Medical Visual Question Answering on SLAKE Closed (AUROC)
73.2AUROCHAC-Platt
Evaluation Results
| Method | Links | |
|---|---|---|
| HAC-PlattModel=InternVL3-2B, Confidence=Samp.2026.04 | 73.2 | |
| HAC-GateModel=InternVL3-38B, Confidence=Samp.2026.04 | 70 | |
| HAC-PlattModel=Qwen3-VL-32B, Confidence=Verb.2026.04 | 69 | |
| HAC-PlattModel=Qwen3-VL-8B, Confidence=Samp.2026.04 | 68.6 | |
| HAC-GateModel=Qwen3-VL-8B, Confidence=Samp.2026.04 | 68.6 | |
| HAC-PlattModel=Qwen3-VL-2B, Confidence=Verb.2026.04 | 65.7 | |
| HAC-PlattModel=Qwen3-VL-2B, Confidence=Samp.2026.04 | 65.6 | |
| HAC-GateModel=Qwen3-VL-2B, Confidence=Samp.2026.04 | 65.6 | |
| HAC-GateModel=Qwen3-VL-2B, Confidence=Verb.2026.04 | 65.4 | |
| UncalibratedModel=Qwen3-VL-2B, Confidence=Samp.2026.04 | 60.7 | |
| Platt ScalingModel=Qwen3-VL-2B, Confidence=Samp.2026.04 | 60.7 | |
| Isotonic Regr.Model=Qwen3-VL-2B, Confidence=Samp.2026.04 | 60.7 | |
| UncalibratedModel=Qwen3-VL-2B, Confidence=Verb.2026.04 | 58.6 | |
| Platt ScalingModel=Qwen3-VL-2B, Confidence=Verb.2026.04 | 58.6 | |
| Isotonic Regr.Model=Qwen3-VL-2B, Confidence=Verb.2026.04 | 58.6 | |
| Hist. BinningModel=Qwen3-VL-2B, Confidence=Samp.2026.04 | 56.6 | |
| Hist. BinningModel=Qwen3-VL-2B, Confidence=Verb.2026.04 | 50 |