Intent Understanding on EgoPro-Bench Intent
72.15PrecisionProAct-Stream
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| ProAct-StreamTraining Strategy=RL, Think Data usage=W/O Think2026.05 | 72.15 | 50.11 | 55.14 | 72.39 | 55.96 | 3.81 | 3.01 | |
| ProAct-StreamTraining Strategy=SFT2026.05 | 71.75 | 50.07 | 55.02 | 72.34 | 56.28 | 3.81 | 3.02 | |
| ProAct-StreamTraining Strategy=RL, Think Data usage=W Think2026.05 | 66.5 | 57.86 | 56.34 | 76.13 | 61.62 | 4.1 | 3.23 | |
| Qwen3VL-8b2026.05 | 15.3 | 76.5 | 19.94 | 68.68 | 57.21 | 3.87 | 2.35 | |
| Qwen3VL-4B2026.05 | 13.29 | 73.09 | 18.04 | 68.33 | 55.19 | 3.49 | 2.1 | |
| Qwen3VL-30B-A3B2026.05 | 10.13 | 90.42 | 16.88 | 66.53 | 65.88 | 4.21 | 2.67 | |
| Qwen2.5VL-72B2026.05 | 9.34 | 98.86 | 16.14 | 57.39 | 68.89 | 3.95 | 1.98 | |
| Qwen2.5VL-32B2026.05 | 8.95 | 99.54 | 15.84 | 55.96 | 68.49 | 4.13 | 2.18 | |
| Qwen2.5vl-3B2026.05 | 8.91 | 100 | 15.8 | 52.7 | 68.71 | 3.9 | 1.68 | |
| Qwen2.5VL-7B2026.05 | 8.76 | 97 | 15.48 | 54.09 | 66.75 | 4.06 | 2.1 | |
| VideoRFT-7B2026.05 | 8.64 | 95.11 | 15.29 | 54.21 | 65.66 | 4.19 | 1.91 | |
| VideoChat-R1.52026.05 | 8.19 | 76.09 | 13.82 | 60.54 | 54.44 | 3.34 | 1.62 | |
| TimeChatOnline-7B2026.05 | 7.48 | 80.59 | 13.11 | 54.05 | 55.4 | 3.46 | 1.6 |