STRAND is a new benchmark that tests multimodal large language models’ ability to track objects, their states, and relationships over time in videos. It evaluates intermediate reasoning by breaking queries into sub‑questions and uses Faithful Accuracy to ensure all parts of an answer are correct. The authors also propose an object‑centric framework that builds structured trajectories and shows reduced hallucinations and better temporal consistency compared to existing models.
arXiv:2605. 08974v2 Announce Type: replace-cross Abstract: While multimodal large language models (MLLMs) have advanced video understanding, they remain highly prone to hallucinations in dynamic scenes.
arXiv:2606. 02522v1 Announce Type: cross Abstract: Video multimodal large language models (MLLMs) have made rapid progress on general and long-form video understanding, yet their ability to preserve brief answer-critical visual evidence remains underexplored.
TimeBlind is a diagnostic benchmark designed to evaluate fine‑grained spatio‑temporal compositionality in video large language models (LLMs). It categorizes temporal understanding into three levels—atomic event recognition, event property characterization, and reasoning about event interdependencies—and uses a minimal‑pairs paradigm where video pairs share identical static content but differ only in temporal structure. Across 20 state‑of‑the‑art MLLMs tested on 600 curated instances, the best model achieved only 48.2% instance accuracy, far below human performance of 98.2%, highlighting a reliance on static visual shortcuts rather than true temporal reasoning.
SYNCR is a synthetic benchmark designed to evaluate multimodal large language models on cross‑video reasoning. It contains 4,000 question‑answer pairs across 4,827 unique videos, covering tasks in temporal alignment, spatial tracking, comparative reasoning, and holistic synthesis. The benchmark reveals a significant performance gap between current models and humans, with models excelling at temporal ordering but struggling with precise physical and spatial reasoning.
arXiv:2607. 02959v1 Announce Type: cross Abstract: We introduce VSeek, an agentic framework that transforms long-video question answering (LVQA) from a passive, single-pass perception task into a multi-turn retrieval process.
AgentVidBench is a new multi‑hop video question‑answering benchmark designed to evaluate spatial, temporal, and causal reasoning in multimodal large language models (MLLMs). Unlike existing tests that focus on simple scene queries or global summaries, AgentVidBench includes step‑by‑step solution traces to assess whether agents gather the necessary evidence to justify their answers. Experiments with 12 MLLMs show limited single‑turn performance, but integrating these models into agentic workflows improves both accuracy and trajectory scores, establishing AgentVidBench as a comprehensive testbed for future research on agentic video understanding.
arXiv:2606. 05702v1 Announce Type: new Abstract: Recent advancements in Vision-Language Models (VLMs) have significantly enhanced their ability to interpret complex visual semantics, yet their capacity for chronological reasoning remains under-explored.
arXiv:2505.01583v2 Announce Type: replace Abstract: Understanding causal event relationships and achieving fine-grained temporal grounding in videos remain challenging for vision-language models (VLM...
The paper introduces S‑MME, a benchmark for diagnosing shortcut‑biased behavior in Multimodal Large Language Models (MLLMs) on long‑video question answering. It proposes CLEAR, a training‑free framework that estimates and mitigates bias by treating the full video‑question pair as a complete view and intervening on hidden states during inference. Experiments on five benchmarks demonstrate that CLEAR improves prediction accuracy and robustness across different MLLMs.
arXiv:2604.01569v2 Announce Type: replace Abstract: Video multimodal large language models achieve strong results on existing benchmarks, but answer accuracy alone does not establish whether they can...
V-STORM introduces a structure-aware framework that converts videos into Spatio-Temporal Scene Graphs (STSGs) and generates Grounded Chain of Thought (GCoT) annotations, linking each reasoning step to explicit spatio-temporal evidence. The dataset contains 260K training samples and a 3K human-verified benchmark, both featuring GCoT annotations that trace reasoning to verifiable evidence. Experiments show that while current multimodal large language models can locate relevant spatio-temporal cues, they struggle to compose these cues into correct multi-step reasoning, and fine‑tuning with V-STORM improves both answer accuracy and grounding faithfulness.
VideoTIR introduces a reinforcement‑learning approach to improve long‑video understanding by encouraging multimodal large language models to use comprehensive multi‑level toolkits efficiently. It combines Zero‑RL and SFT cold‑starting strategies to help models retrieve and focus on meaningful video segments, images, and regions, thereby reducing hallucinations. The method includes Toolkit Action Grouped Policy Optimization (TAGPO) to streamline tool‑calling and a sandbox‑based trajectory synthesis framework for high‑quality data, achieving strong results on three long‑video QA benchmarks.