一作ACL 2026Poster· Natural Language Processing
Beyond Fully Random Masking: Attention-Guided Denoising and Optimization for Diffusion Language Models
Jia Deng · Junyi Li · Wayne Xin Zhao · Jinpeng Wang · Hongyu Lu · Ji-Rong Wen
摘要
Diffusion large language models (dLLMs) offer an efficient alternative to autoregressive models through parallel decoding, yet existing post-training methods largely rely on random masking strategies that overlook intrinsic token dependencies. In this work, we present an empirical analysis of att...