Planning on Countdown (accuracy and average accuracy)
85.6AccuracyGDSD w/ TLC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GDSD w/ TLCSequence Length=1282026.05 | 85.6 | — | |
| GDSDSequence Length=2562026.05 | 84 | — | |
| DMPOSequence Length=5122026.05 | 82.8 | — | |
| GDSDSequence Length=5122026.05 | 82.8 | — | |
| GDSDSequence Length=1282026.05 | 82.4 | — | |
| ESPOSequence Length=2562026.05 | 82 | — | |
| ESPOSequence Length=1282026.05 | 81.6 | — | |
| DMPOSequence Length=2562026.05 | 80.9 | — | |
| GDSD w/ TLCSequence Length=2562026.05 | 80.5 | — | |
| ESPOSequence Length=5122026.05 | 79.3 | — | |
| GDSD w/ TLCSequence Length=5122026.05 | 75 | — | |
| SPGSequence Length=2562026.05 | 70.7 | — | |
| SPGSequence Length=5122026.05 | 70.3 | — | |
| SPGSequence Length=1282026.05 | 68.8 | — | |
| DMPOSequence Length=1282026.05 | 67.2 | — | |
| UniGRPOSequence Length=5122026.05 | 57 | — | |
| wd1Sequence Length=2562026.05 | 51.2 | — | |
| wd1Sequence Length=1282026.05 | 47.7 | — | |
| wd1Sequence Length=5122026.05 | 46.1 | — | |
| UniGRPOSequence Length=1282026.05 | 44.5 | — | |
| UniGRPOSequence Length=2562026.05 | 43 | — | |
| diffu-GRPO (d1)Sequence Length=5122026.05 | 37.1 | — | |
| diffu-GRPO (d1)Sequence Length=1282026.05 | 33.2 | — | |
| diffu-GRPO (d1)Sequence Length=2562026.05 | 31.3 | — | |
| LLaDA-8BSequence Length=1282026.05 | 20.7 | — | |
| LLaDA-8BSequence Length=2562026.05 | 19.5 | — | |
| LLaDA-8BSequence Length=5122026.05 | 16 | — | |
| diffu-GRPO (d1)Sequence Length=Average2026.05 | — | 33.9 | |
| DMPOSequence Length=Average2026.05 | — | 77 | |
| ESPOSequence Length=Average2026.05 | — | 81 | |
| GDSDSequence Length=Average2026.05 | — | 83.1 | |
| GDSD w/ TLCSequence Length=Average2026.05 | — | 80.4 | |
| LLaDA-8BSequence Length=Average2026.05 | — | 18.7 | |
| SPGSequence Length=Average2026.05 | — | 69.9 | |
| UniGRPOSequence Length=Average2026.05 | — | 48.2 | |
| wd1Sequence Length=Average2026.05 | — | 48.3 |