Surgical Video Question Answering on REAL-Colon-Reason (Overall)
58.4EMRL framework over Digital Twin Representations
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RL framework over Digital Twin Representations2026.06 | 58.4 | 64.6 | |
| Qwen3-VL-8B2026.06 | 41.2 | 47.3 | |
| InternVL3-1B2026.06 | 35.2 | 41.4 | |
| Surgical-LVLM2026.06 | 30.2 | 36.3 | |
| PitVQA2026.06 | 29.7 | 36 | |
| Qwen3-VL-4B2026.06 | 27.9 | 33.8 | |
| SurgViVQA2026.06 | 27 | 33.2 | |
| Qwen2.5-VL-3B2026.06 | 22.5 | 28.1 | |
| MedGemma-4B2026.06 | 18.1 | 24.2 | |
| SurgicalGPT2026.06 | 15 | 20 | |
| VideoLLaMA3-2B2026.06 | 12.4 | 17.7 |