String-level response similarity on RA-QA Multiple-choice, Discriminative tasks
0.85BERTScoreCareAQA-operaCT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CareAQA-operaCTBackbone=OPERA-CT2026.03 | 0.85 | 82.96 | |
| RAMoEA-QA2026.03 | 0.85 | 83.17 | |
| CareAQA-operaGTBackbone=OPERA-GT2026.03 | 0.84 | 81.72 | |
| PENGI2026.03 | -0.08 | 0 |