Temporal Reasoning on EgoAVU-Bench
67.84AccuracyEgoAVU-Instruct (Full)
Evaluation Results
| Method | Links | |
|---|---|---|
| EgoAVU-Instruct (Full)Size=7B, Tuning=Full2026.02 | 67.84 | |
| EgoAVU-Instruct (LoRA)Size=7B, Tuning=LoRA2026.02 | 64.31 | |
| Qwen2.5-OmniSize=7B2026.02 | 53.2 | |
| Qwen2.5-OmniSize=3B2026.02 | 46.4 | |
| Phi4-mmSize=8B2026.02 | 45.04 | |
| Intern-OmniSize=8.7B2026.02 | 41.22 | |
| Baichuan-OmniSize=7B2026.02 | 39.85 | |
| VideoLLaMA2Size=7B2026.02 | 37 | |
| MiniCPM-oSize=8B2026.02 | 26.44 |