Traversal Verification for Speculative Tree Decoding

TL;DR

This paper proposes Traversal Verification, a novel speculative decoding method for large language models that improves acceptance length and throughput by rethinking verification from leaf to root, ensuring lossless inference.

Contribution

It introduces a leaf-to-root traversal verification approach that guarantees identical probability distribution to the target model, enhancing efficiency in speculative decoding.

Findings

Improves acceptance length and throughput in large language models

Guarantees lossless inference through theoretical proof

Consistently outperforms existing speculative decoding methods

Abstract

Speculative decoding is a promising approach for accelerating large language models. The primary idea is to use a lightweight draft model to speculate the output of the target model for multiple subsequent timesteps, and then verify them in parallel to determine whether the drafted tokens should be accepted or rejected. To enhance acceptance rates, existing frameworks typically construct token trees containing multiple candidates in each timestep. However, their reliance on token-level verification mechanisms introduces two critical limitations: First, the probability distribution of a sequence differs from that of individual tokens, leading to suboptimal acceptance length. Second, current verification schemes begin from the root node and proceed layer by layer in a top-down manner. Once a parent node is rejected, all its child nodes should be discarded, resulting in inefficient utilization of speculative candidates. This paper introduces Traversal Verification, a novel speculative decoding algorithm that fundamentally rethinks the verification paradigm through leaf-to-root traversal. Our approach considers the acceptance of the entire token sequence from the current node to the root, and preserves potentially valid subsequences that would be prematurely discarded by existing methods. We theoretically prove that the probability distribution obtained through Traversal Verification is identical to that of the target model, guaranteeing lossless inference while achieving substantial acceleration gains. Experimental results across different large language models and multiple tasks show that our method consistently improves acceptance length and throughput over existing methods.