Saber: An Efficient Sampling with Adaptive Acceleration and Backtracking Enhanced Remasking for Diffusion Language Model

arXiv:2510.18165v2 Announce Type: replace Abstract: Diffusion language models (DLMs) are emerging as a powerful and promising alternative to the dominant autoregressive paradigm, offering inherent advantages in parallel generation and bidirectional context modeling. However, the performance of DLMs on code generation tasks, which have stronger structural constraints, is significantly hampered by the critical trade-off between inference speed and output quality. We observed that accelerating the