Surgical Video Question Answering on REAL-Colon-Reason (Level 2)
58.4Exact Match (EM)RL framework over Digital Twin Representations
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RL framework over Digital Twin Representations2026.06 | 58.4 | 64.2 | |
| Qwen3-VL-8B2026.06 | 45.2 | 51.1 | |
| InternVL3-1B2026.06 | 35.4 | 41.2 | |
| Surgical-LVLM2026.06 | 31.2 | 37.5 | |
| Qwen3-VL-4B2026.06 | 28.9 | 34.5 | |
| PitVQA2026.06 | 28.6 | 35.2 | |
| SurgViVQA2026.06 | 26.4 | 32.8 | |
| Qwen2.5-VL-3B2026.06 | 22.4 | 28.1 | |
| MedGemma-4B2026.06 | 18.2 | 24.5 | |
| SurgicalGPT2026.06 | 14.5 | 19.3 | |
| VideoLLaMA3-2B2026.06 | 12.3 | 17.8 |