VWG-Bench
Explore score history ↗VWG-Bench evaluates image-to-video generation models on rule-following, physical and commonsense constraints, and goal realization over time, using 380 annotated initial images across 9 reasoning dimensions and 38 ten-sample task groups, scored with bundled 1–5 metrics and an evaluation toolkit.
- Released
- 2026-09-10
- Evaluates
- General AI, Reasoning, Vision & 3D
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
VWG-Bench paper and dataset
No reported scores are on record for this benchmark yet.