Research map / Video / Video-language understanding
ExpVid: A Benchmark for Experiment Video Understanding & Reasoning
Yicheng Xu, Yue Wu, Jiashuo Yu, Ziang Yan, Tianxiang Jiang, Yinan He, Qingsong Zhao, Kai Chen, Yu Qiao, Limin Wang, Manabu Okumura, Yi Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase open-source model scale (InternVL3-8B to InternVL3.5-38B to InternVL3-78B) (ablation) | InternVL family · ExpVid Level-1/2/3 | average accuracy: L1 39.4 -> 44.0 -> 50.9; L2 23.9 -> 36.0 -> 41.9; L3 27.2 -> 31.9 -> 37.7 | Improved |
| enable thinking mode (ablation) | Gemini-2.5-Flash · ExpVid Level-1/2/3 | average accuracy: L1 58.6 -> 60.2; L2 50.1 -> 49.8; L3 44.1 -> 43.0 | Mixed |
| enable thinking mode (ablation) | Gemini-2.5-Pro · ExpVid Level-1/2/3 | average accuracy: L1 59.2 -> 56.7; L2 53.8 -> 54.3; L3 47.9 -> 47.4 | Mixed |
| enable thinking mode (ablation) | MiMo-VL-7B-RL · ExpVid Level-1/2/3 | average accuracy: L1 42.4 -> 44.3; L2 27.4 -> 34.3; L3 27.3 -> 28.3 | Improved |
| provide video frames as input (ablation) | Gemini-2.5-Flash · ExpVid Level-1 and Level-2 tasks | accuracy | Improved |
| increase number of input frames up to 256 (ablation) | MiMo-VL and Kimi-VL · ExpVid Level-3 | accuracy | Improved |
| increase number of input frames beyond about 128 (ablation) | InternVL3.5 · ExpVid Level-3 | accuracy | Hurt |
Closest papers by meaning
- SciVideoBench: Benchmarking Scientific Video Reasoning in Large Multimodal Models (ICML 2026 · similarity 0.89)
- MicroVQA: A Multimodal Reasoning Benchmark for Microscopy-Based Scientific Research (CVPR 2025 · similarity 0.88)
- VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understanding (NeurIPS 2025 · similarity 0.87)
- Scientists' First Exam: Probing Cognitive Abilities of MLLM via Perception, Understanding, and Reasoning (NeurIPS 2025 · similarity 0.86)
- MME-SCI: A Comprehensive and Challenging Science Benchmark for Multimodal Large Language Models (AAAI 2026 · similarity 0.86)
- VideoReasonBench: Can MLLMs Perform Vision-Centric Complex Video Reasoning? (ICLR 2026 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
