AI RESEARCH
LogicDiff: Logic-Guided Denoising Improves Reasoning in Masked Diffusion Language Models
arXiv CS.LG
•
ArXi:2603.26771v1 Announce Type: cross Masked diffusion language models (MDLMs) generate text by iteratively unmasking tokens from a fully masked sequence, offering parallel generation and bidirectional context. However, their standard confidence-based unmasking strategy systematically defers high-entropy logical connective tokens, the critical branching points in reasoning chains, leading to severely degraded reasoning performance. We