VOCABTRIM: Vocabulary Pruning for Efficient Speculative Decoding in LLMs

TL;DR

VocabTrim is a training-free vocabulary pruning technique that reduces speculative decoding overhead in large language models, significantly improving generation speed especially on memory-bound devices.

Contribution

VocabTrim introduces a simple method to prune the draft model's vocabulary based on target model sampling, enhancing decoding efficiency without retraining.

Findings

Achieves 16% memory-bound speed-up for Llama-3.2-3B-Instruct

Reduces drafting latency significantly in memory-bound environments

Maintains acceptable acceptance rates despite vocabulary pruning

Abstract

In this paper, we introduce a simple training-free technique to improve the performance of drafter-based speculative decoding (SpD) methods that incorporates language modeling head (LM head) during drafting process. A drafter-based speculative decoding leverages one or more smaller language models, a.k.a. drafters or draft models, to sample a draft sequence or tree consisting of multiple tokens, followed by verification by a base LLM, a target model, accepting a subset as its valid generation. As it is usually considered that the speculative decoding requires one-to-one mapping between vocabularies of the target model and the draft model, it has been natural to share the vocabulary between them, or even share the LM head as in EAGLE or Medusa. We first identify that this draft token sampling scheme inherently contains an unnecessary inference overhead in drafting, especially for some target LLMs with very large vocabularies. Then, we propose a simple technique, VocabTrim, to mitigate the drafting overhead to improve the generation speed in memory-bound environment. VocabTrim reconstructs the drafter LM head to contain only a limited set of tokens, selected by the most frequently sampled from the vocabulary of the target model. While limiting the vocabulary in drafting slightly degrades the acceptance rate, it significantly reduces the drafting latency in memory-bound process which is often the case on edge devices, resulting in higher memory-bound speed up (MBSU). We show that our method can boost the memory-bound speed-up for Llama-3 models on Spec-Bench, specifically by 16% for Llama-3.2-3B-Instruct.