DA-DLM: Explicitly Modeling Token Dependencies in Diffusion Language Models
Diffusion Language Models (DLMs) generate text by iteratively denoising a masked sequence, independently predicting multiple tokens at each step. This conditional independence discards inter-token dependencies and degrades coherence-an issue that parallels the multi-modality problem in Non-Autoregressive Translation (N...
Peng-Yu Ji, Zi-Chen Zhang, Xiang Hu et al.
· 0 citations