04:00
2026-07-24
arxiv.org
large-language-models
DC-Leap: Training-Free Acceleration of dLLMs via Draft-Guided Contiguous Leaping Decoding
Researchers propose DC-Leap, a training-free framework that accelerates Diffusion Large Language Models (dLLMs) by up to 53.19x on MBPP for long-sequence generation and up to 105.02x when combined wit…