Medical Question Answering on Medical QA Primary Split n = 314 (dev)
20.7UCCRvanilla answer
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| vanilla answeraction_space=binary-form2026.05 | 20.7 | 78.6 | 97.55 | 49.68 | |
| citation onlyaction_space=binary-form2026.05 | 20.7 | 78.6 | 97.55 | 49.68 | |
| binary answer / abstainaction_space=binary-form2026.05 | 0 | 77.93 | 85.35 | 40.76 | |
| verifier onlyaction_space=binary-form2026.05 | 0 | 76.59 | 84.5 | 39.49 | |
| speech-act-guided proxyaction_space=binary-form2026.05 | 0 | 97.32 | 96.52 | 47.77 | |
| claim-selective + NLI relationaction_space=native, relation_type=NLI2026.05 | 0 | 85.95 | 91.79 | 74.84 | |
| claim-selective + learned relationaction_space=native, relation_type=learned2026.05 | 0 | 97.66 | 98.15 | 87.58 | |
| claim-selective fullaction_space=native2026.05 | 0 | 100 | 99.5 | 92.04 |