AI RESEARCH
On the Reasoning Abilities of Masked Diffusion Language Models
arXiv CS.LG
•
ArXi:2510.13117v3 Announce Type: replace Masked diffusion models (MDMs) for text offer a compelling alternative to traditional autoregressive language models. Parallel generation makes them efficient, but their computational capabilities and the limitations inherent in their parallelism remain largely unexplored. To this end, we characterize what types of reasoning problems MDMs can provably solve and how efficiently.