| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| EgoSchema | Gemini-1.5-Pro | Accuracy72.2 | 61 | 1mo ago | |
| IntentQA (test) | VideoTree | Accuracy66.9 | 17 | 1mo ago | |
| TVQA | FrozenBiLM | Accuracy82 | 14 | 1mo ago | |
| How2QA | FrozenBiLM | Accuracy86.7 | 13 | 1mo ago | |
| MLVU (test) | LLaVA-Video + Video-RAG | Accuracy73.8 | 10 | 1mo ago | |
| TVQA (test) | GPT-4V | Accuracy57.8 | 8 | 1mo ago | |
| LSMDC MC (test) | LAVENDER | Accuracy86 | 7 | 1mo ago | |
| VMB | VideoMindPalace | Spatial Accuracy (S)53 | 6 | 1mo ago | |
| NExT-QA (val) | MoReVQA | Accuracy69.2 | 6 | 1mo ago | |
| TVQA (val) | FrozenBiLM | Accuracy57.9 | 3 | 1mo ago | |
| How2QA (test) | - | Accuracy- | 0 | 1mo ago |
