CLLMs: Consistency Large Language Models

28 Feb 2024  ·  Siqi Kou, Lanxiang Hu, Zhezhi He, Zhijie Deng, Hao Zhang ·

Parallel decoding methods such as Jacobi decoding show promise for more efficient LLM inference as it breaks the sequential nature of the LLM decoding process and transforms it into 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 on 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.

PDF Abstract

Results from the Paper

  Submit results from this paper to get state-of-the-art GitHub badges and help the community compare results to other papers.