AI RESEARCH
CreditDecoding: Accelerating Parallel Decoding in Diffusion Large Language Models with Trace Credit
arXiv CS.CL
•
ArXi:2510.06133v2 Announce Type: replace Diffusion large language models (dLLMs) generate text through iterative denoising. In commonly adopted parallel decoding schemes, each step confirms only high-confidence positions while remasking the others. By analyzing dLLM denoising traces, we uncover a key inefficiency: models often predict the correct target token several steps before its confidence becomes high enough to be decoded. This gap between early prediction and late decoding forces repeated remasking of already-correct tokens, causing redundant iterations and limiting acceleration.