Coding on Coding (val)
100Pass@16GRPO
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GRPODifficulty=Easy, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 100 | 95.8 | |
| CoRPODifficulty=Easy, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 100 | 94.6 | |
| GRPODifficulty=Medium, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 100 | 93.1 | |
| CoRPODifficulty=Medium, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 100 | 88.8 | |
| GRPODifficulty=Easy, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 100 | 90.1 | |
| CoRPODifficulty=Easy, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 100 | 88.9 | |
| CoRPODifficulty=Medium, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 100 | 67.2 | |
| GRPODifficulty=Medium, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 96.6 | 62.5 | |
| GRPODifficulty=Overall, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 87.8 | 67.1 | |
| CoRPODifficulty=Overall, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 87.2 | 65.5 | |
| CoRPODifficulty=Overall, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 83.9 | 47.2 | |
| GRPODifficulty=Overall, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 83.4 | 47 | |
| GRPODifficulty=Hard, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 81.7 | 53.4 | |
| CoRPODifficulty=Hard, Validation Set Domain=In-domain, Train Set=Coding, MSL=16k, Temperature=0.32025.11 | 80.9 | 52.4 | |
| GRPODifficulty=Hard, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 75.6 | 31.3 | |
| CoRPODifficulty=Hard, Validation Set Domain=Out-of-domain, Train Set=Math, MSL=16k, Temperature=0.32025.11 | 75.6 | 30.9 |