AI RESEARCH
Remask, Don't Replace: Token-to-Mask Refinement in Masked Diffusion Language Models
arXiv CS.CL
•
ArXi:2604.18738v1 Announce Type: new Masked diffusion language models such as LLaDA2.1 rely on Token-to-Token (T2T) editing to correct their own generation errors: whenever a different token crosses a confidence threshold, the committed token is overwritten. We identify three structural failure modes of this rule.