Surgical Video Question Answering on REAL-Colon-Reason Level 3
51.5Exact Match (EM)RL framework over Digital Twin Representations
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RL framework over Digital Twin Representations2026.06 | 51.5 | 58.9 | |
| Qwen3-VL-8B2026.06 | 35.6 | 42.3 | |
| InternVL3-1B2026.06 | 22.1 | 29.5 | |
| Surgical-LVLM2026.06 | 20.8 | 27.2 | |
| Qwen3-VL-4B2026.06 | 19.4 | 25.8 | |
| PitVQA2026.06 | 15.3 | 22.1 | |
| Qwen2.5-VL-3B2026.06 | 14.8 | 20.5 | |
| SurgViVQA2026.06 | 14.2 | 20.5 | |
| MedGemma-4B2026.06 | 10.5 | 16.8 | |
| SurgicalGPT2026.06 | 8.2 | 12.5 | |
| VideoLLaMA3-2B2026.06 | 6.4 | 11.2 |