Grounded Video Question-Answering on ReXTime
29.92mIoUVTimeLLM
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| VTimeLLMSize=7B, Fine-tuned=true2025.03 | 29.92 | 43.69 | 26.13 | 57.58 | 17.13 | |
| VideoTemp-o3-7B-RLModel parameters=7B, Training stage=RL2026.02 | 29.5 | — | — | 74.4 | — | |
| VideoMindSize=7B, Fine-tuned=false2025.03 | 27.61 | 38.22 | 25.52 | 74.59 | 20.2 | |
| TimeChatSize=7B, Fine-tuned=true2025.03 | 26.29 | 40.13 | 21.42 | 49.46 | 10.92 | |
| VideoMindSize=2B, Fine-tuned=false2025.03 | 24.83 | 34.31 | 22.69 | 69.06 | 17.26 | |
| VideoTemp-o3-7B-SFTModel parameters=7B, Training stage=SFT2026.02 | 23.7 | — | — | 71.4 | — | |
| LITASize=13B, Fine-tuned=false2025.03 | 21.49 | 29.49 | 16.29 | 34.44 | — | |
| VTimeLLMSize=7B, Fine-tuned=false2025.03 | 20.14 | 28.84 | 17.41 | 36.16 | — | |
| VTimeLLM-7BModel parameters=7B2026.02 | 20.1 | — | — | 36.1 | — | |
| TimeChatSize=7B, Fine-tuned=false2025.03 | 11.65 | 14.42 | 7.61 | 40.04 | — | |
| TimeChat-7BModel parameters=7B2026.02 | 11.6 | — | — | 40 | — | |
| VideoChat-R1-7BModel parameters=7B2026.02 | 10.3 | — | — | 71.1 | — | |
| Qwen2.5-VL-7BModel parameters=7B2026.02 | 8.7 | — | — | 73.4 | — | |
| Video-R1-7BModel parameters=7B2026.02 | 7.3 | — | — | 70.5 | — |