Erasure Dependence Evaluation on Aggregate (AIME24, AIME25, ZebraLogic, AutoLogi, GPQA-D)
-1.6ASGCERL full
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| CERL fullTraining=SFT+RL2026.05 | -1.6 | 99.8 | 8.8 |
Loading the SOTA2 catalog…
| Method | Links | |||
|---|---|---|---|---|
| CERL fullTraining=SFT+RL2026.05 | -1.6 | 99.8 | 8.8 |