Beyond Hard Masks: Progressive Token Evolution for Diffusion Language Models
Open paper
Benchmark
Model
Rank
Results
mathematical-reasoning-on-math-500
EvoToken-DLM (LLaDA-Instruct-8B, 10k FT)
#130
Accuracy: 36.33
arithmetic-reasoning-on-gsm8k
EvoToken-DLM (LLaDA-Instruct-8B, 10k FT)
#148
Accuracy: 80.74
Rank counts only results with a code link.