Multi-Block Diffusion Language Models
Open paper
Benchmark
Model
Rank
Results
arithmetic-reasoning-on-gsm8k
MBD-LLaDA2-Mini
#55
Accuracy: 91.96
mathematical-reasoning-on-math-500
MBD-LLaDA2-Mini
#95
Accuracy: 79.2
Rank counts only results with a code link.