Papers for
video content analysts
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
Vision language models gain clearer concept level understanding
A General Concept-based Decomposition for Vision-Language Embeddings
Abstract: Vision-Language Models are designed to capture the compositional structure of the world through multi-modal alignment; however, it remains unclear whether their representations truly align with the way a human would naturally describe the same visual input. Prior works attempt to decompose visual embeddings into their concept-level contributions, but their instance-level optimization is inherently local and does not capture the model's global semantic structure, which is essential for compositional generalization across different inputs, tasks, and domains. We introduce GRACE, a method that learns this concept-based structure at the feature space level, producing semantic and general concept decompositions. To achieve this, we propose a novel training objective that preserves the geometric relations of different samples in the concept space, enabling a better understanding of how concepts compose and relate to each other. We validate GRACE on several image and video benchmarks, showing that its concept decomposition is aligned with human perception and achieves faithful decompositions across different tasks and domains. With minimal overhead, GRACE makes existing Vision-Language Models more interpretable, while preserving most of their original performance.
Human centered graphs improve video question answering accuracy
THG-Omni: Traceable Human-Centric Graph Supervision for Video Reasoning
Abstract: Omni-modal models can answer questions about video, but their predictions rarely show which participant, event, or interval supports an answer. Moreover, answer-level supervision alone does not teach a model to interpret structured video evidence. To systematically mitigate this issue, we present THG-Omni, a human-centric graph-supervision framework that links Segment Anything Model~3 participant tracks with temporally grounded Qwen3-VL annotations. Candidate semantic claims are checked against the source video and available audio before they are used to construct training questions, while track geometry provides explicit spatial and temporal supervision. The resulting questions teach an omni-modal reasoning model to interpret participant attributes, actions, affect, interactions, and their timing from a structured graph. At inference, THG-Omni uses a predicted graph constructed from the evaluation video. Experiments on IntentBench show consistent improvements over the unadapted backbone across social reasoning, emotion understanding, and deception perception. On NExT-QA, THG-Omni achieves 89.54\% accuracy with the matched graph, compared with 85.10\% without graph input. Further validation experiments show that compact graph input achieves 78.22\% accuracy, compared with 78.02\% using the full graph and 77.58\% without graph input. A complementary diagnostic demonstrates that the predicted graph preserves evidence useful for answering questions even without the source video and audio. Together, these findings show that THG-Omni improves human-centric video reasoning while grounding its evidence in identifiable participants and events. Our codes are available at https://anonymous.4open.science/r/review-n94hhfcwp5666/
DeformStruct improves action timing with deformable structure modeling
DeformStruct: Deformable Structure Modeling for Point-Supervised Temporal Action Localization
Abstract: Point-supervised Temporal Action Localization (PTAL) aims to localize action instances using only a single annotated timestamp for each instance. Such sparse supervision offers limited guidance for modeling the structural organization of an action instance. Existing PTAL methods primarily infer action regions from classification responses, leaving the structural organization of multiple action-related parts within each instance insufficiently modeled. To address this problem, we propose DeformStruct, an instance-level deformable structure modeling framework for PTAL. DeformStruct progressively constructs the structure of each action instance through three stages. First, Structural Anchor Mining discovers multiple anchors from annotated points and action-responsive regions, providing structural nodes grounded in the current instance. Second, Deformable Part Alignment associates latent structural parts with these anchors through deformation-aware matching, allowing their temporal supports to adapt to instance-dependent variations. Third, Relational Structural Organization explicitly models the relative temporal relations among the aligned parts, forming a relational representation that characterizes how different action-related parts are organized together. Extensive experiments on THUMOS14, ActivityNet-v1.3, GTEA, and BEOID demonstrate consistent improvements in point-supervised temporal localization and validate the effectiveness of deformable structural modeling.
Video based research agents create interactive reports linking moments
Beyond Watching and Searching: Benchmarking Agents for Video-centered Deep Research Reports
Abstract: Deep research agents synthesize open-web evidence into long reports, but current benchmarks do not evaluate research centered on a source video. We introduce Video-Centered Deep Research Report Generation. Given a video and user request, an agent identifies research questions, retrieves external multimodal evidence, and produces an interactive report linked to relevant video moments. This task has direct practical value because it helps users locate relevant moments, verify claims, reproduce demonstrated procedures, and make informed decisions from video content. We present a skill-based workflow and OmniDR-Bench for this task. For evaluation, we propose Omni Atomic Rubric Trees (Omni-ARTs). Each tree contains three dimensions, 23 shared directions, and task-specific atomic binary criteria. The directions test audiovisual grounding, multimodal evidence use, and interactive presentation. Audited Rubric Construction (ARC) builds these trees through independent evidence collection, automated checks, behavioral calibration, and iterative human review. OmniDR-Bench contains 50 Chinese and English tasks across 19 realworld subcategories. It includes 6,414 criteria. Across 18 agent configurations, the best system scores 63.3 out of 100. Current agents still need broader content coverage, stronger causal and logical analysis across modalities, and more accurate citation support. We further analyze models, harnesses, skill configurations, video categories, and research trajectories. The code and data are released at https://github.com/marinero4972/OmniDR-Bench.
Procedural mistake recognition improves by linking video to language
Through a PRISM, Clearly: Step-Sequence Reasoning for Mistake Recognition in Procedural Videos
Abstract: Procedural mistake recognition requires identifying deviations from valid multi-step activities. We study post-completion recognition from videos with benchmark-provided step boundaries. We introduce PRISM (Procedural mistake Recognition via Interpretable Semantic Modeling), which grounds video segments in language and maps their descriptions to procedural states using a lightweight hidden Markov model trained on correct executions. Reference alignment and semantic checks identify errors in step content and order. To support evaluation across heterogeneous benchmarks, we map dataset-specific annotations to four edit-based error types: insertion, deletion, substitution, and transposition. Across three benchmarks, we compare gold and VLM-generated descriptions in raw, semantic-role, and action-object forms. Decoder controls show consistent gains from the trained local decoder over nearest-candidate matching, with additional, dataset-dependent benefits from sequence decoding. Semantic role labeling improves Any-F1 in all six matched grounded comparisons, although the gains vary across datasets and do not translate into uniform improvements under separate configuration selection. These results establish language grounding as an effective interface for explicit procedural mistake reasoning and characterize the contributions of representation and decoding.
Grounded descriptions improve recognizing everyone in multi-person videos
Grounded Human-Attributed Description and Activity Recognition in Videos
Abstract: Understanding characteristics and activities of individuals in complex multi-person environments is crucial for real-world applications. However, existing datasets often simplify this problem to grounded group activity recognition, single-person activity recognition, or close-set activity recognition, thus limiting the generalizability of models trained on these datasets. In this work, we introduce the task of Grounded Human-Attributed Description and Activity Recognition (GHADAR), which involves describing the characteristics and activity descriptions of \textbf{every} person in a video, provided the location of the person in the video, which is more practical for real-world applications. To facilitate this, we introduce a new dataset derived from AVA-Actions by generating open-set captions for the person description and activity. In addition, we propose a novel method to effectively utilize the information contained in grounding during training by constraining the cross-attention masks during training in VLMs to improve performance for this task. Our experiments show that our method outperforms SOTA VLMs on this task. Finally, we demonstrate the limitations of existing evaluation metrics, which are overly reliant on human-annotations and exact text-text matching. As an added video-based evaluation, we propose a holistic VLM-based evaluation schema that compares concepts {\em directly} between the video and the generated predictions. Thus, in this work, we develop a complete framework for GHADAR, including a dataset, a novel method and an evaluation schema, thereby establishing a strong foundation for future research in this domain.
Adaptive frame filtering reduces video analysis cost and boosts accuracy
AdaFrame: Hierarchical Multimodal Deduplication with Adaptive Information Budgeting for Cost-Efficient Video Advertisement Analysis
Abstract: Uniform frame sampling for vision-language model (VLM)-based video analysis often floods multimodal backbones with redundant content, while existing keyframe selection methods are typically optimized for human viewing rather than downstream extraction accuracy. We propose AdaFrame, a content-adaptive preprocessing cascade that reduces visual redundancy without sacrificing extraction fidelity. Three progressively finer filters, perceptual hashing, learned perceptual image patch similarity (LPIPS), and CLIP-based semantic clustering, efficiently eliminate redundant frames before costly cross-modal inference. Per-video frame budgets are determined using Intrinsic Semantic Dimensionality (ISD), which estimates each video's semantic complexity through singular value decomposition of its frame embeddings. Experiments on 500 video advertisements from the Pitt Ads Dataset show that AdaFrame reduces the frame count by 48 percent and VLM inference cost by 75 percent compared with uniform sampling, while achieving 86.0 percent topic accuracy, an improvement of 1.6 percentage points, and 79.4 percent sentiment accuracy, an improvement of 7.5 percentage points.
LongEmo improves emotion understanding in extended videos
LongEmo: Towards Emotion Understanding and Reasoning in Long Videos
Abstract: While recent Multimodal Large Language Models (MLLMs) have shown promise in affective computing, their reasoning capabilities are largely confined to short video clips with limited interactions. However, real-world emotions are not merely isolated instantaneous reactions but dynamic and cumulative processes deeply shaped by past experiences and ongoing events. To bridge this gap, we introduce LongEmoBench, a benchmark dedicated to emotion understanding and reasoning in long videos. It assesses progressive capabilities scaling from continuous scene interactions to complex episodic developments. Furthermore, we propose LongEmo, a novel memory-augmented agentic framework designed to tackle the immense challenges of long-range affective reasoning. LongEmo processes continuous video streams to construct an Event Memory Graph, explicitly modeling long-range dependencies and capturing emotional dynamics across discrete events. Given a question, the agent retrieves a query-relevant event stream from the graph, iteratively integrating multimodal memories and relational dependencies to deduce the final answer. Extensive evaluations of 17 representative methods reveal that they struggle significantly with emotion understanding and reasoning in long videos. In contrast, LongEmo achieves state-of-the-art performance, demonstrating the efficacy of its event-centric memory architecture.
Vision language models improve limb motion captions with graded error scoring
Beyond Binary Preferences: Graded Preference Optimization for Limb-Motion Captioning
Abstract: Vision-Language Models (VLMs) can generate rich video captions, yet often misidentify which person performs an action or which limb is involved, particularly across camera cuts. Improving these details requires evaluation and training that distinguish missing information from incorrect assertions. We introduce FlexBench, a benchmark spanning 3,105 shots and 18,161 evaluation queries, with human-verified identities and systematic per-person coverage of fine-grained limb actions and states. Its reference-derived checklists support automated assessment of complete captions in their person and shot contexts. Our Graded Physical Alignment score (GPA) awards credit for correct content and deducts points for incorrect or fabricated actions, making these errors explicit in the aggregate score. Building on this rubric, we propose Graded Margin Direct Preference Optimization (GM-DPO), which assigns stronger preference margins and greater training weight to more severe action errors. Across three VLM backbones, GM-DPO achieves the highest substantive-action and GPA scores among the evaluated preference objectives, improving GPA over DPO by 2.02-3.40 points. On Qwen3-8B, it reduces the weighted hallucination rate by 21.3% relative to DPO. These gains accompany sustained long-form output, improved shot structure, and competitive performance on three additional multimodal benchmarks.
Query-guided summarization improves locating events in long videos
Summarize Before Grounding: Query-Guided Chunk Condensation for Long-Video Temporal Grounding
Abstract: Video temporal grounding (VTG) aims to localize the video interval corresponding to a language query. Recent large vision-language models (LVLMs) show great potential in solving such a multi-modal reasoning task. However, long videos often contain large amounts of redundant information that disturbs LVLMs to mine query-relevant evidence. Instead of dense frame sampling which incurs prohibitive training memory, previous reinforcement learning with verifiable rewards (RLVR) works typically utilize sparse sampling, which makes training feasible but may miss critical evidence. In this paper, we propose a ``summarize before grounding'' framework (named ``SumGround'') for long-video temporal grounding. The key of SumGround is to perform query-guided chunk condensation to aggregate and retrieve query-relevant evidence. Specifically, we split the video into several chunks and perform two-level chunk condensation. First, we introduce query-guided latent summaries, which is represented as KV states of query-guided prompts, to compress redundant visual tokens into compact query-relevant chunk summaries. Furthermore, we design an associative summary retrieval scheme to rank and select chunk summaries that are most likely to contain the event interval. Both query-guided latent summary and associative summary retrieval schemes are enabled by RLVR. To reduce memory consumption, we propose a length-aware gradient gating module to selectively stop gradient back-propagated to visual tokens. Extensive experiments demonstrate that SumGround performs favorably against previous state-of-the-art methods across multiple downstream datasets, with remarkable gains on long videos.
Livestream videos get smarter at matching products with show moments
Grounded Product Understanding in Livestream Videos
Abstract: E-commerce livestreams have emerged as an important channel for presenting products to online consumers, containing multiple products whose information is scattered in different moments. This poses significant challenges for downstream product understanding applications, such as product-centric livestream clipping, where models need to identify the product and its relevant segments for information gathering. However, existing benchmarks for general product understanding typically evaluate product retrieval and temporal localization in isolation, leaving the critical correspondence between product identity and temporal evidence largely unassessed. To address this limitation, we introduce GPUB, a large-scale benchmark comprising 3,000 livestream instances with quality-controlled multi-moment temporal annotations and a catalog of over 31K fashion products. GPUB supports three evaluation tasks: the main task Grounded Product Understanding (GPrU) requires jointly identifying the target product and localizing its supporting moments from a livestream video and a candidate product set; Product Retrieval and Product Moment Localization serve as two complementary subtasks. Evaluation of existing multimodal models shows that GPrU remains highly challenging, with the best-performing baseline achieving only 10.13% Pair mAP@.3. To narrow the performance gap, we further develop UniPro, a unified product understanding model that derives product-aligned and temporally structured representations from shared multimodal encoding, improving Pair mAP@.3 to 21.53% while achieving 37.23% Joint R@1@.3 on GPrU.
Video sentiment analysis improves by splitting polarity and intensity tasks
Divide and Conquer: Mixture-of-Bottleneck Experts in Informative Ordinal Space for Video-based Multimodal Sentiment Analysis
Abstract: Video-based Multimodal sentiment analysis (MSA) must handle information from text, audio, and image sequence in human speaking videos, yet current methods often fail to integrate modalities with task awareness. Most models treat video sentiment prediction as a single task, overlooking its ordinal nature, and their fusion strategies struggle to capture diverse unique and synergic cues across modalities. To address these limitations, we adopt a divide-and-conquer perspective by reformulating MSA as an ordinal regression problem and decoupling it into polarity recognition and intensity prediction. Driven by information theory, we introduce a Mixture-of-Bottleneck (MoB) framework that assigns different latents to polarity- and intensity-specific experts for different modalities. With the learning of information bottleneck, each expert learns compact and task-relevant representations while filtering out redundancy and noise. A multimodal bottleneck routing fusion module then fuses these expert latents with hard mining strategy, guiding the prediction in the ordinal sentiment space. Extensive experiments on 4 MSA datasets and 4 language models show that MoB effectively leverages informative latents from diverse modalities and captures general sentiment structure. Beyond stronger performance, MoB comprehensively captures fine-grained intra- and inter-modal dynamics, enabling more trustworthy localization of nuanced video sentiment signals.