Breast Ultrasound Diagnosis on BrEaST OOD (test)
0.685AUCUltrasoundAgents
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| UltrasoundAgentsBackbone=Qwen2.5-3B, Training Strategy=Three-stage RL (GRPO) and SFT, Inference Workflow=ROI -> attribute -> diagnosis evidence-chain2026.03 | 0.685 | 72.5 | 15.7 | 0.037 | |
| Qwen2.5-3B-COT-SFTBackbone=Qwen2.5-3B, Training Strategy=SFT on template-based reasoning data, Inference Workflow=Intermediate attribute evidence2026.03 | 0.586 | 62.7 | 17.6 | 0.06 | |
| Think-with-ImageBackbone=Qwen2.5-3B, Training Strategy=DeepEyes-style single-agent workflow, Inference Workflow=ROI localization -> crop-and-zoom -> re-analyze2026.03 | 0.526 | 64.7 | 19.6 | 0.005 | |
| Qwen2.5-3B-Zero-ShotBackbone=Qwen2.5-3B, Training Strategy=Zero-shot, Inference Workflow=Directly prompts on full image2026.03 | 0.484 | 60.8 | 7.8 | -0.065 |