MDLMs lag autoregressive models in performance because parallel modeling weakens inter-token dependencies, yet they adapt generation order to task demands and show promise in a generate-then-edit paradigm.
Abulhair Saparov and He He
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Parallelism and Generation Order in Masked Diffusion Language Models: Limits Today, Potential Tomorrow
MDLMs lag autoregressive models in performance because parallel modeling weakens inter-token dependencies, yet they adapt generation order to task demands and show promise in a generate-then-edit paradigm.