Future Video Question Answering on FutureVQA
61.9Acc@1sQwen2.5-VL-7b
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| Qwen2.5-VL-7b2026.03 | 61.9 | 49.5 | 40.7 | -21.2 | 47.2 | 0.31 | 0.68 | |
| FutureAgenttemporal label training=true2026.03 | 60.8 | 50.7 | 43.6 | -16.6 | 50.1 | 0.21 | 0.78 | |
| FutureAgenttemporal label training=true, Chain of Thought=false2026.03 | 60.5 | 48.4 | 41.3 | -19.2 | 48.2 | 0.3 | 0.75 | |
| Baselinetemporal label training=true2026.03 | 60.2 | 48.2 | 38.1 | -22.7 | 46.1 | 0.36 | 0.73 | |
| GPT-4o2026.03 | 59.1 | 41.1 | 31.6 | -27.5 | 42.2 | 0.42 | 0.64 | |
| LLV-Video2026.03 | 53.7 | 46.5 | 43.4 | -10.3 | 46.8 | 0.18 | 0.81 | |
| Vid-LMA22026.03 | 52.4 | 41.2 | 37.2 | -15.2 | 42.4 | 0.28 | 0.78 | |
| Baselinetemporal label training=false2026.03 | 49.8 | 44.1 | 33.1 | -16.7 | 38.6 | 0.33 | 0.75 | |
| FutureAgenttemporal label training=false2026.03 | 49.2 | 46.7 | 36 | -13.2 | 41.4 | 0.25 | 0.79 | |
| GPT-4o-mini2026.03 | 47.7 | 36 | 32 | -15.7 | 37.7 | 0.29 | 0.69 | |
| Yi-VL-34b2026.03 | 38.1 | 30 | 26.1 | -12 | 28.4 | 0.29 | 0.7 | |
| LLV-v1.5-13b2026.03 | 37.8 | 30.9 | 26.3 | -11.5 | 30.7 | 0.27 | 0.73 | |
| LLV-v1.5-7b2026.03 | 24 | 18.1 | 16 | -8 | 18.6 | 0.24 | 0.55 | |
| CogVLM-17b2026.03 | 22.8 | 19.4 | 14 | -8.8 | 17.2 | 0.3 | 0.59 | |
| LLV-Next-13b2026.03 | 15.4 | 9.3 | 4.2 | -11.2 | 7.3 | 0.6 | 0.39 |