DataVista is a new benchmark for data video understanding, comprising 961 real‑world videos and 6,775 evaluation questions. It is organized into a three‑level capability framework—data perception, temporal reasoning, and narrative understanding—with 10 fine‑grained question types across five topic domains. Evaluation of 19 mainstream multimodal large language models shows that the best model, Gemini‑3.1‑Pro, scores 70.0% overall accuracy, lagging far behind human experts, especially on causal reasoning and narrative structure.
arXiv:2608.29711v1 Announce Type: new Abstract: While MLLMs have made significant strides in chart comprehension and video understanding, current evaluations largely isolate these capabilities, leavi...
While MLLMs have made significant strides in chart comprehension and video understanding, current evaluations largely isolate these capabilities, leaving a critical gap in understanding temporally evo...
arXiv:2606. 07433v1 Announce Type: cross Abstract: Video understanding is being rapidly transformed by multimodal large language models (MLLMs), as research moves from short clips to long, multimodal, and knowledge-intensive video scenarios.
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.
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.
arXiv:2601. 01095v3 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have achieved impressive progress in vision-language reasoning, yet their ability to understand temporally unfolding narratives in videos remains underexplored.
arXiv:2604. 25220v2 Announce Type: replace Abstract: Data videos combine animated visualizations with synchronized narration to communicate quantitative information and are widely used in journalism, education, and public communication.
The paper introduces REVEAL, a diagnostic benchmark that stresses Video‑Language Models (VidLMs) on five controlled probes—camera‑motion sensitivity, cross‑frame integration, video sycophancy, language‑only shortcuts, and temporal expectation bias—to assess how well these models encode and use visual evidence. Experiments on 12 VidLMs reveal systematic failures: some visual signals are never reliably encoded, while others are overridden by model priors, leading to performance below chance on several probes that humans solve with high accuracy. Mechanistic probes further pinpoint where and why visual evidence is lost, demonstrating that under assertive prompts a model’s output becomes nearly invariant to real versus random video input, rendering visual evidence causally inert.
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...
The paper introduces Short‑Films 20K (SF20K), a large publicly available movie dataset comprising 20,143 amateur films totaling 3,582 hours, with an average length of 12 minutes per film. Accompanying the dataset is SF20K‑Test, a manual open‑ended question‑answering benchmark featuring 95 movies and 979 question‑answer pairs. Analysis of the benchmark shows limited data leakage, highlights the necessity of long‑term reasoning, and demonstrates that instruction tuning on the large‑scale dataset significantly boosts vision‑language model performance.
arXiv:2608.05592v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) have made strong progress in video understanding, yet long videos remain difficult: the visual token budge...
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.