Code Generation on Aggregate All evaluated datasets (test)
48.4Precision@1JPlag-RLVR
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| JPlag-RLVRChange=↑%2026.05 | 48.4 | 46.2 | 32.3 | 77.4 | 73.8 | |
| PKPOChange=↑%2026.05 | 45.7 | 42.4 | 30 | 60 | 57.6 | |
| Base-RLVRChange=↑%2026.05 | 45.5 | 37.6 | 24.5 | 36.1 | 33.3 | |
| Pass@k-RLVRChange=↑%2026.05 | 38.9 | 35.4 | 25.8 | 66.9 | 73.3 | |
| Pass@k-RLVRChange=↓%2026.05 | 29.6 | 19.7 | 5.5 | 32.4 | 26.6 | |
| JPlag-RLVRChange=↓%2026.05 | 28.7 | 12.1 | 2.2 | 22.6 | 26.2 | |
| PKPOChange=↓%2026.05 | 26.7 | 15.7 | 3.6 | 38.9 | 42.4 | |
| JPlag-RLVRChange=Δ2026.05 | 16.4 | 19.7 | 17.3 | 0.298 | 0.33 | |
| Base-RLVRChange=↓%2026.05 | 13.5 | 10.6 | 5.6 | 57.2 | 65.3 | |
| Base-RLVRChange=Δ2026.05 | 9.5 | 11.5 | 10.6 | -0.046 | -0.53 | |
| PKPOChange=Δ2026.05 | 7.2 | 14.9 | 15.5 | 0.083 | -0.13 | |
| Pass@k-RLVRChange=Δ2026.05 | 4.5 | 11.3 | 12.4 | 0.123 | 0.32 |