DataVista: Diagnosing Multimodal LLMs on Data Video Understanding
DataVista finds Gemini-3.1-Pro best of 19 MLLMs at 70% on data-video understanding.
DataVista is a benchmark for understanding data videos that combine animated charts with narrative. It contains 961 real-world videos and 6,775 questions across data perception, temporal reasoning, and narrative understanding, with 10 question types and five topic domains. Among 19 mainstream multimodal LLMs, Gemini-3.1-Pro scores 70.0% overall, still far below human experts, and models are weakest on Causal Reasoning and Narrative Structure. Extra frames and subtitles help perception and temporal reasoning more than narrative understanding.
- DataVista has 961 real data videos and 6,775 questions.
- Capabilities span data perception, temporal reasoning, and narrative understanding.
- Gemini-3.1-Pro leads 19 MLLMs at 70.0% accuracy, well below human experts.
- Models are weakest on Causal Reasoning and Narrative Structure.
- Vendors
- Products
- DataVista
- Organizations
- AI models
- Gemini-3.1-Pro
From the original article
1 min read at huggingface.co
Data video is a media form that integrates data visualization with video narrative, widely adopted in news reporting and business analysis. Compared with general video understanding, data video understanding places greater emphasis on accurately reading data from animated charts, integrating evidence across charts and time, and understanding how narrative organization and visual design communicate information.
By Yupeng Xie, Zhenyang Wang, Jiayi Zhu · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.