Turning Trash into Treasure: Accelerating Inference of Large Language Models with Token Recycling

ACL ARR 2024 December Submission862 Authors

Abstract:

The rapid growth in the parameters of LLMs has made inference latency a fundamental bottleneck. Speculative decoding represents a lossless approach to accelerate inference through a guess-and-verify paradigm. Some methods rely on additional architectures to guess draft tokens, which need extra training before use.
Alternatively, retrieval-based train-free techniques build libraries from pre-existing corpora or by n-gram generation. However, they face challenges like large storage requirements, time-consuming retrieval, and limited adaptability.
Observing that candidate tokens generated during the decoding process are likely to reoccur in future sequences, we propose Token Recycling.
This approach stores candidate tokens in an adjacency matrix and employs a breadth-first-search (BFS)-like algorithm to construct a draft tree, which is then validated through tree attention.
New candidate tokens from the decoding process are then used to update the matrix.
Token Recycling requires \textless2MB of additional storage and achieves approximately 2x speedup across all sizes of LLMs. It significantly outperforms existing train-free methods by 30\% and even a training method by 25\%.

Paper Type: Long

Research Area: Efficient/Low-Resource Methods for NLP

Research Area Keywords: Inference Acceleration; Efficient Inference

Contribution Types: NLP engineering experiment, Reproduction study, Approaches to low-resource settings, Approaches low compute settings-efficiency, Publicly available software and/or pre-trained models

Languages Studied: English;Python

Submission Number: 862