Video Question Answering on HD-EPIC 1 (test)
96.7RecipeSample Human Baseline
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Sample Human Baseline2026.04 | 96.7 | 96.7 | 85 | 92.5 | 93.8 | 92.7 | 75 | 90.3 | |
| Gemini ProModality=Video-Language2026.04 | 60.5 | 46.2 | 34.7 | 39.6 | 32.5 | 20.8 | 28.7 | 37.6 | |
| Belief + VLM + RLModality=Video-Language, Backbone=InternVL-2B2026.04 | 54.9 | 59.2 | 56.8 | 43 | 42.1 | 42.3 | 43.5 | 48.8 | |
| VLM + RLModality=Video-Language, Backbone=InternVL-2B2026.04 | 49.5 | 51.7 | 53.2 | 40.5 | 40.2 | 38.6 | 43.5 | 44.6 | |
| Gemini ProModality=Blind - Language Only2026.04 | 38 | 26.8 | 30 | 22.1 | 21.5 | 27.7 | 20.5 | 26.7 | |
| LLaVA-VideoModality=Video-Language2026.04 | 36.3 | 33.5 | 38.7 | 43 | 27.3 | 18.9 | 29.3 | 32.4 | |
| Llama 3.2Modality=Blind - Language Only2026.04 | 33.5 | 25 | 36.7 | 23.3 | 22.3 | 25.5 | 19.5 | 26.5 | |
| VideoLlama 2Modality=Video-Language2026.04 | 30.8 | 25.7 | 32.7 | 27.2 | 25.7 | 28.5 | 21.2 | 27.4 | |
| LongVAModality=Video-Language2026.04 | 29.6 | 30.8 | 33.7 | 30.7 | 32.9 | 22.7 | 24.5 | 29.3 |