Skip to yearly menu bar Skip to main content


Poster

CLLMs: Consistency Large Language Models

Siqi Kou · Lanxiang Hu · Zhezhi He · Zhijie Deng · Hao Zhang

Hall C 4-9 #604
[ ] [ Project Page ] [ Paper PDF ]
Wed 24 Jul 4:30 a.m. PDT — 6 a.m. PDT

Abstract: Jacobi decoding shows promise for more efficient LLM inference as it breaks the sequential nature of the LLM decoding process and transforms it into more parallelizable computation. However, in practice, it achieves little speedup compared to traditional autoregressive (AR) decoding, primarily because Jacobi decoding seldom accurately predicts more than one token in a single fixed-point iteration step. To address this, we develop a new approach aimed at realizing fast convergence from any state to the fixed point in a Jacobi trajectory. This is accomplished by refining the target LLM to consistently predict the fixed point given any state as input. Extensive experiments demonstrate the effectiveness of our method, showing 2.4$\times$ to 3.4$\times$ improvements in generation speed while preserving generation quality across both domain-specific and open-domain benchmarks.

Chat is not available.