Code Generation on MBPP (3) (pass@1)
43.6Pass@1ESPO
Evaluation Results
| Method | Links | |
|---|---|---|
| ESPOSequence Length=1282026.05 | 43.6 | |
| GDSDSequence Length=5122026.05 | 43.6 | |
| GDSD w/ TLCSequence Length=2562026.05 | 43.6 | |
| GDSD w/ TLCSequence Length=Avg.2026.05 | 43.3 | |
| ESPOSequence Length=2562026.05 | 43.2 | |
| GDSD w/ TLCSequence Length=5122026.05 | 43.2 | |
| GDSD w/ TLCSequence Length=1282026.05 | 43 | |
| ESPOSequence Length=Avg.2026.05 | 42.7 | |
| LLaDASequence Length=2562026.05 | 42 | |
| GDSDSequence Length=Avg.2026.05 | 42 | |
| GDSDSequence Length=2562026.05 | 41.8 | |
| ESPOSequence Length=5122026.05 | 41.2 | |
| SPGSequence Length=2562026.05 | 40.8 | |
| GDSDSequence Length=1282026.05 | 40.6 | |
| SPGSequence Length=Avg.2026.05 | 40.5 | |
| SPGSequence Length=1282026.05 | 40.4 | |
| SPGSequence Length=5122026.05 | 40.4 | |
| LLaDASequence Length=Avg.2026.05 | 38.8 | |
| LLaDASequence Length=5122026.05 | 38.1 | |
| diffu-GRPO (d1)Sequence Length=5122026.05 | 38 | |
| wd1Sequence Length=1282026.05 | 38 | |
| wd1Sequence Length=2562026.05 | 37.2 | |
| diffu-GRPO (d1)Sequence Length=2562026.05 | 36.6 | |
| diffu-GRPO (d1)Sequence Length=Avg.2026.05 | 36.5 | |
| wd1Sequence Length=Avg.2026.05 | 36.5 | |
| LLaDASequence Length=1282026.05 | 36.2 | |
| diffu-GRPO (d1)Sequence Length=1282026.05 | 34.8 | |
| wd1Sequence Length=5122026.05 | 34.4 |