Temporal Visual Reasoning on SVU-31K 1.0 (test)
11.3BLEU-4MedScope
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| MedScopeTool Use=true, Reasoning=true, Zero-shot=true2026.02 | 11.3 | 4.56 | 16.89 | 34.83 | |
| SurgVidLMTool Use=false, Reasoning=false, Zero-shot=true2026.02 | 10.1 | 3.83 | 14.13 | 31.58 | |
| VideoR1-7BTool Use=false, Reasoning=true, Zero-shot=true2026.02 | 8.25 | 3.13 | 14.23 | 24.41 | |
| VideoLLaMA3-7BTool Use=false, Reasoning=false, Zero-shot=true2026.02 | 7.68 | 1.92 | 11.53 | 23.19 | |
| VideoRFT-7BTool Use=true, Reasoning=true, Zero-shot=true2026.02 | 6.45 | 2.52 | 11.44 | 20.43 | |
| LongVT-7B-RFTTool Use=true, Reasoning=true, Zero-shot=true2026.02 | 6.22 | 1.77 | 14.99 | 21.26 | |
| Video-LLaVA-7BTool Use=false, Reasoning=false, Zero-shot=true2026.02 | 5.65 | 2.17 | 11.59 | 19.73 | |
| Qwen2.5-VL-7BTool Use=false, Reasoning=false, Zero-shot=true2026.02 | 4.41 | 1.55 | 13.09 | 16.27 | |
| VideoChat-R1-7BTool Use=true, Reasoning=true, Zero-shot=true2026.02 | 4.3 | 1.56 | 11.89 | 19.83 | |
| InternVL3-8BTool Use=false, Reasoning=false, Zero-shot=true2026.02 | 3.83 | 1.13 | 11.28 | 18.32 | |
| GroundingGPT-7BTool Use=false, Reasoning=false, Zero-shot=true2026.02 | 2.81 | 1.44 | 10.52 | 17.86 |