Mathematical Reasoning on GSM8K (Acc, TPF)
92.65AccuracyMultiBD (training-free)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MultiBD (training-free)Backbone=LLaDA2-Mini, bufsz=2, blksz=322026.06 | 92.65 | 2.76 | |
| MBD-LLaDA2-MiniBackbone=LLaDA2-Mini, bufsz=2, blksz=322026.06 | 91.96 | 5.55 | |
| SingleBD (Native)Backbone=LLaDA2-Mini-DMax, bufsz=2, blksz=322026.06 | 91.89 | 5.7 | |
| SingleBD (Native)Backbone=LLaDA2-Mini, bufsz=2, blksz=322026.06 | 91.89 | 2.27 | |
| MBD-SDAR-8B-Chat-b4Backbone=SDAR-8B-Chat-b4, bufsz=4, blksz=42026.06 | 91.81 | 2.28 | |
| MBD-LLaDA2-Mini-DMaxBackbone=LLaDA2-Mini-DMax, bufsz=2, blksz=322026.06 | 91.74 | 8.95 | |
| SingleBD (Native)Backbone=SDAR-8B-Chat-b4, bufsz=4, blksz=42026.06 | 91.05 | 1.33 | |
| MultiBD (training-free)Backbone=SDAR-8B-Chat-b4, bufsz=4, blksz=42026.06 | 90.45 | 2.39 | |
| SingleBD (Native)Backbone=SDAR-8B-Chat-b32, bufsz=4, blksz=322026.06 | 90.07 | 2.52 | |
| MultiBD (training-free)Backbone=LLaDA2-Mini-DMax, bufsz=2, blksz=322026.06 | 89.84 | 8.76 | |
| MBD-SDAR-8B-Chat-b32Backbone=SDAR-8B-Chat-b32, bufsz=4, blksz=322026.06 | 89.16 | 3.08 | |
| MultiBD (training-free)Backbone=SDAR-8B-Chat-b32, bufsz=4, blksz=322026.06 | 89.01 | 2.78 |