Videos

Video large language models (Video-LLMs) have made strong progress in general video understanding, but their ability to maintain temporal object consistency remains underexplored. Existing benchmarks often emphasize event recognition, action understanding, or coarse temporal reasoning, while rarely testing whether models can preserve the identity, state, and continuity of the same object across occlusion, disappearance, reappearance, state transitions, and cross-object interactions. We introduce TOC-Bench, a diagnostic benchmark for evaluating temporal object consistency in Video-LLMs. TOC-Bench is object-track grounded: each queried subject is linked to a per-frame trajectory and a structured temporal event timeline. To ensure that questions require temporally ordered visual evidence rather than language priors, single-frame shortcuts, or unordered frame cues, we design a three-layer temporal-necessity filtering protocol, which removes 60.7% of candidate QA pairs and retains 17,900 temporally dependent items across 10 diagnostic dimensions. From this pool, we construct a human-verified benchmark with 2,323 high-quality QA pairs over 1,951 videos. Experiments on representative Video-LLMs show that temporal object consistency remains a major unsolved challenge, with notable weaknesses in event counting, event ordering, identity-sensitive reasoning, and hallucination-aware verification, even when models perform well on general video understanding benchmarks. These results suggest that object-centric temporal coherence is a key bottleneck for current Video-LLMs, and that TOC-Bench provides a focused platform for diagnosing and improving object-aware temporal reasoning. The resource is available at https://github.com/cjzcjz666/toc_bench.git.

Next-generation visual assistants, such as smart glasses, embodied agents, and always-on life-logging systems, must reason over an entire day or more of continuous visual experience. In ultra-long video settings, relevant information is sparsely distributed across hours or days, making memory a fundamental challenge: models must accumulate information over time, recall prior states, track temporal order, and abstract recurring patterns. However, existing week-long video benchmarks are primarily designed for perception and recognition, such as moment localization or global summarization, rather than reasoning that requires integrating evidence across multiple days. To address this gap, we introduce EgoMemReason, a comprehensive benchmark that systematically evaluates week-long egocentric video understanding through memory-driven reasoning. EgoMemReason evaluates three complementary memory types: entity memory, tracking how object states evolve and change across days; event memory, recalling and ordering activities separated by hours or days; and behavior memory, abstracting recurring patterns from sparse, repeated observations over the whole week period. EgoMemReason comprises 500 questions across three memory types and six core challenges, with an average of 5.1 video segments of evidence per question and 25.9 hours of memory backtracking. We evaluate EgoMemReason on 17 methods across MLLMs and agentic frameworks, revealing that even the best model achieves only 39.6% overall accuracy. Further analysis shows that the three memory types fail for distinct reasons and that performance degrades as evidence spans longer temporal horizons, revealing that long-horizon memory remains far from solved. We believe EgoMemReason establishes a strong foundation for evaluating and advancing long-context, memory-aware multimodal systems.

World models for deformable objects should recover not only geometry and appearance, but also underlying physical dynamics, interaction grounding, and material behavior. Learning such a model from real videos is challenging because deformable linear, planar, and volumetric objects evolve under high-dimensional deformation, noisy interactions, and complex material response. The model must therefore infer a physical state from visual observations, roll it forward under new interactions, and render the resulting dynamics with high visual fidelity. We present DeformMaster, a video-derived interactive physics-neural world model that turns real interaction videos into an online interactive model of deformable objects within a unified dynamics-and-appearance framework. DeformMaster preserves structured physical rollout while using a neural residual to compensate for unmodeled effects, grounds sparse hand motion as distributed compliant actuator for hand-continuum interaction, represents material response with spatially varying constitutive experts, and drives high-fidelity 4D appearance from the predicted physical evolution. Experiments on real-world deformable-object sequences demonstrate DeformMaster's ability to roll out future dynamics and render dynamic appearance, outperforming state-of-the-art baselines while supporting novel action rollout, material-parameter variation, and dynamic novel-view synthesis. Project page: https://can-lee.github.io/deformmaster-web/

Sign Language Translation (SLT) converts sign language videos into spoken-language text, bridging communication between Deaf and hearing communities. Current gloss-free approaches rely on large encoder-decoder models, limiting deployment. We propose a compact 77M-parameter pipeline that couples MMPose skeletal pose extraction with a single linear projection into T5-small. By varying the input frame rate, we expose a practical efficiency trade-off: at 12 fps the model halves its sequence length, achieving a 75% reduction in encoder quadratic self-attention computational complexity while incurring only a modest BLEU-4 drop (9.53 vs. 10.06 at 24 fps on How2Sign). Our system is roughly 3x smaller than prior T5-base systems, demonstrating that a lightweight architecture can remain competitive without hierarchical encoders or large-scale models.

Video MLLMs face a persistent tension between spatial fidelity and temporal coverage: preserving fine-grained visual details requires many spatial tokens, while capturing short-lived events requires dense temporal sampling. We propose \textbf{Fre-Res}, a budget-adaptive dual-track video-token compression framework that separates these two forms of evidence. Fre-Res preserves sparse high-fidelity spatial anchors and represents dense temporal evolution through compact residual-frequency tokens. Specifically, it applies temporal 1D-DCT to inter-frame residual trajectories in vision-latent space, where we observe strong low-frequency concentration. To align frequency-domain dynamics with native visual embeddings, Fre-Res introduces a Spatial-Guided Absorber that injects temporal residual information into spatially corresponding anchor tokens. Across fine-grained short-video and long-video reasoning benchmarks, Fre-Res achieves a favorable accuracy--efficiency trade-off, matching or approaching full-token performance while substantially reducing visual-token length. Extensive ablations further show that temporal-frequency residuals preserve causal transition cues, while spatial anchors remain essential for fine-grained object and layout reasoning.

Selecting informative frames from long videos is a combinatorial problem that existing methods address either through efficient heuristics without explicit modeling of query-conditioned temporal structure, or through multi stage retrieval pipelines with substantial preprocessing cost. We propose \textbf{CREST}, a training-free frame selection method grounded in the temporal geometry of query--frame relevance. CREST is based on the observation that relevance over time exhibits structured local variation: sharp curvature around salient events and flatter regions in redundant segments. By using local curvature to guide selection, CREST allocates a fixed frame budget more effectively across brief decisive events and slowly evolving evidence. Under a fixed backbone and frame budget, CREST achieves higher accuracy than AKS, a lightweight relevance--coverage baseline, on LongVideoBench and VideoMME, while retaining 93--95% of the accuracy of MIRA, a stronger multi-stage retrieval pipeline, at only 3--4% of its preprocessing cost.\footnote{Code and implementation details are included in the supplementary material and will be released publicly upon acceptance.} On TempRel, our diagnostic benchmark for temporal frame selection, CREST achieves a 6.88% relative improvement over AKS. Pairwise LLM-as-a-judge evaluation further shows that CREST-selected frames yield more coherent frame-conditioned descriptions, with win rates of 60.58% and 54.50% on the two benchmarks. These results show that local temporal geometry provides a simple and efficient basis for long-video frame selection.

We propose Compressed Video Aggregator (CVA), a lightweight micro-video recommendation module that decouples video information from preference learning. It aggregates frozen VFM embeddings, and uses latent reasoning without cross-attention projection, producing compact video embeddings for recommenders. Due to the redundancy in the frame count of the original benchmark and its overly coarse sampling, we used titles to re-select key frames based on CLIP. Experiments on MicroLens and Short-Video show consistent gains with orders-of-magnitude reductions in training time and GPU memory, and re-selected frames can further enhance the performance of all methods, including CVA. Furthermore, we also discussed the impact of several scenarios involving erroneous titles on our method. Code will be released soon.

Recent "Thinking with Video" approaches use Video Generation Models (VGMs) for visual reasoning by producing temporally coherent Chain-of-Frames as reasoning artifacts. Even strong VGMs, however, exhibit two recurring failure modes on goal-directed tasks: long-horizon drift on multi-step tasks and mid-clip simulation errors that compound. Both stem from the absence of explicit reasoning built upon the VGM's short-horizon visual prior, a role naturally filled by Vision-Language Models (VLMs), but where to place the VLM is non-trivial: upfront plans commit before any frame is generated and post-hoc critiques over whole videos intervene too late. We propose VLM-VGM Collaborative Video Reasoning (CollabVR), a closed-loop framework that couples the VLM with the VGM at step-level granularity: the VLM plans the immediate next action, inspects the clip the VGM generates, and folds the verifier's diagnosis directly into the next action prompt to repair detected failures. On Gen-ViRe and VBVR-Bench, CollabVR improves both open-source and closed-source VGMs over single-inference, Pass@$k$, and prior test-time scaling baselines at matched compute, with the largest gains on the hardest tasks. It also yields further improvements on top of a reasoning-fine-tuned VGM, indicating that step-level VLM supervision is orthogonal to and stackable with reasoning-oriented fine-tuning. We provide video samples and additional qualitative results at our project page: https://joow0n-kim.github.io/collabvr-project-page.

Video anomaly detection (VAD) systems often prioritize accuracy while overlooking privacy concerns, limiting their suitability for real-world deployment. We propose the Orthogonal Projection Layer (OPL), a lightweight module that removes task-irrelevant variations to produce representations focused on anomaly-relevant cues. To address privacy risks in human-centered scenarios, we introduce Guided OPL (G-OPL), which suppresses facial attributes using weak supervision from face-presence signals while preserving non-identifying features such as pose and motion. A cosine alignment objective enforces consistent capture and removal of facial information without identity labels or adversarial training. We further present a privacy-aware evaluation framework that jointly assesses detection performance and privacy preservation, and enables analysis of how sensitive information is filtered. Experiments show that embedding privacy constraints into model design reduces sensitive information while maintaining or improving detection accuracy, supporting projection-based architectures as a principled approach for privacy-aware VAD.

Multimodal Large Language Models (MLLMs) have made rapid progress in single-video understanding, yet their ability to reason across multiple independent video streams remains poorly understood. Existing multi-video benchmarks rely largely on human-annotated real-world footage, limiting the precision of spatial, temporal, and physical ground truth and making it difficult to diagnose model failures. We introduce SYNCR, a controlled synthetic benchmark for cross-video reasoning with programmatically verified grounding. Built using Habitat, Kubric, and CLEVRER simulator engines, SYNCR contains 8,163 multi-video question-answer pairs grounded in 9,650 unique videos. It evaluates MLLMs across eight tasks spanning four diagnostic pillars: Temporal Alignment, Spatial Tracking, Comparative Reasoning, and Holistic Synthesis. Our zero-shot evaluation of leading open- and closed-weight MLLMs reveals a substantial gap between current models and humans: the best model achieves only 52.5% average accuracy, compared to an 89.5% human baseline. Models perform relatively well on temporal ordering but struggle with precise physical and spatial reasoning, with the best model reaching only 26.0% accuracy on Kinematic Comparison. We further find that parameter scaling and reasoning-specialized post-training improve temporal alignment capabilities, but do not reliably address fine-grained physical tracking or global spatial synthesis. Finally, an exploratory sim-to-real correlation analysis suggests that several SYNCR tasks track model-level trends on real-world multi-video benchmarks, while also exposing reasoning capabilities underrepresented by existing evaluations. Code available at https://github.com/SaraGhazanfari/SYNCR.

Video anomaly detection (VAD) aims to automatically identify events that deviate from normal patterns in untrimmed surveillance videos. Existing methods universally depend on large-scale annotations or task-specific training procedures, severely limiting their rapid deployment to novel scenes. We observe that intermediate-layer features of pre-trained multimodal large language models (MLLMs) already encode rich anomaly semantics, yet existing approaches rely on the language output pathway and fail to exploit the geometric discriminability latent in these representations. Based on this finding, we propose SphereVAD, a fully training-free, zero-shot VAD framework that recasts anomaly discrimination as von Mises-Fisher (vMF) likelihood-ratio geodesic inference on the unit hypersphere, unleashing latent discriminability through principled geometric reasoning rather than learning new representations. Specifically, SphereVAD first applies Frechet mean centering to unfold feature distributions and eliminate domain biases, then employs Holistic Scene Attention (HSA) to reinforce feature consistency using cross-video priors, and finally performs vMF-guided Spherical Geodesic Pulling (SGP) to align ambiguous segments with directional prototypes on the spherical manifold. This training-free pipeline requires only minimal synthetic images for calibration. SphereVAD establishes new state-of-the-art results among training-free approaches on three major benchmarks and remains competitive with fully supervised baselines. Code will be available upon acceptance.

Multimodal reward models have advanced substantially in text and image domains, yet progress in video understanding reward modeling remains severely limited by the lack of robust evaluation benchmarks and high-quality preference data. To address this, we propose a unified framework spanning benchmark design, data construction, and reward model training. We introduce Video Understanding Reward Bench (VURB), a benchmark featuring 2,100 preference pairs with long chain-of-thought reasoning traces (averaging 1,143 tokens) and majority voting evaluation across general, long, and reasoning-oriented video tasks. We further construct Video Understanding Preference Dataset (VUP-35K) via a fully automated pipeline, providing large-scale high-quality supervision for video reward training. Building on the data, we train VideoDRM and VideoGRM, a discriminative and a generative reward model, both achieving state-of-the-art performance on VURB and VideoRewardBench. Further analysis confirms that VUP-35K enhances both reward performance and model reasoning capability, while VideoDRM and VideoGRM yield significant gains under best-of-$N$ test-time scaling.

High-fidelity surgical video generation can greatly improve medical training and the development of AI, adapting these generative models for precise video editing remains a formidable challenge. Modifying surgical attributes, such as instrument tissue interactions or procedural phases is challenging due to the strict anatomical and temporal constraints. In this paper, we propose OphEdit, a novel training-free framework for the text-guided editing of ophthalmic surgical videos. Our approach leverages a deterministic second-order ODE inversion pipeline to capture Attention Value (V) tensors from the original video. By selectively injecting these stored tensors into the conditional Classifier-Free Guidance (CFG) branch during the denoising phase, OphEdit rigorously preserves the intricate anatomical geometry of the eye while seamlessly mapping text-driven semantic modifications onto the video stream. Clinical evaluations demonstrates that OphEdit effectively handles complex surgical transformations, such as instrument swaps and procedural variations, with superior structural fidelity and temporal consistency compared to natural-domain video editors. Our work represents the first application of training-free video editing in the ophthalmic surgical domain, offering a scalable solution for generating diverse, annotated medical datasets without the need for exhaustive manual recording or costly model fine-tuning. The code and prompts can be accessed at https://github.com/ophedit/OphEdit

Real-world audio-visual understanding requires chaining evidence that is sparse, temporally dispersed, and split across the visual and auditory streams, whereas existing benchmarks largely fail to evaluate this capability. They restrict videos to short clips, isolate modalities, or reduce questions to one-hop perception. We introduce TraceAV-Bench, the first benchmark to jointly evaluate multi-hop reasoning over long audio-visual trajectories and multimodal hallucination robustness. TraceAV-Bench comprises 2,200 rigorously validated multiple-choice questions over 578 long videos, totaling 339.5 hours, spanning 4 evaluation dimensions and 15 sub-tasks. Each question is grounded in an explicit reasoning chain that averages 3.68 hops across a 15.1-minute temporal span. The dataset is built by a three-step semi-automated pipeline followed by a strict quality assurance process. Evaluation of multiple representative OmniLLMs on TraceAV-Bench reveals that the benchmark poses a persistent challenge across all models, with the strongest closed-source model (Gemini 3.1 Pro) reaching only 68.29% on general tasks, and the best open-source model (Ming-Flash-Omni-2.0) reaching 51.70%, leaving substantial headroom. Moreover, we find that robustness to multimodal hallucination is largely decoupled from general multimodal reasoning performance. We anticipate that TraceAV-Bench will stimulate further research toward OmniLLMs that can reason coherently and faithfully over long-form audio-visual content.

Proactive streaming video understanding requires Video-LLMs to decide when to respond as a video unfolds, a task where existing methods often fall short due to their implicit, query-agnostic modeling of visual evidence. We introduce Response-G1, a novel framework that establishes explicit, structured alignment between the accumulated video evidence and the query's expected response conditions via scene graphs. The framework operates in three fine-tuning-free stages: (1) online query-guided scene graph generation from streaming clips; (2) memory-based retrieval of the most semantically relevant historical scene graphs; and (3) retrieval-augmented trigger prompting for per-frame "silence/response" decisions. By grounding both evidence and conditions in a shared graph representation, Response-G1 achieves more interpretable and accurate response timing decisions. Experimental results on established benchmarks demonstrate the superiority of our method in both proactive and reactive tasks, validating the advantage of explicit scene graph modeling and retrieval in streaming video understanding.

Video-language models (VLMs) face rapid inference costs as visual token counts scale with video length. For example, 32 frames at $448{\times}448$ resolution already yield >8,000 visual tokens in Qwen3-VL, making LLM prefill the dominant throughput bottleneck. Existing methods often rely on global similarity or attention-guided compression, incurring offsets to their gains. We propose \textbf{Temporal Token Fusion (TTF)}, a training-free, plug-and-play pre-LLM token compression framework that exploits structured temporal redundancy in video. TTF automatically selects an anchor frame, then for each subsequent frame, performs a local window similarity search (e.g.,$3\times 3$), fusing tokens that exceed a threshold. The compressed sequence maintains positional consistency across both prefill and decoding through coordinate realignment, enabling seamless integration with existing VLM pipelines. On Qwen3-VL-8B with threshold t=0.70, TTF removes about 67% of visual tokens while retaining 99.5% of the baseline accuracy and introducing only ${\approx}0.16$,GFLOPs of matching overhead. Overall, TTF offers a practical, efficient solution for video understanding. The code is available at \href{https://github.com/Cominder/ttf}{https://github.com/Cominder/ttf}

Understanding ultra-long videos such as egocentric recordings, live streams, or surveillance footage spanning days to weeks, remains a challenge. For current multimodal LLMs: even with million-token context windows, frame budgets cover only tens of minutes of densely sampled video, and most evidence is discarded before inference begins. Memory-augmented and agentic approaches help with scale, but their retrieval remains fragmented across modalities and lacks long-range narrative summaries that span days or weeks. We propose \textbf{MAGIC-Video}, a training-free framework built around a multimodal memory graph with interleaved narrative chain: the graph unifies episodic, semantic, and visual content through six typed edges and supports cross-modal retrieval, while the chain distils long-horizon entity biographies and recurring activity events. At inference time, an agentic loop interleaves graph retrieval with narrative fact injection, covering both the modality and time dimensions of ultra-long video in a single retrieval pipeline. On EgoLifeQA, Ego-R1 and MM-Lifelong, MAGIC-Video consistently outperforms strong general-purpose, long-video, and agentic baselines, with gains of 10.1, 7.4, and 5.9 points over the prior best agentic system on each benchmark. Code is available at https://github.com/lijiazheng0917/MAGIC-video.

Transformers dominate video recognition. They split videos into tokens, and processing them has expensive superlinear computational cost. Yet videos are filled with redundancy, so we can question the need for this expense. We introduce LookWhen, a selector-extractor framework that factorizes video recognition into learning when, where, and what to compute. Our shallow selector gets a scaled-down video and quickly scores all tokens across space-time, while our deep extractor gets the top-K selected tokens to approximate full-video representations without actually processing all the tokens. A key challenge is defining effective supervision for selection and extraction. For selection pre-training, we introduce a score on representations that ranks tokens by uniqueness using a simple nearest-neighbor distance. For extraction pre-training, we distill both a video teacher and an image teacher, for which we normalize its frame-wise representations to learn what changes within videos. Through these strategies, our selector-extractor learns general and efficient representations for feature extraction or fine-tuning to a task. Through experiments on Kinetics-400, SSv2, Epic-Kitchens, Diving48, Jester, and Charades, we show that LookWhen achieves a better accuracy-computation trade-off than efficient models and upgraded baselines of similar size. LookWhen Pareto-dominates in accuracy-FLOPs on 9 of 12 cases (6 tasks x 2 settings) and roughly matches on 3. In accuracy-throughput, measuring time in practice, LookWhen is more efficient still at 6.7x faster than InternVideo2-B at equal accuracy.

Medical multimodal large language models (MLLMs) have advanced image understanding and short-video analysis, but real clinical review often requires full-procedure video understanding. Unlike general long videos, medical procedures contain highly redundant anatomical views, while decisive evidence is temporally sparse, spatially subtle, and context dependent. Existing benchmarks often assume this evidence has already been localized through images, short clips, or pre-segmented videos, leaving the retrieval-before-reasoning problem under-tested. We introduce MedHorizon, an in-the-wild benchmark for long-context medical video understanding. MedHorizon preserves 759 hours of full-length clinical procedures and provides 1,253 evidence-grounded multiple-choice questionsthat jointly evaluate sparse evidence understanding and multi-hop clinical reasoning. Its evidence is extremely sparse, with only 0.166% evidence frames on average, requiring models to search noisy procedural streams before interpreting and aggregating findings. We evaluate representative general-domain, medical-domain, and long-video MLLMs. The best model reaches only 41.1% accuracy, showing that current systems remain far from robust full-procedure understanding. Further analysis yields four key findings: performance does not scale reliably with more frames, evidence retrieval and clinical interpretation remain primary bottlenecks; these bottlenecks are rooted in weak procedural reasoning and attention drift under redundancy, and generic sampling methods only partially balances local detail with global coverage. MedHorizon provides a rigorous testbed for MLLMs that retrieve sparse evidence and reason over complete clinical workflows.

In recent years, open-source efforts like Senorita-2M have propelled video editing toward natural language instruction. However, current publicly available datasets predominantly focus on local editing or style transfer, which largely preserve the original scene structure and are easier to scale. In contrast, Background Replacement, a task central to creative applications such as film production and advertising, requires synthesizing entirely new, temporally consistent scenes while maintaining accurate foreground-background interactions, making large-scale data generation significantly more challenging. Consequently, this complex task remains largely underexplored due to a scarcity of high-quality training data. This gap is evident in poorly performing state-of-the-art models, e.g., Kiwi-Edit, because the primary open-source dataset that contains this task, i.e., OpenVE-3M, frequently produces static, unnatural backgrounds. In this paper, we trace this quality degradation to a lack of precise background guidance during data synthesis. Accordingly, we design a scalable pipeline that generates foreground and background guidance in a decoupled manner with strict quality filtering. Building on this pipeline, we introduce Sparkle, a dataset of ~140K video pairs spanning five common background-change themes, alongside Sparkle-Bench, the largest evaluation benchmark tailored for background replacement to date. Experiments demonstrate that our dataset and the model trained on it achieve substantially better performance than all existing baselines on both OpenVE-Bench and Sparkle-Bench. Our proposed dataset, benchmark, and model are fully open-sourced at https://showlab.github.io/Sparkle/.

This research aims to solve the challenge of video retrieval from massive datasets, caused by ambiguous user queries. Prevailing single-round retrieval paradigms face a performance bottleneck, as they lack effective feedback mechanisms to handle complex search intentions. The root cause is the "Intent-Query Gap", where users' intent cannot be captured by a simple text query. To solve this, we propose the ADEPT framework: a training-free agent that pioneers an entropy-driven decision engine to efficiently guide dialogue by dynamically selecting between ASK and REFINE strategies. Experiments on two challenging datasets demonstrate that ADEPT significantly outperforms all non-interactive, heuristic, and Video-LLM baselines. The core contribution of this work is an efficient and interpretable entropy-driven interactive strategy that sets a new performance benchmark for the field of interactive video retrieval.

Video semantic search in densely crowded scenes remains a challenging task due to visual encoders tendency to prioritize salient foreground regions while neglecting contextually important, background areas. We propose an Inverse Attention Embedding mechanism that explicitly captures and highlights these overlooked regions. By combining inverse attention embeddings with traditional visual embeddings, our method significantly enhances semantic retrieval performance without additional training. Initial experiments and ablation studies demonstrate promising improvements over existing approaches in recall for video semantic search in crowded environments.

Recent large vision-language models have achieved strong performance on short- and medium-length video understanding, yet they remain inadequate for ultra-long or even infinite video reasoning, where models must preserve coherent memory over extended durations and infer causal dependencies across temporally distant events. Existing end-to-end video understanding methods are fundamentally limited by the $O(n^2)$ complexity of self-attention, while recent retrieval-augmented generation (RAG) approaches still suffer from fragmented clip-level memory, weak modeling of temporal and causal structure, and high storage and online inference costs. We present Event-Causal RAG, a lightweight retrieval-augmented framework for infinite long-video reasoning. Instead of indexing fixed-length clips, our method segments streaming videos into semantically coherent events and represents each event as a structured State-Event-State (SES) graph, capturing the event together with its surrounding state transitions. These graphs are merged into a global Event Knowledge Graph and stored in a dual-store memory that supports both semantic matching and causal-topological retrieval. On top of this memory, we design a bidirectional retrieval strategy to efficiently identify the most relevant event causal chains and provide them, together with the associated video evidence, to a backbone video foundation model for answer generation. Experiments on long-video understanding benchmarks demonstrate that Event-Causal RAG consistently outperforms strong clip-based retrieval baselines and long-context video models, particularly on questions requiring multi-event integration and causal inference across long temporal gaps, while also achieving improved memory efficiency and robust streaming performance.

Partially relevant video retrieval aims to retrieve untrimmed videos using text queries that describe only partial content. However, the inherent asymmetry between brief queries and rich video content inevitably introduces uncertainty into the retrieval process. In this setting, vague queries often induce semantic ambiguity across videos, a challenge that is further exacerbated by the sparse temporal supervision within videos, which fails to provide sufficient matching evidence. To address this, we propose Holmes, a hierarchical evidential learning framework that aggregates multi-granular cross-modal evidence to quantify and model uncertainty explicitly. At the inter-video level, similarity scores are interpreted as evidential support and modeled via a Dirichlet distribution. Based on the proposed three-fold principle, we perform fine-grained query identification, which then guides query-adaptive calibrated learning. At the intra-video level, to accumulate denser evidence, we formulate a soft query-clip alignment via flexible optimal transport with an adaptive dustbin, which alleviates sparse temporal supervision while suppressing spurious local responses. Extensive experiments demonstrate that Holmes outperforms state-of-the-art methods. Code is released at https://github.com/lijun2005/ICML26-Holmes.

The large-scale digitization of historical archives has created a paradox: "dark data"-digital objects lacking metadata for retrieval. Manual archival description is slow and expensive, limiting discovery and reuse. We propose Vidya, a modular pipeline that orchestrates Large Language Models (LLMs) and FOSS tools to automate semantic enrichment and archival ingestion at scale. Vidya constrains generations using YAML-defined ontologies and Pydantic validation, producing deterministic, structured JSON outputs from probabilistic models. Developed at Laboratory for Digital Humanities and Innovation (LAMUHDI) of the State University of Ponta Grossa (UEPG), Vidya applies Maker principles and open-source practices to enable low-cost deployment in memory institutions using modest hardware. We compare LLM performance and present a cost-benefit analysis showing major gains, reducing processing time from decades to days while complying with NOBRADE and ISAD(G).

Modern AI-generated videos are photorealistic at the single-frame level, leaving inter-frame dynamics as the main remaining axis for detection. Existing detectors typically handle this temporal evidence in three ways: feeding the full frame sequence to a generic temporal backbone, reducing one dominant temporal cue to fixed video-level descriptors, or comparing temporal features to real-video statistics through a detection metric. These strategies degrade sharply under cross-generator evaluation, where artifact type and timescale vary across generators. On caption-paired benchmark, GenVidBench, we identify two signatures that prior detectors do not jointly exploit: AI-generated videos exhibit smoother frame-to-frame temporal residuals at the pixel level, and more compact trajectories in the semantic feature space, indicating a temporal smoothness gap at both levels. We further observe that, when raw video is fed into a Spiking Neural Networks (SNNs), fake clips elicit firing predominantly at object and motion boundaries, unlike real clips, suggesting that the SNN responds to temporal artifacts localized at edges. These cues are sparse, asynchronous, and concentrated at moments of change, which makes SNNs a natural choice for this task: their event-driven, sparsely-activated dynamics align with the structure of the residual signal in a way that dense ANN backbones do not. Building on this observation, we propose MAST, a detector that processes multi-channel temporal residuals with a spike-driven temporal branch alongside a frozen semantic encoder for cross-generator generalization. On the GenVideo benchmark, MAST achieves 93.14% mean accuracy across 10 unseen generators under strict cross-generator evaluation, matching or surpassing the strongest ANN-based detectors and demonstrating the practical applicability of SNNs to AI-generated video detection.

Video large multimodal models increasingly face a scalability bottleneck: long videos produce excessively long visual-token sequences, which sharply increase memory and latency during inference. While existing compression methods are effective in specific settings, most are either weakly query-aware or apply a fixed compression policy across frames, proving suboptimal when visual evidence is unevenly distributed over time. To address this, we present VideoRouter, a query-adaptive dual-router framework built on InternVL for budgeted evidence allocation. The Semantic Router predicts the dominant allocation policy, choosing between broad temporal coverage and adaptive high-resolution preservation, while the Image Router uses early LLM layers to score frame relevance. This enables aggressive compression on less relevant frames while preserving detail on critical evidence frames. To train both routers, we build Video-QTR-10K for allocation-policy supervision and Video-FLR-200K for frame-relevance supervision. Experiments on VideoMME, MLVU, and LongVideoBench show that VideoRouter consistently improves over the InternVL baseline under comparable or lower budgets, achieving up to a 67.9% token reduction.

Existing affective understanding studies have mainly focused on recognizing emotions from images, audio signals, or pre-cliped video clips, where the affective evidence is already given. This passive and clip-centered setting does not fully reflect real-world scenarios, in which users often interact with long videos and express their needs through natural-language queries. In this paper, we study \textbf{Vague-Query-driven video Affective Understanding (VQAU)}, a new task that requires models to localize affective moments in long videos, predict their emotion categories, and generate evidence-grounded rationales under vague user queries. To support this task, we construct \textbf{VQAU-Bench}, a benchmark that integrates long videos, vague affective queries, temporal clip annotations, emotion labels, and rationale explanations into a unified evaluation framework. VQAU-Bench enables systematic assessment of semantic-temporal-affective alignment, affective moment localization, emotion classification, and rationale generation. To address the multi-step reasoning challenges of VQAU, we further propose \textbf{AffectSeek}, an agentic framework that actively seeks, verifies, and explains affective moments in long videos. AffectSeek decomposes VQAU into intent interpretation, candidate localization, clip verification, emotion reasoning, and rationale generation, and progressively aligns vague user intent with long-video evidence through role-specialized reasoning and cross-stage verification. Experiments show that VQAU remains challenging for existing affective recognition models and single-step vision-language models, while AffectSeek provides a simple yet effective framework for agentic long-video affective understanding.

Open-vocabulary human-object interaction (HOI) detection requires recognizing interaction phrases that may not appear as annotated categories during training. Recent vision-language HOI detectors improve semantic transfer by matching human-object features with text embeddings, but their predictions are often dominated by object affordance and phrase-level co-occurrence. As a result, a model may predict \textit{cut cake} from the presence of a knife and a cake without verifying whether the hand, tool, target, contact pattern, and object state jointly support the action. We propose \textbf{ScriptHOI}, a structured framework that represents each interaction phrase as a soft scripted state transition. Rather than treating a phrase as a single class token, ScriptHOI decomposes it into body-role, contact, geometry, affordance, motion, and object-state slots. A visual state tokenizer parses each detected human-object pair into corresponding state tokens, and a slot-wise matcher estimates both script coverage and script conflict. These two quantities calibrate HOI logits, expose missing visual evidence, and provide training constraints for incomplete annotations. To avoid suppressing valid but unannotated interactions, we further introduce interval partial-label learning, which constrains unannotated candidates with script-derived lower and upper probability bounds instead of assigning closed-world negatives. A counterfactual script contrast loss swaps individual script slots to discourage object-only shortcuts. Experiments on HICO-DET, V-COCO, and open-vocabulary HOI splits show that ScriptHOI improves rare and unseen interaction recognition while substantially reducing affordance-conflict false positives.

This companion paper reports the ICPR 2026 TVRID competition on privacy-aware top-view person re-identification. We present the competition setting, the released RGB-Depth dataset, and a summary of final results with descriptions of the top entries. TVRID contains 86 identities captured by four synchronized overhead Intel RealSense D455 cameras, with paired RGB/Depth streams and structured geometric variation across flat, ascent, descent, and oblique viewpoints. The evaluation protocol includes three tracks: RGB Re-ID, Depth Re-ID, and RGB$\leftrightarrow$Depth cross-modal retrieval. Submissions are ranked using mAP and CMC-1 under a unified server-side evaluation. The final results show a clear difficulty ordering (RGB $>$ Depth $>$ Cross-Modal), highlighting both the challenge of modality-constrained retrieval and the feasibility of strong performance with modality-invariant learning. By releasing the dataset at https://zenodo.org/records/17909410, the evaluation scripts at https://github.com/RaphaelDel/ICPR-TVRID, and the accompanying documentation, TVRID establishes a reproducible benchmark for top-view, depth-based, and cross-modal person re-id.

The exponential expansion of digital commerce in Indonesia has significantly shifted consumer interactions toward video-centric social networks, particularly YouTube. Consequently, the sheer volume of unstructured, multi-contextual comments poses a tremendous challenge for manual sentiment tracking. This study investigates and constructs a predictive model for customer satisfaction leveraging the Extreme Gradient Boosting (XGBoost) architecture coupled with Term Frequency-Inverse Document Frequency (TF-IDF) vectorization. By utilizing a secondary dataset of YouTube comments retrieved from e-commerce review videos, the raw text underwent rigorous preprocessing to generate normalized numerical features. The experimental results demonstrate that the PyCaret-optimized machine learning framework delivers superior classification resilience. Beyond standard performance metrics, lexical evaluations and feature-importance mapping uncover a notable phenomenon: e-commerce discourse is heavily infiltrated by socio-political terminologies, which ultimately influence the polarity of audience satisfaction.

Video text-based visual question answering (Video TextVQA) aims to answer questions by reasoning over visual textual content appearing in videos. Despite the strong multimodal video understanding capabilities of recent Video-LLMs, their performance on existing Video TextVQA benchmarks remains limited. To better understand this gap, we conduct an upper-bound analysis through frame-wise question answering, counting a sample as correct if any frame yields the right answer, which significantly outperforms direct video-based inference and reveals a substantial performance gap. The results suggest that the primary bottleneck lies in the localization of key question-relevant evidence, rather than in reasoning capacity itself. Building on this insight, we propose a question-guided agent framework that explicitly anchors the relevant keyframes before answering. The approach operates effectively in a training-free setting and consistently surpasses direct video inference. With additional supervised fine-tuning (SFT) and reinforcement learning (RL), it achieves an average improvement of +12.12 in accuracy and +11.15 in ANLS across benchmarks, establishing new state-of-the-art results. Our study underscores the critical role of explicit keyframe anchoring for advancing Video TextVQA. The code will be publicly released.

While Video Large Language Models (Video-LLMs) excel in general understanding, they exhibit systematic deficits in fine-grained physical reasoning. Existing interventions not only suffer from limited generalization but fundamentally conflate generative artifacts with genuine physical fallacies. Furthermore, we find that models fail systematically not only in anti-physics anomalies but also in counter-intuitive scenarios where visual facts contradict statistical expectations. Accordingly, we propose the Unified Attribution Theory: this dual failure stems not from perception deficiency, but from Semantic Prior Dominance -- the reasoning mechanism is deeply hijacked by internal narrative scripts. To address this, we construct the Programmatic Adversarial Curriculum (PACC), the first high-fidelity adversarial video dataset synthesized based on physical laws, thoroughly decoupling visual artifacts from logical errors. Concurrently, we design the Visual-Anchored Reasoning Chain (VARC) to force models to explicitly ground their judgments in low-level visual facts prior to logical adjudication. Experiments demonstrate that without invasive architectural modifications, standard LoRA fine-tuning with the PACC curriculum effectively neutralizes prior interference in state-of-the-art (SOTA) models, yielding a substantial leap in physical reasoning capabilities.

While Test-Time Scaling (TTS) offers a promising direction to enhance video generation without the surging costs of training, current test-time video generation methods based on diffusion models suffer from exorbitant candidate exploration costs and lack temporal guidance. To address these structural bottlenecks, we propose shifting the focus to streaming video generation. We identify that its chunk-level synthesis and few denoising steps are intrinsically suited for TTS, significantly lowering computational overhead while enabling fine-grained temporal control. Driven by this insight, we introduced Stream-T1, a pioneering comprehensive TTS framework exclusively tailored for streaming video generation. Specifically, Stream-T1 is composed of three units: (1) Stream -Scaled Noise Propagation, which actively refines the initial latent noise of the generating chunk using historically proven, high-quality previous chunk noise, effectively establishes temporal dependency and utilizing the historical Gaussian prior to guide the current generation; (2) Stream -Scaled Reward Pruning, which comprehensively evaluates generated candidates to strike an optimal balance between local spatial aesthetics and global temporal coherence by integrating immediate short-term assessments with sliding-window-based long-term evaluations; (3) Stream-Scaled Memory Sinking, which dynamically routes the context evicted from KV-cache into distinct updating pathways guided by the reward feedback, ensuring that previously generated visual information effectively anchors and guides the subsequent video stream. Evaluated on both 5s and 30s comprehensive video benchmarks, Stream-T1 demonstrates profound superiority, significantly improving temporal consistency, motion smoothness, and frame-level visual quality.

Video vision-language models (VLMs) keep paying for visual state the stream already told us was stable. The factory wall did not move, but most VLM pipelines still hand the model dense RGB frames or a fresh prefix again. We study that waste as training-free anti-recomputation: reuse state when validation says it survives, and buy fresh evidence when the scene, query, or cache topology requires it. The largest measured win is after ingest. On frozen Qwen2.5-VL-7B-Instruct-4bit, adaptive same-video follow-up reuse preserves paired choices and correctness on a 93-query VideoMME breadth setting while reducing follow-up latency by 14.90-35.92x. The first query is still cold; the win starts when later questions reuse the same video state. Stress tests bound the result: repeated-question schedules hold through 50 turns, while dense-answer-anchored prompt variation separates conservative fixed K=1 repair from faster aggressive policies that drift. Fresh-video pruning is smaller but real. C-VISION skips timed vision-tower work before the first answer is generated. On Gemma 4-E4B-4bit, the clean 32f short cell reaches 1.316x first-query speedup with no paired drift or parse failures on 20 items; Qwen shows the fidelity/speed boundary. Stage-share ceiling (C-CEILING) is the accounting guardrail: a component speedup becomes an end-to-end speedup only in proportion to the wall-clock share it accelerates, so C-VISION and after-ingest follow-up reuse do not multiply. Candidate C-STREAM remains a native-rate target, not a headline result here. The broader direction is VLM-native media that expose change, motion, uncertainty, object state, sensor time, and active tiles directly, so models do not have to rediscover the world from dense RGB every frame.

Videos are unique in their ability to capture actions which transcend multiple frames. Accordingly, for many years action recognition was the quintessential task for video understanding. Unfortunately, due to a lack of sufficiently diverse and challenging data, modern vision-language models (VLMs) are no longer evaluated on their action recognition capabilities. To revitalize action recognition in the era of VLMs, we advocate for a returned focus on domain-specific actions. To this end, we introduce VideoNet, a domain-specific action recognition benchmark covering 1,000 distinct actions from 37 domains. We begin with a multiple-choice evaluation setting, where the difference between closed and open models is stark: Gemini 3.1 Pro attains 69.9% accuracy while Qwen3-VL-8B gets a mere 45.0%. To understand why VLMs struggle on VideoNet, we relax the questions into a binary setting, where random chance is 50%. Still, Qwen achieves only 59.2% accuracy. Further relaxing the evaluation setup, we provide $k\in\{1,2,3\}$ in-context examples of the action. Some models excel in the few-shot setting, while others falter; Qwen improves $+7.0\%$, while Gemini declines $-4.8\%$. Notably, these gains fall short of the $+13.6\%$ improvement in non-expert humans when given few-shot examples. Finding that VLMs struggle to fully exploit in-context examples, we shift from test-time improvements to the training side. We collect the first large-scale training dataset for domain-specific actions, totaling nearly 500k video question-answer pairs. Fine-tuning a Molmo2-4B model on our data, we surpass all open-weight 8B models on the VideoNet benchmark.

Long-video understanding with multimodal language models suffers from three compounding bottlenecks: heavy decode cost to obtain dense RGB frames, quadratic token growth with frame count, and weak motion perception under sparse keyframe sampling. We present HY-Himmel, a hierarchical video-language framework that allocates semantic and motion capacity separately. A small set of sparse anchor I-frames is routed to the expensive host ViT to ground object identity and scene layout, while the far denser inter-frame intervals are encoded by a lightweight compressed-domain tri-stream adapter that distils motion evidence from motion-vector maps, residual maps, and I-frame context into aligned motion tokens. These tokens are injected into the LLM via a differentiable placeholder mechanism after a dedicated Stage-1 contrastive alignment that places the motion representation in a geometry compatible with the frozen visual backbone. On Video-MME, HY-Himmel surpasses the dense 32-frame baseline by +2.3 pp (61.2 to 63.5%) while using 3.6x fewer context tokens. Extensive ablations over stream composition, motion encoder family, fusion mode, alignment objective, anchor count, LoRA rank, and video duration confirm that the full tri-stream is necessary and sufficient for the observed gains.

Recently, video language models (VLMs) have been applied in various fields. However, the visual token sequence of the VLM is too long, which may cause intolerant inference latency and GPU memory usage. Existing methods propose mixed-precision quantization to the key-value (KV) cache in VLMs based on token granularity, which is time-consuming in the search process and hardware inefficient during computation. This paper introduces a novel approach called WindowQuant, which employs window-adaptive mixed-precision quantization to optimize the KV cache. WindowQuant consists of two modules: window-level quantization search and window-level KV cache computation. Window-level quantization search quickly determines the optimal bit-width configuration of the KV cache windows based on the similarity scores between the corresponding visual token windows and the text prompt, maintaining the model accuracy. Furthermore, window-level KV cache computation reorders the KV cache windows before quantization, avoiding the hardware inefficiency caused by mixed-precision quantization in inference computation. Extensive experiments demonstrate that WindowQuant outperforms state-of-the-art VLM models and KV cache quantization methods on various datasets.

Capturing user interests across extensive watch histories is critical for short-form video recommendation, yet scaling sequence length is limited by two bottlenecks: the semantic sparsity of atomic Video IDs and the quadratic computational complexity of Transformers. Traditional orthogonal Video IDs fail to capture content relationships and demand large embedding tables, while the quadratic complexity of self-attention restricts the maximum sequence length under strict industrial latency and resource constraints. In this work, we present a production-deployed framework for modeling ultra-long user behavior sequences at a billion-user scale. We first address the representation bottleneck by adopting content-native Semantic IDs. By utilizing depth-truncated, coarse-grained Semantic IDs, we shrink the embedding table size from corpus cardinality. This compact representation naturally generalizes to cold-start content through shared semantic prefixes. Second, to overcome the sequence scaling barrier, we introduce a Global-Aware Compression Transformer that leverages non-parametric temporal folding and unified global query integration to effectively condense the sequence, alleviating both the memory and computational bottlenecks of standard self-attention. Offline profiling on our computing infrastructure demonstrates an order-of-magnitude reduction in peak memory footprint and a drastic decrease in computational overhead. This efficiency gain enables supporting longer sequence lengths at an affordable cost in production, yielding substantial online gains in satisfied user engagement and satisfied content consumption in large-scale online A/B tests.

Streaming video understanding requires processing unbounded video streams with limited memory and computation, posing two key challenges. First, continuously constructing new and evicting old key-value(KV) caches is required for unbounded streams. Secondly, due to the high cost of collecting and training on unbounded streams, models must learn from short sequences while generalizing to long streams. Existing streaming VideoVLLMs fail to scale to unbounded video streams or focus on cache reuse strategies, leaving the impact of cache construction underexplored. In this paper, we propose Decoupled Streaming Cache(DSCache), a training-free cache construction mechanism that adapts pretrained offline models to streaming settings. DSCache maintains a cumulative past KV cache while constructing a separate instant cache on-demand, decoupled from past caches to preserve the informativeness of recent inputs. To enable position extrapolation beyond the training length, DSCache further incorporates a position-agnostic encoding strategy, ensuring KV caches to support unseen positions and preventing position overflow. Experiments on Streaming Video QA benchmarks demonstrate DSCache's state-of-the-art performance, with an average 2.5% accuracy gains over prior methods.

Bandwidth-constrained robotic and surveillance systems often rely on a single compressed video stream to support both continuous scene awareness and downstream machine perception. In practice, this creates a mismatch: low-bitrate video can preserve motion and coarse context, but often loses the fine local detail needed for reliable object recognition and decision-making. Motivated by a hybrid architecture in which low-resolution video supports dynamic scene understanding while eventdriven high-detail regions of interest (ROIs) support close-up identification and analytics, this paper formalizes a two-channel visual telemetry scheme in which a continuous low-bitrate video stream is augmented by selectively transmitted high-detail still ROIs. This first paper does not attempt to prove the superiority of a new still-image codec. Instead, it establishes the hybrid transmission paradigm itself using a practical and reproducible codec stack: x265/HEVC for the base video stream and JPEG stills for ROI refinement. We formulate the problem as bitrate-constrained information selection for robotic vision and define an experimental protocol in which video-only and hybrid schemes are compared under matched total communication budgets. The study is designed around UAV-oriented datasets, two practical bitrate regimes, several ROI triggering policies, and object-level classification refinement on selectively transmitted ROI stills. The resulting paper lays the methodological foundation for a second-stage investigation of JPEG AI as the semantic still-image channel within the same hybrid architecture.

Constructing controllable visual data is a major bottleneck for image editing and multimodal understanding. Useful supervision is rarely produced by a single rendering pass; instead it emerges through iterative generation, inspection, correction, filtering, and export. We present DataEvolver, a closed-loop visual data engine that organizes this process around explicit goals, persistent artifacts, bounded corrective actions, and acceptance decisions. DataEvolver supports multiple artifact types, including RGB images, masks, depth maps, normal maps, meshes, poses, trajectories, and review traces. In the current release, the system operates through two coupled loops: generation-time self-correction within each sample and validation-time self-expansion across dataset rounds. We validate the framework on an image-level object-rotation setting. With a fixed Qwen-Edit LoRA probe, our final Ours+DualGate model outperforms both the unadapted base model and a public multi-angle LoRA on SpatialEdit and a held-out evaluation set. Ablations show a consistent improvement path from scene-aware generation to feedback-driven correction and dual-gated validation. Beyond the released rotation data, our main contribution is a reusable framework for building visual datasets through explicit goal tracking, review, correction, and acceptance loops.

Existing large-scale sign language resources typically provide supervision only at the level of raw video-text alignment and are often produced in laboratory settings. While such resources are important for semantic understanding, they do not directly provide a unified interface for open-world recognition and translation, or for modern pose-driven sign language video generation frameworks: 1. RGB-based pretrained recognition models depend heavily on fixed backgrounds or clothing conditions during recording, and are less robust in open-world settings than style-agnostic pose-processing models. 2. Recent pose-guided image/video generation models mostly use a unified keypoint representation such as DWPose as their control interface. At present, the sign language field still lacks a data resource that can directly interface with this modern pose-native paradigm while also targeting real-world open scenarios. We present SignVerse-2M, a large-scale multilingual pose-native dataset for sign language pose modeling and evaluation. Built from publicly available multilingual sign language video resources, it applies DWPose in a unified preprocessing pipeline to convert raw videos into 2D pose sequences that can be used directly for modeling, resulting in a consolidated corpus of about two million clips covering more than 55 sign languages. Unlike many laboratory datasets, this resource preserves the recording conditions and speaker diversity of real-world videos while reducing appearance variation through a unified pose representation. Toward this goal, we further provide the data construction pipeline, task definitions, and a simple SignDW Transformer baseline, demonstrating the feasibility of this resource for multilingual pose-space modeling and its compatibility with modern pose-driven pipelines, while discussing the evaluation claims it can support as well as its current limitations.

Large vision-language models (VLMs) have advanced multimodal tasks such as video question answering (QA). However, VLMs face the challenge of selecting frames effectively and efficiently, as standard uniform sampling is expensive and performance may plateau. Inspired by active perception theory, which posits that models gain information by acquiring data that differs from their expectations, we introduce Video Active Perception (VAP), a training-free method to enhance long-form video QA using VLMs. Our approach treats keyframe selection as data acquisition in active perception and leverages a lightweight text-conditioned video generation model to represent prior world knowledge. Empirically, VAP achieves state-of-the-art zero-shot results on long-form or reasoning video QA datasets such as EgoSchema, NExT-QA, ActivityNet-QA, IntentQA, and CLEVRER, achieving an increase of up to 5.6 x frame efficiency by frames per question over standard GPT-4o, Gemini 1.5 Pro, and LLaVA-OV. Moreover, VAP shows stronger reasoning abilities than previous methods and effectively selects keyframes relevant to questions. These findings highlight the potential of leveraging active perception to improve the frame effectiveness and efficiency of long-form video QA.

The rapid growth of video content across domains such as surveillance, education, and social media has made efficient content understanding increasingly critical. Video summarization addresses this challenge by generating concise yet semantically meaningful representations, but existing approaches often rely on expensive manual annotations, struggle to generalize across domains, and incur significant computational costs due to complex architectures. Moreover, unsupervised and weakly supervised methods typically underperform compared to supervised counterparts in capturing long-range temporal dependencies and semantic structure. In this work, we propose TRIMMER (Temporal Relative Information Maximization for Multi-objective Efficient Reinforcement), a novel self-supervised reinforcement learning framework for video summarization. TRIMMER operates in two stages: it first learns robust representations via self-supervised learning and then performs spatio-temporal decision making through reinforcement learning guided by information-theoretic reward functions. Unlike prior approaches that rely on similarity-based objectives, our method introduces entropy-based metrics to capture higher-order temporal dynamics and semantic diversity, while computing rewards directly over selected frame indices to improve computational efficiency. Extensive experiments on standard benchmarks demonstrate that TRIMMER achieves state-of-the-art performance among unsupervised and self-supervised methods, while remaining competitive with leading supervised approaches, highlighting its effectiveness for scalable and generalizable video summarization.

This report presents the results and findings of the first edition of the Short-Films 20K (SF20K) Competition, held in conjunction with the SLoMO Workshop at ICCV 2025. The competition is designed to advance story-level video understanding beyond short-clip action recognition, introducing an open-ended video question-answering task built on a corpus of amateur short films. This setup ensures that models must rely on multimodal understanding rather than memorization of popular movies. Evaluation is conducted using the SF20K-Test benchmark (95 movies, 979 question-answer pairs) and scored via LLM-QA-Eval, an automated judge based on GPT-4.1-nano. The competition attracted 22 teams and 286 submissions across two tracks: a Main Track with unrestricted model size and a Special Track limited to models under 8 billion parameters. The winning team achieved 65.7% accuracy on the Main Track and 48.7% on the Special Track, against a human performance ceiling of 91.7%. Our analysis reveals several key findings: narrative-aware, shot-level processing consistently outperforms uniform frame sampling; well-designed multi-stage pipelines using smaller models can match or exceed end-to-end inference with models over 30x larger; and subtitle quality is a dominant factor in performance. These results highlight that the primary bottleneck in long-form video QA lies in information selection and reasoning structure rather than raw model capacity, and that a substantial gap remains between current methods and human-level narrative comprehension.

Tensor completion has emerged as a powerful framework for recovering missing data in multidimensional signals by exploiting low-rank tensor structures. Among existing approaches, linear transform-based tensor nuclear norm (TNN) methods have achieved considerable success by enforcing low-rankness on transformed frontal slices. However, the low-rank structure revealed by linear transforms remains inherently limited. To better capture intrinsic correlations, nonlinear transform-based TNN (NTTNN) models have been proposed, significantly enhancing low-rank representation through composite transforms. Despite their effectiveness, existing NTTNN methods are restricted to real-valued tensors and fail to model quaternion-valued data, which are essential for preserving inter-channel dependencies in color images and videos. Extending nonlinear TNN models to the quaternion domain is challenging due to the non-commutativity of quaternion multiplication and the complexity of quaternion singular value decomposition. To address the limitations encountered in prior works, we propose a quaternion nonlinear transform-induced tensor nuclear norm (QNTTNN) via a real embedding of quaternions, enabling tractable nuclear norm definitions and efficient optimization. Building upon QNTTNN, we formulate a quaternion tensor completion model and develop a proximal alternating minimization algorithm with rigorous convergence guarantees. Extensive experiments on benchmark color video inpainting datasets validate the superior performance of the proposed method over existing approaches.

The growing availability of health-related instructional videos creates new opportunities for clinical training, patient rehabilitation, and health education, yet existing retrieval systems remain largely single-turn: a user submits one query and receives one ranked list. This interaction is brittle in health scenarios, where information needs are often vague at first and become clinically meaningful only after follow-up constraints such as posture, hand placement, contraindications, equipment, or patient condition are specified. We introduce interactive multi-turn semantic retrieval for health videos and construct MHVRC, a Multi-Turn Health Video Retrieval Corpus, by combining video-grounded descriptions from VideoChat-Flash with query refinements generated by DeepSeek. We further propose DATR, a Dialogue-Aware Two-Stage Retrieval framework. DATR first performs efficient coarse retrieval with a CLIP-style dual encoder and sparse frame sampling, then re-ranks the top candidates through multi-turn query fusion and a lightweight cross-encoder scoring module. Experiments on MHVRC show consistent gains over strong text-video retrieval baselines, while user studies indicate that refined multi-turn queries better capture fine-grained procedural semantics than single-turn annotations. The work establishes a benchmark and a scalable technical recipe for interactive health video retrieval.

Existing benchmarks for Vision-Language Models (VLMs) primarily evaluate spatio-temporal understanding on simple single-action videos, closed attribute sets and restricted entity types, failing to capture the freeform, multi-action interactions between diverse entities which characterize real-world video understanding. Furthermore, the lack of a systematic framework for analyzing model failures across complementary spatio-temporal axes hinders comprehensive evaluation. To address these gaps, we introduce VISTA, a Video Interaction Spatio-Temporal Analysis benchmark designed for open-set, multi-entity and multi-action spatio-temporal understanding in VLMs. VISTA decomposes videos into interpretable entities, their associated actions, and relational dynamics, enabling multi-axis diagnostics and unified assessment of relational, spatial, and temporal understanding. Our benchmark integrates multiple datasets into a single interaction-aware taxonomy and comprises ~12K curated video-query pairs spanning diverse scenes and complexities. We systematically evaluate 11 state-of-the-art VLMs on VISTA, and break down aggregate performance across our taxonomy to reveal shortcomings and pronounced spatio-temporal biases obscured by traditional metrics. By providing detailed, taxonomy-driven diagnostics on a challenging dataset, VISTA offers a nuanced framework to guide advances in model design, pretraining strategies, and evaluation protocols. Overall, VISTA is the first, large-scale, interaction-aware diagnostic benchmark for spatio-temporal understanding in VLMs.

Although reinforcement learning (RL) has significantly advanced reasoning capabilities in large multimodal language models (MLLMs), its efficacy remains limited for lightweight models essential for edge deployments. To address this issue, we leverage causal analysis and experiment to reveal the underlying phenomenon of perceptual bias, demonstrating that RL-based fine-tuning compels lightweight models to preferentially adopt perceptual shortcuts induced by data biases, rather than developing genuine reasoning abilities. Motivated by this insight, we propose VideoThinker, a causal-inspired framework that cultivates robust reasoning in lightweight models through a two-stage debiasing process. First, the Bias Aware Training stage forges a dedicated "bias model" to embody these shortcut behaviors. Then, the Causal Debiasing Policy Optimization (CDPO) algorithm fine-tunes the primary model, employing an innovative repulsive objective to actively push it away from the bias model's flawed logic while simultaneously pulling it toward correct, generalizable solutions. Our model, VideoThinker-R1, establishes a new state-of-the-art in video reasoning efficiency. For same-scale comparison, requiring no Supervised Fine-Tuning (SFT) and using only 1 of the training data for RL, it surpasses VideoRFT-3B with a 3.2% average gain on widely-used benchmarks and a 7% lead on VideoMME. For cross-scale comparison, it outperforms the larger Video-UTR-7B model on multiple benchmarks, including a 2.1% gain on MVBench and a 3.8% gain on TempCompass. Code is available at https://github.com/falonss703/VideoThinker.

The development of video game streaming has grown rapidly, with major platforms such as YouTube and Twitch using different codecs. To support quality assessment models that work consistently across any codec, it is necessary to have access to large, diverse subjective gaming quality datasets. Currently, there are only a few available, each having limitations. To address this gap, we present the largest gaming video quality dataset to date, incorporating both user-generated content (UGC) and professional-generated content (PGC) with extensive visual diversity. Our dataset covers the most widely used codecs - H.264, H.265, and AV1 - and consists of 4,048 video samples, each annotated by an average of 37 mean opinion score (MOS) ratings. In addition to overall quality scores, we collect coarse-grained quality attributes, enabling a better understanding of perceptual factors. We study the performance of leading video quality assessment methods on this dataset, including a vision language model that outperforms all the benchmarks. To the best of our knowledge, this is the first dataset that comprehensively addresses gaming video quality assessment across multiple codecs and content types with quality attributes. Our dataset is publicly available at https://rajeshsureddi.github.io/GameScope/.

Understanding human actions from visual observations is essential for human--robot interaction, particularly when semantic interpretation of unfamiliar or hard-to-annotate actions is required. In scenarios such as rapid and less common activities, collecting sufficient labeled data for supervised learning is challenging, making zero-shot approaches a practical alternative for semantic understanding without task-specific training. While recent advances in large-scale pretrained models enable such zero-shot reasoning, the impact of temporal resolution, especially for rapid and fine-grained motions, remains underexplored. In this study, we investigate how temporal resolution affects zero-shot semantic understanding of high-speed human actions. Using kendo as a representative case of rapid and subtle motion patterns, we propose a training-free pipeline that combines a pre-trained video-language model for semantic representation with large language model-based reasoning for pairwise action comparison. Through controlled experiments across multiple frame rates (120 Hz, 60 Hz, and 30 Hz), we show that higher temporal resolution significantly improves semantic separability in zero-shot settings. We further analyze the role of tracking-based human joint information under both full and partial observation scenarios. Quantitative evaluation using a nearest-class prototype strategy demonstrates that high-speed video provides more stable and interpretable semantic representations for fast actions. These findings highlight the importance of temporal resolution in training-free action recognition and suggest that high-speed perception can enhance semantic understanding capabilities.

Multi-modal large language models (MLLMs) advance vision language understanding but face inherent limitations in long-video tasks due to bounded perception context budgets. Existing agentic methods mitigate this via rule-based preprocessing, yet often suffer from information loss, high cost, and reliance on textual intermediates. We propose MACF, an end-to-end Multi-Agent Collaboration Framework that decouples per-agent perception budgets from global video complexity, enabling scalable video understanding while preserving visual fidelity. MACF partitions videos into segments for locally budgeted agents and enables holistic reasoning via an agent-native latent communication protocol. Each agent encodes partial observations into compact, task-sufficient tokens in a shared embedding space, allowing efficient and information-preserving collaboration by a central coordinator. We introduce a curriculum training strategy that progressively enforces semantic alignment, evidence summarization, and cross-agent coordination. Extensive experiments on diverse video understanding benchmarks show that MACF consistently outperforms state-of-the-art MLLMs and multi-agent systems under identical budget constraints, demonstrating the effectiveness of our latent collaboration for scalable video understanding.

Video platforms such as YouTube have reshaped how users engage with entertainment and information, emphasizing brief, highly engaging content such as Shorts. Within this ecosystem, certain content occupies a gray area where it remains allowed but may still have unintended negative effects on some audiences. To study this problem, we introduce TwistedHumor, a dataset of 1,211 YouTube Shorts paired with 33,041 related comments, with hand annotations for humor presence, humor type, harm, topic, rhetorical devices, and stand up context. Beyond dataset creation, we present a multi view analysis of how humor and harm appear in short form social media. Using LLooM based concept induction over video descriptions, we find that dark humor frequently clusters around themes of critique, coping, awkwardness, and identity expression rather than appearing as a single uniform category. We further analyze audience response through linked comments and show that regular humor is associated with more positive sentiment, while dark humor receives more mixed, neutral, and sometimes more toxic reactions. Finally, we evaluate large language models against human annotations and find that they perform better on stand up comedy compared to shorter jokes. Together, these results position TwistedHumor not only as a new benchmark, but as an empirical study of the gray area between humor and harm in short form video, highlighting the need for context aware moderation and more robust multimodal evaluation.

Due to the scarcity of large-scale in-the-wild triplet data and the improper use of masks, the performance of video virtual try-on models remains limited. In this paper, we first introduce TripVVT-10K, the largest and most diverse in-the-wild triplet dataset to date, providing explicit video-level cross-garment supervision that existing video datasets lack. Built upon this resource, we develop TripVVT, a Diffusion Transformer-based framework that replaces fragile garment masks with a simple, stable human-mask prior, enabling reliable background preservation while remaining robust to real-world motion, occlusion, and cluttered scenes. To support comprehensive evaluation, we further establish TripVVT-Bench, a 100-case benchmark covering diverse garments, complex environments, and multi-person scenarios, with metrics spanning video quality, try-on fidelity, background consistency, and temporal coherence. Compared to state-of-the-art academic and commercial systems, TripVVT achieves superior video quality and garment fidelity while markedly improving generalization to challenging in-the-wild videos. We publicly release the dataset and benchmark, which we believe provide a solid foundation for advancing controllable, realistic, and temporally stable video virtual try-on.

Existing talking avatar methods typically adopt an image-to-video pipeline conditioned on a static reference image within the same scene as the target generation. This restricted, single-view perspective lacks sufficient temporal and expression cues, limiting the ability to synthesize high-fidelity talking avatars in customized backgrounds. To this end, we introduce Talking Avatar generation from Video Reference (TAVR), a novel framework that shifts the paradigm by leveraging cross-scene video inputs. To effectively process these extended temporal contexts and bridge cross-scene domain gaps, TAVR integrates a token selection module alongside a comprehensive three-stage training scheme. Specifically, same-scene video pretraining establishes foundational appearance copying, which is subsequently expanded by cross-scene reference fine-tuning for robust cross-scene adaptation. Finally, task-specific reinforcement learning aligns the generated outputs with identity-based rewards to maximize identity similarity. To systematically evaluate cross-scene robustness, we construct a new benchmark comprising 158 carefully curated cross-scene video pairs. Extensive experiments show that TAVR benefits from flexible inference-time video referencing and consistently surpasses existing baselines both quantitatively and qualitatively. This work has been deployed to production. For more related research, please visit \href{https://www.heygen.com/research}{HeyGen Research} and \href{https://www.heygen.com/research/avatar-v-model}{HeyGen Avatar-V}.

A critical bottleneck hindering further advancement in embodied AI and robotics is the challenge of scaling robot data. To address this, the field of learning robot manipulation skills from human video data has attracted rapidly growing attention in recent years, driven by the abundance of human activity videos and advances in computer vision. This line of research promises to enable robots to acquire skills passively from the vast and readily available resource of human demonstrations, substantially favoring scalable learning for generalist robotic systems. Therefore, we present this survey to provide a comprehensive and up-to-date review of human-video-based learning techniques in robotics, focusing on both human-robot skill transfer and data foundations. We first review the policy learning foundations in robotics, and then describe the fundamental interfaces to incorporate human videos. Subsequently, we introduce a hierarchical taxonomy of transferring human videos to robot skills, covering task-, observation-, and action-oriented pathways, along with a cross-family analysis of their couplings with different data configurations and learning paradigms. In addition, we investigate the data foundations including widely-used human video datasets and video generation schemes, and provide large-scale statistical trends in dataset development and utilization. Ultimately, we emphasize the challenges and limitations intrinsic to this field, and delineate potential avenues for future research. The paper list of our survey is available at https://github.com/IRMVLab/awesome-robot-learning-from-human-videos.

This paper proposes an algorithm for real-time learning without explicit feedback. The algorithm combines the ideas of semi-supervised learning on graphs and online learning. In particular, it iteratively builds a graphical representation of its world and updates it with observed examples. Labeled examples constitute the initial bias of the algorithm and are provided offline, and a stream of unlabeled examples is collected online to update this bias. We motivate the algorithm, discuss how to implement it efficiently, prove a regret bound on the quality of its solutions, and apply it to the problem of real-time face recognition. Our recognizer runs in real time, and achieves superior precision and recall on 3 challenging video datasets.

The task of capturing and rendering 3D dynamic scenes from 2D images has become increasingly popular in recent years. However, most conventional cameras are bandwidth-limited to 30-60 FPS, restricting these methods to static or slowly evolving scenes. While overcoming bandwidth limitations is difficult for general scenes, recent years have seen a flurry of computational imaging methods that yield high-speed videos using conventional cameras for specific applications (e.g., motion capture and particle image velocimetry). However, most of these methods require modifications to a camera's optics or the addition of mechanically moving components, limiting them to a single-view high-speed capture. Consequently, these methods cannot be readily used to capture a 3D representation of rapid scene motion. In this paper, we propose a novel method to capture and reconstruct a volumetric representation of a high-speed scene using only unaugmented low-speed cameras. Instead of modifying the hardware or optics of each individual camera, we encode high-speed scene dynamics by illuminating the scene with a rapid, sequential color-coded sequence. This results in simultaneous multi-view capture of the scene, where high-speed temporal information is encoded in the spatial intensity and color variations of the captured images. To construct a high-speed volumetric representation of the dynamic scene, we develop a novel dynamic Gaussian Splatting-based approach that decodes the temporal information from the images. We evaluate our approach on simulated scenes and real-world experiments using a multi-camera imaging setup, showing first-of-a-kind high-speed volumetric scene reconstructions.

Recent advances in self-evolution video understanding frameworks have demonstrated the potential of autonomous learning without human annotations. However, existing methods often suffer from weakly controlled optimization and uncontrolled difficulty progression, as they lack structured guidance throughout the iterative learning process. To address these limitations, we propose CurEvo, a curriculum-guided self-evolution framework that introduces curriculum learning into self-evolution to achieve more structured and progressive model improvement. CurEvo dynamically regulates task difficulty, refines evaluation criteria, and balances data diversity according to model competence, forming a curriculum-guided feedback loop that aligns learning complexity with model capability. Built upon this principle, we develop a multi-dimensional adaptive QA framework that jointly evolves question generation and answer evaluation across perception, recognition, and understanding dimensions, ensuring coherent and measurable curriculum progression. Through this integration, CurEvo transforms weakly controlled self-evolution into a more structured learning process for autonomous video understanding. Across seven backbones, CurEvo consistently improves both benchmark accuracy and evaluator-based semantic score on four VideoQA benchmarks, validating the effectiveness of curriculum-guided self-evolution for video understanding.