{\ie} image-to-video transfer learning has become a dominant paradigm. To achieve superior performance, it raises as an effective strategy among recent advances to employ Mixture-of-Experts (MoE) to enhance VLMs' temporal modeling capabilities. However, conventional MoE designs suffer from expert homogenization, where all experts act as identical generalists, inefficiently learning spatio-temporal features from undifferentiated video streams. To overcome this problem, we propose VidPrism, a novel heterogeneous temporal Mixture-of-Experts framework. VidPrism pioneers a division of labor by deploying functionally specialized experts, each assuming a role ranging from spatial understanding to temporal modeling. To feed these specialists appropriately, we introduce a content-aware, multi-rate sampling module that dynamically generates streams ranging from semantically rich to motion-focused representations, providing specialized inputs for experts. Furthermore, a dynamic, bidirectional fusion mechanism enables synergistic information exchange between these pathways, leading to a comprehensive video representation. Extensive experiments on various video recognition benchmarks demonstrate that VidPrism achieves state-of-the-art performance and effectively fosters expert specialization. Our source code is available at \href{this https URL}{this https URL}.
Abstract (translated)
URL
https://arxiv.org/abs/2605.28229
PDF
-
VCap: Hypergeometric Rewards for Weak-to-Strong Visual Captioning
2026-05-27 06:27:04Xingyu Lu, Jinpeng Wang, Yi-Fan Zhang, Yankai Yang, Yancheng Long, Yiyang Fan, Xuanyu Zheng, Haonan Fan, Kaiyu Jiang, Tianke Zhang, Changyi Liu, Bin Wen, Fan Yang, Tingting Gao, Han Li, Chun YuanAbstract
Visual captioning requires models to capture visual content faithfully while minimizing both omission and hallucination. As the dominant paradigm for captioning, MLLMs have achieved strong performance through scaling and high-quality data. Recently, RL has emerged as a key route to driving MLLMs toward higher precision and broader coverage, however, existing reward designs for captioning fail to provide fine-grained and reliable signals for factual verification, limiting their effectiveness. To address this, we propose VCap, a Witness-Adjudicator reward that pairs the reference caption (a witness) with the visual signal (an adjudicator). By explicitly verifying factual consistency between the reference and policy-generated captions grounded in the visual signal, VCap delivers a reward signal with hypergeometric-distribution-level precision for caption quality verification. This design enables effective learning even from imperfect references, facilitating weak-to-strong generalization in RL training. In our experiments, an 8B model trained with VCap outperforms open- and closed-source SOTA models on multiple image and video captioning benchmarks. Human evaluation further confirms its strong alignment with factual correctness. Additionally, VCap improves MLLM perceptual capability, generalizes across tasks, and surpasses best-of-N distillation, challenging prior assumptions about RLVR.
Abstract (translated)
URL
https://arxiv.org/abs/2605.28023
PDF
-
Tensor Memory: Fixed-Size Recurrent State for Long-Horizon Transformers
2026-05-26 21:03:42Kabir Swain, Sijie Han, Daniel Karl I. Weidele, Mauro Martino, Antonio TorralbaAbstract
Transformers process images and videos by flattening space and time into long token sequences. While attention and KV caching preserve past features, their memory grows with sequence length and they lack an explicit, persistent spatial state, making long-horizon video understanding and occlusion-sensitive reasoning difficult. We propose Tensor Memory, a lightweight module that augments Transformer blocks with a fixed-size recurrent 3D memory tensor: tokens write into a voxel grid via a differentiable soft write that deposits content as a Gaussian-weighted volume around a predicted continuous 3D location, the memory is updated with an efficient local interaction operator and gated recurrent dynamics, and tokens read back context via continuous sampling with gated residual fusion. Because the memory tensor has a constant size, Tensor Memory decouples state capacity from input length while preserving a spatial inductive bias. We evaluate the module on standard language, image, and video benchmarks and on a controlled toy diagnostic suite designed to isolate when persistent state is beneficial; it integrates with standard Transformer training pipelines and can be attached to or removed from existing blocks without other architectural changes.
Abstract (translated)
URL
https://arxiv.org/abs/2605.27686
PDF
-
OmniMem: Perturbation-aware Memory Compression for Streaming Audio-Visual LLMs
2026-05-26 19:12:13Guangzhi Sun, Yixuan Li, Yudong Yang, Chao ZhangAbstract
Audio-visual large language models (LLMs) hold strong promise for long-form video understanding, yet their long-video inference is fundamentally limited by the linear growth of video tokens and key-value (KV) caches. We present OmniMem, a memory-efficient streaming framework designed specifically for audio-visual LLMs. Unlike existing compression methods that treat all tokens uniformly, OmniMem introduces a modality-aware memory allocation strategy that separately manages visual and audio contexts, addressing the severe token imbalance between the two modalities. OmniMem further preserves informative and non-redundant KV states through perturbation-aware memory selection, enabling compact memory without sacrificing long-range understanding. To strengthen compression under realistic deployment constraints, we also explore budget-aware fine-tuning, which encourages the model to consolidate useful information into retained memory. Experiments on VideoMME Long, LVBench, and LVOmniBench with video-SALMONN 2+ and Qwen-2.5-Omni show that OmniMem consistently improves over strong training-free compression baselines by 2-4% absolute accuracy under the same memory budgets, with an additional 1-2% gain after fine-tuning.
Abstract (translated)
URL
https://arxiv.org/abs/2606.07577
PDF
