Logical Planning on Countdown (test)
84.8AccuracyGDSD w/ TLC
Evaluation Results
| Method | Links | |
|---|---|---|
| GDSD w/ TLCSeq Len=512, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 84.8 | |
| GDSD w/ TLCSeq Len=Avg., Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 83.5 | |
| GDSD w/ TLCSeq Len=128, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 83.2 | |
| GDSD w/ TLCSeq Len=256, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 82.4 | |
| GDSDSeq Len=128, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 70.3 | |
| ESPOSeq Len=128, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 68.8 | |
| GDSDSeq Len=256, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 68.4 | |
| GDSDSeq Len=Avg., Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 67.5 | |
| ESPOSeq Len=256, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 66.8 | |
| ESPOSeq Len=Avg., Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 66.8 | |
| ESPOSeq Len=512, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 64.8 | |
| GDSDSeq Len=512, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 63.7 | |
| wd1Seq Len=512, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 42.2 | |
| wd1Seq Len=256, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 37.9 | |
| diffu-GRPO (d1)Seq Len=512, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 37.5 | |
| wd1Seq Len=Avg., Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 36.3 | |
| diffu-GRPO (d1)Seq Len=Avg., Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 30.8 | |
| wd1Seq Len=128, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 28.9 | |
| diffu-GRPO (d1)Seq Len=256, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 27.7 | |
| diffu-GRPO (d1)Seq Len=128, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 27.3 | |
| Dream-7BSeq Len=512, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 17.4 | |
| Dream-7BSeq Len=Avg., Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 11.2 | |
| Dream-7BSeq Len=128, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 8.5 | |
| Dream-7BSeq Len=256, Zero-shot=true, Base Model=Dream-7B-Instruct2026.05 | 7.8 |