Research map / Video / Video generation
UniVBench: Towards Unified Evaluation for Video Foundation Models
Jianhui Wei, Xiaotian Zhang, Yichen Li, Yuan Wang, Yan Zhang, Ziyi Chen, Zhihang Tang, Wei Xu, Zuozhu Liu
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 |
|---|---|---|---|
| use commercial model Gemini 2.5 Pro versus unified model Showo-2 for video understanding | V2T · UniVBench V2T | average score: 16.3% -> 54.1% | Improved |
| unified model Omni-Video for T2V versus specialist Seedance-1.0-Pro | T2V · UniVBench T2V | average score: 77.9% -> 56.2% | Hurt |
| scale Wan2.1-VACE from 1.5B to 14B | V2V reconstruction · UniVBench V2V | average score: 34.9% -> 62.7% | Improved |
| reconstruct from model-generated caption instead of ground-truth caption | Omni-Video · UniVBench | qualitative consistency | Hurt |
| action dimension versus others | video models · UniVBench | dimension score: Action 27.0% vs Color 65.8% (Gemini V2T) | Hurt |
Closest papers by meaning
- MMBench-Video: A Long-Form Multi-Shot Benchmark for Holistic Video Understanding (NeurIPS 2024 · similarity 0.90)
- Needle In A Video Haystack: A Scalable Synthetic Evaluator for Video MLLMs (ICLR 2025 · similarity 0.90)
- VBench: Comprehensive Benchmark Suite for Video Generative Models (CVPR 2024 · similarity 0.90)
- MVBench: A Comprehensive Multi-modal Video Understanding Benchmark (CVPR 2024 · similarity 0.89)
- Video-Bench: Human-Aligned Video Generation Benchmark (CVPR 2025 · similarity 0.89)
- Apollo: An Exploration of Video Understanding in Large Multimodal Models (CVPR 2025 · similarity 0.89)
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.
