Surgical Video Question Answering on REAL-Colon-Reason (Level 1)
65.3EMRL framework over Digital Twin Representations
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RL framework over Digital Twin Representations2026.06 | 65.3 | 70.8 | |
| InternVL3-1B2026.06 | 48.2 | 53.5 | |
| PitVQA2026.06 | 45.1 | 50.8 | |
| Qwen3-VL-8B2026.06 | 42.8 | 48.5 | |
| SurgViVQA2026.06 | 40.5 | 46.2 | |
| Surgical-LVLM2026.06 | 38.5 | 44.1 | |
| Qwen3-VL-4B2026.06 | 35.5 | 41.2 | |
| Qwen2.5-VL-3B2026.06 | 30.2 | 35.6 | |
| MedGemma-4B2026.06 | 25.6 | 31.4 | |
| SurgicalGPT2026.06 | 22.4 | 28.2 | |
| VideoLLaMA3-2B2026.06 | 18.5 | 24.1 |