Mathematical Question Answering on GSM8K
72.4AUROCB1 max entropy
Evaluation results
| Method | Links | ||
|---|---|---|---|
| B1 max entropyCost=Free2026.03 | 72.4 | 65 | |
| Combined (4 entropy feat, CV)Cost=Free2026.03 | 72.4 | — | |
| P(True)Cost=1 call2026.03 | 60.8 | 52 |
| Method | Links | ||
|---|---|---|---|
| B1 max entropyCost=Free2026.03 | 72.4 | 65 | |
| Combined (4 entropy feat, CV)Cost=Free2026.03 | 72.4 | — | |
| P(True)Cost=1 call2026.03 | 60.8 | 52 |