Medical Question Answering on MedQA (val)
81AccuracyRLCR
Evaluation results
| Method | Links | ||
|---|---|---|---|
| RLCRτ=0.952026.07 | 81 | 1.7 | |
| RLSRτ=1.02026.07 | 78.9 | 3 | |
| RLVR†τ=1.02026.07 | 66.6 | — | |
| RLVR-Temp†τ=0.622026.07 | 66.6 | — |
| Method | Links | ||
|---|---|---|---|
| RLCRτ=0.952026.07 | 81 | 1.7 | |
| RLSRτ=1.02026.07 | 78.9 | 3 | |
| RLVR†τ=1.02026.07 | 66.6 | — | |
| RLVR-Temp†τ=0.622026.07 | 66.6 | — |