Action Recognition on SAR-RARP50
48.1AccuracySurg-R1
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Surg-R1Thinking=true2026.03 | 48.1 | 29.95 | 45.49 | 22.55 | |
| Gemma3-27B-itThinking=false2026.03 | 33.2 | 13.86 | 18.01 | 5.32 | |
| Gemini-3.0-Pro-ThinkingThinking=true2026.03 | 31.81 | 18.74 | 37.41 | 10.56 | |
| Qwen2.5-VL-32B-InstructThinking=false2026.03 | 31.8 | 13.65 | 41.35 | 5.21 | |
| Qwen2.5-VL-7B-Surg-CholecT50Thinking=false2026.03 | 31.5 | 12.73 | 6.89 | 4.28 | |
| Qwen2.5-VL-7B-InstructThinking=false2026.03 | 31.1 | 12.5 | 3.89 | 3.89 | |
| MiniCPM-O-2_6Thinking=false2026.03 | 30.8 | 13.13 | 20.12 | 5.74 | |
| GPT-5.1-ThinkingThinking=true2026.03 | 30.64 | 19.42 | 19.9 | 11.21 | |
| InternVL3-78BThinking=false2026.03 | 29.5 | 12.87 | 22.82 | 6.28 | |
| InternVL3-8BThinking=false2026.03 | 29.3 | 12.78 | 9.96 | 6.13 | |
| Qwen2.5-MaxThinking=false2026.03 | 28.3 | 14.6 | 10.2 | 7.14 | |
| Qwen2.5-VL-72B-InstructThinking=false2026.03 | 28.2 | 13.04 | 22.3 | 5.94 | |
| GPT-4o-2024-0806Thinking=false2026.03 | 28.1 | 15.99 | 16.15 | 9.16 | |
| Llama-4-Scout-17B-16E-InstructThinking=false2026.03 | 25.1 | 12.5 | 3.14 | 3.14 | |
| Gemini-2.0-FlashThinking=false2026.03 | 24.4 | 17.51 | 18.54 | 7.27 | |
| MiniCPM-V-2_6Thinking=false2026.03 | 24.3 | 12.07 | 13.43 | 7.25 | |
| Phi-4-MultimodalThinking=false2026.03 | 15.1 | 12.83 | 11.41 | 3.47 | |
| Mistral-Small-3.1-24B-Instruct-2503Thinking=false2026.03 | 12.5 | 12.68 | 5.89 | 1.78 | |
| Skywork-R1V-38BThinking=false2026.03 | 12.3 | 13.02 | 19.83 | 3.08 | |
| LLaVA-1.5-7BThinking=false2026.03 | 5.1 | 12.5 | 0.64 | 0.64 |