Code Suggestion on DS-1000
51Binary RewardShapE-GRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| ShapE-GRPObackbone=QWEN3-8B2026.03 | 51 | |
| GRPObackbone=QWEN3-8B2026.03 | 46.5 | |
| WTAbackbone=QWEN3-8B2026.03 | 41.8 | |
| QWEN3-8Bstatus=pretrained2026.03 | 39.8 |
| Method | Links | |
|---|---|---|
| ShapE-GRPObackbone=QWEN3-8B2026.03 | 51 | |
| GRPObackbone=QWEN3-8B2026.03 | 46.5 | |
| WTAbackbone=QWEN3-8B2026.03 | 41.8 | |
| QWEN3-8Bstatus=pretrained2026.03 | 39.8 |