Temporal Video Understanding

Spatio-temporal reasoning is a core capability for Multimodal Large Language Models (MLLMs) operating in the real world. As such, evaluating it precisely has become an essential challenge. However, existing spatio-temporal reasoning benchmark datasets primarily rely on static image sets or passively curated video data, which limits the evaluation of fine-grained reasoning capabilities. In this paper, we introduce VGenST-Bench, a video benchmark that employs generative models to actively synthesize highly controlled and diverse evaluation scenarios. To construct VGenST-Bench, we propose a multi-agent pipeline incorporating a human quality control stage, ensuring the quality of all generated videos and QA pairs. We establish a comprehensive 3x2x2 video taxonomy, encompassing Spatial Scale, Perspective, and Scene Dynamics to span diverse scenarios. Furthermore, we design a hierarchical task suite that decouples low-level visual perception from high-level spatio-temporal reasoning. By shifting the paradigm from passive curation to active synthesis, VGenST-Bench enables fine-grained diagnosis of spatio-temporal understanding in MLLMs.

Video Capsule Endoscopy (VCE) poses a challenging multi-label temporal classification problem, requiring simultaneous localization of 8 anatomical regions and detection of 9 pathological findings across tens of thousands of frames. We present GALAR-TemporalNet v2, a hierarchical temporal model that addresses three core challenges: extreme class imbalance, long-range temporal dependencies, and pathology--anatomy entanglement. Our architecture combines windowed self-attention for local modeling, a Dual-Graph GCN for global frame relationships, and Bidirectional Mamba for selective boundary context encoding. A novel anatomy prototype residual pathway decouples pathological deviation signals from normal organ appearance, and a frame-level GCN skip connection stabilizes training of visually confusable rare classes. The competition version, GALAR-TemporalNet, achieved an overall [email protected] of 0.2644 and [email protected] of 0.2353 on the RARE-VISION test set. Following the competition, the redesigned GALAR-TemporalNet v2 -- incorporating a restructured pathology branch, refined loss functions, and extended post-processing -- improved these results to [email protected] of 0.3409 and [email protected] of 0.3333.

Zero-shot temporal action localization (ZS-TAL) consists of classifying and localizing actions in untrimmed videos, where action classes are unseen at training time. Existing work uses Vision and Language Models (VLMs), taking advantage of their strong zero-shot transfer capabilities. Yet, these models face evident challenges with fine-grained action classification, making it difficult to directly use them to distinguish between the presence and absence of an action. Most current methods for ZS-TAL address these challenges by training models on large-scale video datasets, which require annotated data and often result in limited generalization performance. Recently, approaches discarding the use of labeled data have emerged as an alternative. Following this direction, we propose a novel approach, ``Textual Guidance for finer localization of actions in videos'' (TEGU), that compensates for the lack of supervision from training data by exploiting rich textual information derived from large language models and structured text extracted from captions. This additional linguistic context can improve fine-grained discrimination by providing richer cues about fine-grained action differences within videos. We validate the effectiveness of the proposed method by conducting experiments on the THUMOS14 and the ActivityNet-v1.3 datasets. Our results show that, by exploiting rich textual information for improved action localization, TEGU outperforms state-of-the-art ZS-TAL approaches that do not involve training

Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in general visual understanding. However, their application to safety-critical driving scenarios remains limited by an inability to accurately perceive and reason about rare high-stakes dynamic events, such as collisions or near-collisions. To address this, we introduce a pipeline that enhances MLLM perception by fusing downsampled video frames with synchronized high-frequency telematics data (IMU and GPS) and semantic insights from specialized computer vision models. Our pipeline generates high-quality pseudo-labels, including descriptive captions and question-answer pairs, specifically designed to train MLLMs to identify and describe Safety-Critical Events (SCEs) in real-world driving footage. We show the effectiveness of our approach fine-tuning the open-source QwenVL-2.5 model via DoRA adapters: our experiments demonstrate significant improvements in identifying and explaining safety-critical events, with fewer than 50M trainable parameters and limited computational budget.

Multimodal Large Language Models (MLLMs) face significant computational overhead when processing long videos due to the massive number of visual tokens required. To improve efficiency, existing methods primarily reduce redundancy by pruning or merging tokens based on importance or similarity. However, these approaches largely overlook a critical dimension of video content, i.e., changes and turning points, and they lack a collaborative model for spatio-temporal relationships. To address this, we propose a new perspective: similarity is for identifying redundancy, while difference is for capturing key events. Based on this, we designed a training-free framework named ST-SimDiff. We first construct a spatio-temporal graph from the visual tokens to uniformly model their complex associations. Subsequently, we employ a parallel dual-selection strategy: 1) similarity-based selection uses community detection to retain representative tokens, compressing static information; 2) temporal difference-based selection precisely locates content-changing points to preserve tokens that capture key dynamic shifts. This allows it to preserve both static and dynamic content with a minimal number of tokens. Extensive experiments show our method significantly outperforms state-of-the-art approaches while substantially reducing computational costs. Our code is available in https://github.com/bingjunluo/ST-SimDiff.

Capsule endoscopy event detection is challenging because clinically relevant findings are sparse, visually heterogeneous, and evaluated at the event level rather than by frame accuracy. We propose VISTA, a metric-aligned multi-backbone framework for the RAREVISION task. VISTA combines EndoFM-LV for temporal context and DINOv3 ViTL/16 for frame-level visual semantics, followed by a Diverse Head Ensemble (DHE), Validation-Guided Weighted Fusion (VGWF), and Anatomy-Aware Temporal Event Decoding (ATED). The original official submission achieved hidden-test temporal [email protected] of 0.3530 and [email protected] of 0.3235. After the competition, extending local threshold refinement with a global coarse search improved performance to 0.3726 [email protected] and 0.3431 [email protected], ranking Team ACVLab second in the post-competition evaluation.

Recent advances in Multimodal Large Language Models (MLLMs) have significantly advanced video understanding tasks, yet challenges remain in efficiently compressing visual tokens while preserving spatiotemporal interactions. Existing methods, such as LLaVA family, utilize simplistic pooling or interpolation techniques that overlook the intricate dynamics of visual tokens. To bridge this gap, we propose ST-GridPool, a novel training-free visual token enhancement method designed specifically for Video LLMs. Our approach integrates Pyramid Temporal Gridding (PTG), which captures multi-grained spatiotemporal interactions through hierarchical temporal gridding, and Norm-based Spatial Pooling (NSP), which preserves high-information visual regions by leveraging the correlation between token norms and semantic richness. Extensive experiments on various benchmarks demonstrate that ST-GridPool consistently enhances performance of Video LLMs without requiring costly retraining. Our method offers an efficient and plug-and-play solution for improving visual token representations. Our code is available in https://github.com/bingjunluo/ST-GridPool.

Video large language models (Video LLMs) can achieve strong video-QA accuracy without reliably tracking spatiotemporal dynamics. A model may answer a motion question from static cues, for example, and give the same prediction even after the underlying motion is reversed. Correctness-based reinforcement learning does not directly address this problem because it rewards the final answer without requiring the policy to respond to task-relevant video dynamics. We propose Counterfactual Relational Policy Optimization (CRPO), which explicitly trains Video LLMs to respond to controlled changes in the visual input. For each training example, CRPO constructs a counterfactual video, such as a horizontally flipped or temporally reversed version of the original, and jointly optimizes rollouts from both videos under a shared policy. Factual supervision anchors what the model should answer, while counterfactual supervision constrains when that answer should change: predictions should change when an intervention alters task-relevant dynamics and remain stable when the queried property is preserved. This coupling provides a direct behavioral learning signal without requiring ground-truth labels for transformed videos or annotated spatiotemporal reasoning traces, while discouraging indiscriminate answer changes. To evaluate this property, we introduce DyBench, a paired counterfactual benchmark with 3{,}014 videos and a strict pair-accuracy metric. Across paired spatiotemporal evaluations and standard video benchmarks, CRPO improves sensitivity to motion and temporal changes while improving performance on general video understanding. The gains also extend to segment reordering, a transformation never used during training, suggesting that CRPO learns sensitivity to video dynamics beyond the training interventions. The project website can be found at https://ddz16.github.io/crpo.github.io/ .

The emergence of Large Vision-Language Models (LVLMs) has significantly advanced video understanding capabilities. However, existing benchmarks focus predominantly on coarse-grained tasks such as action segmentation, classification, captioning, and retrieval. Furthermore, these benchmarks often rely on entities that can be easily identified verbally, like household objects, animals, human subjects, etc., limiting their applicability to complex, in-the-wild video scenarios. But, many applications such as furniture assembly, cooking, etc., require step-by-step fine-grained spatio-temporal understanding of the video, which is not sufficiently evaluated in current benchmarks. To address this gap, we introduce Flat-Pack Bench, a novel benchmark centered on furniture assembly tasks. Our benchmark evaluates LVLMs on nuanced tasks, including temporal ordering of assembly actions, temporal localization of assembly state, understanding part mating, and tracking, using multiple-choice questions paired with visual prompts highlighting relevant parts as references for fine-grained questions. Our experiments reveal that state-of-the-art LVLMs struggle significantly with fine-grained spatio-temporal reasoning, highlighting their limitations in effectively leveraging temporal information from videos, limited tracking ability, and understanding of spatial interactions like physical contact.

Vision-language models (VLMs) are increasingly being explored for video game quality assurance, especially gameplay glitch detection. Most existing evaluations, however, treat glitches as static visual anomalies, asking models to detect failures from a single frame. We argue that this framing misses a key distinction: some glitches are spatial and visible in an isolated frame, whereas others are temporal and become evident only through changes across ordered frames. A preliminary study confirms this gap, showing that temporal glitches are substantially harder for VLMs to detect than spatial ones. To enable systematic evaluation of this underexplored setting, we introduce TempGlitch, a controlled gameplay video benchmark for temporal glitch detection. TempGlitch covers five temporal glitch types with balanced per-category samples, together with paired glitch-free videos that enable reliable binary evaluation. We evaluate 12 proprietary and open-weight VLMs across multiple frame-sampling settings. Our results show that current VLMs remain near chance on TempGlitch, often collapsing into either overly conservative behavior that misses most glitches or overly sensitive behavior that flags clean videos as glitchy. Moreover, denser frame sampling and larger model size do not reliably resolve these failures. TempGlitch provides a focused testbed for temporal reasoning, robust gameplay understanding, and automated glitch detection with VLMs. Code and data are available at the project website.

Detecting what has changed in an environment is essential for long-term autonomy, yet most change detection settings assume fixed viewpoints, mild misalignment, or only a few changed objects. We introduce Video-based Scene Change Detection (VSCD), which predicts a pixel-wise change mask for each query frame, given a reference and a query RGB video of the same indoor space recorded at different times under unconstrained camera motion. The two videos are not temporally synchronized, and many object instances may appear or disappear. To study this setting, we build a large-scale benchmark with over 1.1 million frames annotated with pixel-accurate change masks, together with a real-world test set for evaluating transfer beyond simulation. We propose a query-centric multi-reference model that learns temporal matching implicitly from change-mask supervision, aligns candidate reference features to the query via local patch correspondence, and fuses per-candidate change features using frame-level and patch-level confidence before decoding a high-resolution mask once per frame. Our approach achieves state-of-the-art performance against strong image- and video-based baselines, and we validate its real-world impact by deploying it on a mobile robot for two downstream applications -- visual surveillance and object incremental learning.

While Multimodal Large Language Models (MLLMs) exhibit strong performance on standard video tasks, their ability to faithfully summarize and reason over complex narratives remains poorly evaluated. Existing summarization benchmarks fragment supervision across isolated granularities, such as keyframes, key shots, or disjointed text summaries, failing to capture the inherently hierarchical structure of cross-modal alignment. To address this critical gap, we introduce HAVEN, a hierarchically aligned multimodal benchmark for unified video understanding. HAVEN pioneers a fully granular (frame, shot, and video levels) and fully multimodal (video and text) dataset architecture, complete with explicit, continuous alignment between modalities. Built upon this unified annotation paradigm, we propose a comprehensive evaluation suite spanning summarization, temporal reasoning, multimodal grounding, and saliency ranking. Extensive benchmarking of state-of-the-art MLLMs exposes a persistent gap between surface-level textual fluency and grounded multimodal understanding. Ultimately, HAVEN advances the evaluation of multimodal systems beyond traditional QA formats, offering a rigorous, standardized testbed to drive future research in interpretable, hierarchical video understanding. We publicly release the dataset, benchmark suite, and evaluation protocols.

Video foundation models achieve strong performance across many video understanding tasks, but typically require large-scale pre-training on massive video datasets, resulting in substantial data and compute costs. In contrast, modern image foundation models already provide powerful spatial representations. This raises an important question: can competitive video models be built by reusing these spatial representations and pre-training only for temporal reasoning? We take initial steps toward exploring a lightweight training paradigm that freezes a pre-trained image foundation model and trains only a recurrent temporal module to process streaming video. By reusing an image foundation model as a spatial encoder, this approach could significantly reduce the amount of video data and compute required compared to end-to-end video pre-training. In this work, we explore the feasibility of this approach before investing in computing for video pre-training. Our empirical findings across multiple video understanding tasks suggest that strong temporal performance can emerge without large-scale video pre-training, motivating future work on recurrent video foundation models obtained by pre-training a temporal module on top of a frozen image foundation model. Code: https://github.com/tue-mps/towards-video-image-frozen .

Law enforcement agencies are accumulating vast amounts of body-worn camera (BWC) footage. However, this remains operationally opaque. That is, analysts and trainers still have to invest considerable time watching full-length videos to pinpoint the start of key encounters and identify the points where activity shifts to something more physically intense. We present an approach to process BWC video into a time-aligned sequence of fixed-length 10-second windows, processed and labeled using a privacy-conscious protocol. Each window is labeled with two dimensions of information: (i) the operational context of the window and (ii) the level of motion intensity within the window, with low-evidence labels for windows for which insufficient evidence exists due to darkness, blur or occlusion. We train models to classify windows based on these two axes using frames sampled from each window encoded using CLIP model and aggregated into a window-level representation. We extract dense optical flow statistics for each window to capture motion intensity. On test windows the best context model achieves 78.75% accuracy, and the best-accuracy activity model achieves 88.33%. We also included integrity audits to show the results and how the visual timeline representations support faster incident review and make the officer training workflow more practical.

Vision-language models (VLMs) have recently emerged as a promising paradigm for video anomaly detection (VAD) due to their strong visual reasoning ability and natural language-based explainability. In this paper, we aim to address a key limitation of such pipelines, which perform segment-level inference independently owing to token constraints and reason without structured temporal context, allowing VLMs to interpret anomalies as deviations from evolving video dynamics rather than producing fragmented predictions and explanations. To specify, we propose a context-aware framework named LATERN, which reformulates VAD as a temporal evidence aggregation process. LATERN consists of two complementary modules: Context-Aware Anomaly Scoring (CEA) and Recursive Evidence Aggregation (REA). CEA introduces a novel image-grounded memory mechanism, which selectively chooses historical content via frame diversity and visual-textual alignment as expanded context to help generate reliable anomaly scores. Building upon these scores, REA performs recursive temporal aggregation to identify coherent anomaly intervals and produce event-level decisions and explanations grounded in visual-textual evidence. Extensive experiments on challenging benchmarks, including UCF-Crime and XD-Violence, show that LATERN enhances detection accuracy and explanation consistency for frozen VLMs during test time, while generating temporally coherent and semantically grounded event-level explanations.

Online surgical phase recognition (SPR) underpins context-aware operating-room systems and requires committing to a prediction at every frame from past context alone. Surgical video poses three demands that natural-video recognizers do not jointly address: procedures span tens of thousands of frames, time flows non-uniformly as long routine stretches are punctuated by brief phase-defining transitions, and the visual domain is narrow so backbone features are strongly correlated across channels. Existing recognizers either let per-frame cost grow with elapsed length, or hold cost bounded but advance state at a uniform rate with channel-independent dynamics, leaving the latter two demands unaddressed. We present SurgicalMamba, a causal SPR model built on Mamba2's structured state-space duality (SSD) that holds per-frame cost at O(d). It introduces three SSD-compatible components that jointly address these demands: a dual-path SSD block that separates long- and short-term regimes at the level of recurrent state; intensity-modulated stepping, a continuous-time time-warp that adapts the slow path's effective rate to phase-relevant information; and state regramming, a per-chunk Cayley rotation that opens cross-channel mixing in the otherwise axis-aligned SSM recurrence. The learned rotation planes inherit a phase-aligned structure without any direct supervision, offering an interpretable internal signature of surgical workflow. Across seven public SPR benchmarks, SurgicalMamba reaches state-of-the-art accuracy and phase-level Jaccard under strict online evaluation: 94.6%/82.7% on Cholec80 (+0.7 pp/+2.2 pp over the strongest prior) and 89.5%/68.9% on AutoLaparo (+1.7 pp/+2.0 pp), at 238.74 fps on a single GPU. Ablations isolate the contribution of each component. The code is publicly available at https://github.com/sukjuoh/Surgical-Mamba.

Self-evolution offers a promising path for improving reasoning models without relying on intensive human annotation. However, extending this paradigm to video understanding remains underexplored and challenging: videos are long, dynamic, and redundant, while the evidence needed for reasoning is often sparse and temporally localized. Naively generating difficult question-answer pairs from full videos can therefore produce supervision that appears challenging but is weakly grounded, relying on static cues or language priors rather than temporal evidence. In this work, we argue that the key bottleneck of video self-evolution is not difficulty alone, but grounding. We propose Video-Zero, an annotation-free Questioner--Solver co-evolution framework that centers self-evolution on temporally localized evidence. The Questioner discovers informative evidence segments and generates evidence-grounded questions, while the Solver learns to answer and align its predictions with the supporting evidence. This closes an iterative loop of evidence discovery, grounded supervision, and evidence-aligned learning. Across 13 benchmarks spanning temporal grounding, long-video understanding, and video reasoning, Video-Zero consistently improves multiple video VLM backbones, demonstrating the effectiveness and transferability of evidence-centered self-evolution.

Scene understanding is central to general physical intelligence, and video is a primary modality for capturing both state and temporal dynamics of a scene. Yet understanding physical processes remains difficult, as models must combine object localization, hand-object interactions, relational parsing, temporal reasoning, and step-level procedural inference. Existing benchmarks usually evaluate these capabilities separately, limiting diagnosis of why models fail on procedural tasks. We introduce BARISTA, a densely annotated egocentric dataset and benchmark of 185 real-world coffee-preparation videos covering fully automatic, portafilter-based, and capsule-based workflows. BARISTA provides verified per-frame scene graphs linking persistent object identities to masks, tracks, boxes, attributes, typed relations, hand-object interactions, activities, and process steps. From these graphs, we derive zero-shot language-based tasks spanning phrase grounding, hand-object interaction recognition, referring, activity recognition, relation extraction, and temporal visual question answering. Experiments reveal strong variation across task families and no consistently dominant model family, positioning BARISTA as a challenging diagnostic benchmark for procedural video understanding. Code and dataset available at https://huggingface.co/datasets/ramblr/BARISTA.

Online Surgical Phase Recognition (SPR) models can reach high frame-wise accuracy, yet their predictions often lack temporal stability, fragmenting workflow understanding and reducing the reliability of downstream assistance. We show that this instability is not random noise but arises from two mechanisms: early misclassifications corrupt temporal feature states and propagate forward to form error cascades, and phase transitions follow evidence-accumulation dynamics whereas most online SPR systems rely on memoryless frame-wise decisions, making them sensitive to transient confidence fluctuations. We propose a unified Train-Inference-Evaluation framework that explicitly stabilizes temporal inference dynamics using model-agnostic, plug-and-play components. For training, the Temporal Error-Cascade (TEC) loss suppresses error onset and mitigates forward error propagation by stabilizing temporal feature evolution. For inference, the Evidence-Gated Transition Predictor (EGTP) enforces evidence-driven state transitions, allowing phase changes only when accumulated evidence exceeds a confidence boundary. For evaluation, we introduce the Temporal Fragmentation Index (TFI), a reliability-aware metric that quantifies instability-induced temporal disagreement beyond conventional frame-wise and token-based measures. Experiments on Cholec80 and AutoLaparo across three representative backbones show that the proposed framework substantially improves temporal stability and reduces prediction fragmentation, while maintaining or modestly improving frame-wise performance.

Proactive streaming video understanding requires models to continuously process video streams and decide when to respond, rather than merely what to respond. This naturally introduces a decision-making problem under partial observations, where models must balance early prediction against sufficient evidence. However, existing benchmarks largely follow a "see-then-answer" paradigm, where responses are triggered only after explicit evidence appears, effectively reducing proactive reasoning to delayed perception. As a result, they fail to evaluate a model's ability to make timely and reliable decisions under incomplete observations. Moreover, training proactive models is inherently challenging due to the extreme imbalance between silence and response signals in streaming trajectories, as well as the need to jointly optimize response correctness and timing. To address these challenges, we introduce StreamPro-Bench, a new benchmark that evaluates streaming models from three complementary perspectives: Perception Understanding, Temporal Reasoning, and Proactive Agency, where the last measures a model's ability to make early yet reliable decisions under partial observations. We further propose StreamPro, a two-stage training framework for proactive learning. First, we introduce CB-Stream Loss to mitigate the severe supervision imbalance during supervised fine-tuning (SFT). Then, we apply Group Relative Policy Optimization (GRPO) with a multi-grained reward design that involves both turn-level and trajectory-level rewards. Experiments show that StreamPro significantly improves proactive performance. On StreamPro-Bench, it achieves 41.5, substantially outperforming the previous best (10.4), while also maintaining strong performance on real-time streaming benchmarks, achieving 78.9 on StreamingBench-RTVU.

Online Temporal Action Localization (On-TAL) aims to detect the occurrence time and category of actions in untrimmed streaming videos immediately upon their completion. Recent advancements in this field focus on developing more sophisticated frameworks, shifting from Online Action Detection (OAD)-based aggregation paradigm to instance-level understanding. However, existing approaches are typically trained on specific domains and often exhibit limited generalization capabilities when applied to arbitrary videos, particularly in the presence of previously unseen actions. In this paper, we introduce a new task called Online Zero-shot Temporal Action Localization (OZ-TAL), which aims to detect previously unseen actions in an online fashion. Furthermore, we propose a training-free framework that leverages off-the-shelf Vision-Language Models (VLMs) while introducing additional mechanisms to enhance visual representations and mitigate their inherent biases. We establish new benchmarks and representative baselines for OZ-TAL on THUMOS14 and ActivityNet-1.3, and extensive experiments demonstrate that our method substantially outperforms existing state-of-the-art approaches under both offline and online zero-shot settings.

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.

This article presents a novel approach to keyframe detection in ultrasound videos, with a particular focus on fetal brain imaging. The proposed model is a composite neural network architecture that combines a Convolutional Neural Network (CNN) with a Recurrent Neural Network (RNN). The CNN extracts spatial features from individual video frames, while the RNN captures temporal dependencies between consecutive frames within each video sequence. The proposed model may improve the efficiency and accuracy of fetal brain ultrasound analysis, thereby supporting earlier detection, diagnosis, and treatment planning for selected fetal brain conditions.

Vision-language models (VLMs) have recently shown strong potential in soccer video understanding. However, given the high complexity of soccer videos due to large viewpoint variations, rapid shot transitions, and cluttered scenes, it remains unclear on whether VLMs rely on meaningful visual evidence or exploit spurious correlations and shortcut learning. Existing evaluation protocols focus primarily on classification accuracy and do not assess visual grounding. To address this limitation, we introduce SoccerLens, a benchmark for grounded soccer video understanding. The benchmark contains annotated video segments spanning $13$ common soccer events, with structured visual cues organized into three levels of semantic relevance. We further extend the attribution method of Chefer [arXiv:2103.15679] to jointly model spatial and temporal attention, and introduce evaluation metrics that measure whether model attention aligns with annotated cues or drifts toward spurious regions. Our evaluation of state-of-the-art soccer VLMs shows that, despite strong classification accuracy, current models fail to exceed $50\%$ grounding performance even under the loosest cue definitions and consistently underutilize temporal information. These results reveal a substantial gap between predictive performance and true visual grounding, highlighting the need for grounded evaluation in complex spatio-temporal domains such as soccer.

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.

While multimodal large language models (MLLMs) have advanced video understanding, they remain highly prone to hallucinations in dynamic scenes. We argue this stems from a failure in spatio-temporal monitoring, the ability to persistently track object identities, states, and relations over time. Existing benchmarks obscure this deficit by relying on single final-answer evaluations for queries that can often be resolved via local visual cues or statistical priors. To rigorously diagnose this, we introduce STEMO-Bench (Spatio-TEmporal MOnitoring), a benchmark of human-verified object-centric facts that evaluates intermediate reasoning by decomposing queries into sub-questions, distinguishing genuine temporal understanding from coincidental correctness. To address failure modes exposed by STEMO, we propose STEMO-Track, a novel object-centric framework that explicitly constructs and reasons over structured object trajectories via chunk-wise state extraction and temporal aggregation. Extensive experiments demonstrate that our object-centric framework significantly reduces hallucinated answers and improves spatio-temporal reasoning consistency over state-of-the-art MLLMs.

Long-term multimodal action quality assessment (AQA) evaluates action execution in several-minute audiovisual sequences by mining discriminative quality cues for score prediction. Existing multimodal methods usually model entire sequences with a single temporal encoder and fuse modality features by direct alignment or concatenation, causing key cues to be obscured by global trends, weakened by modal redundancy, and distorted during one-shot score mapping. To address this issue, we reformulate long-term multimodal AQA as a quality cue organization problem and propose MLCR, a multi-level cue refinement framework. MLCR organizes quality evidence at three levels: intra-modal representation, cross-modal interaction, and stage-wise aggregation. Specifically, the intra-modal decoupling encoder (IMDE) preserves modality identity while refining global temporal context and local frequency details. The cross-modal dynamic complementarity-aware retrieval (CMDCR) module retrieves incremental evidence conditioned on the evolving fused state and suppresses redundant responses. The stage-wise multimodal integration (SMI) block progressively accumulates intra-modal and cross-modal cues to refine the fused representation. Experiments on the Rhythmic Gymnastics and Fis-V datasets show that MLCR achieves the best or second-best performance in both Spearman correlation and prediction error, demonstrating its effectiveness and robustness.

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.

The Arrow-of-Time (AoT) task, determining whether a video plays forward or backward by recognizing temporal irreversibility, is one humans solve with near-perfect accuracy, yet frontier Video Large Language Models (Video-LLMs) perform only modestly above chance. This gap raises a key question: do visual backbones fail to encode temporal information, or does information bottleneck lie elsewhere in the Video-LLM architecture? We address this question by isolating the vision encoder from the Video-LLM and tracing temporal information across the encoder, projector, and LLM. We find that video-centric encoders with explicit temporal modeling encode strong temporal signals, whereas frame-centric encoders do not. However, when video-centric representations are passed through a standard Video-LLM architecture, performance often collapses, revealing a bottleneck of temporal information flow. We identify projector design as a key factor: Q-Former disrupts temporal information, while a time-preserved MLP projection substantially improves the LLM's access to such information. Our layer-wise analysis further shows temporal representation dynamics across encoder layers. Guided by these findings, we build a Video-LLM with temporal-aware video-centric encoder, time-preserved projector, and AoT supervision, surpassing human performance on AoT$_{PPB}$ with 98.1% accuracy, and improving broader temporal reasoning tasks by up to 6.0 points on VITATECS-Direction and 1.3 points on TVBench. Our results show that temporal reasoning in Video-LLMs requires both effective temporal encoding and reliable transfer of this information to the LLM.

Human Motion Segmentation (HMS), which aims to partition a video into non-overlapping segments corresponding to different human motions, has recently attracted increasing research attention. Existing HMS approaches are predominantly based on subspace clustering, which are grounded on the assumption that the distribution of high-dimensional temporal features well aligns with a Union-of-Subspaces (UoS). For videos in the real world, however, the raw frame-level features often violate the UoS assumption and yield unsatisfactory segmentation performance. To address this issue, we propose an efficient and effective approach for HMS, named Temporal Deep Self-expressive subspace Clustering (TDSC), which jointly learns temporally consistent structured representations and stabilized affinity for accurate and robust HMS. Specifically, in TDSC, we alternately learn structured representations of the input frame features and self-expressive coefficients via a properly regularized self-expressive model, in which a coding-rate maximization regularizer is incorporated to avoid representation collapse and conform the learned representations to span a desired UoS distribution, and meanwhile, temporal constraints are incorporated to promote temporally adjacent frames to be partitioned into the same groups. Moreover, we develop a temporal momentum averaging mechanism to stabilize affinity evolution and design a reparameterization strategy to enable efficient optimization. We conduct extensive experiments on five benchmark HMS datasets using both conventional (HoG) and up-to-date deep features (i.e., CLIP, DINOv2) to validate the effectiveness of our approach.