Token-Budget-Aware LLM Reasoning

TL;DR

This paper introduces a token-budget-aware framework for LLM reasoning that dynamically adjusts reasoning length to reduce token costs while maintaining performance, addressing efficiency concerns in Chain-of-Thought methods.

Contribution

It proposes a novel dynamic token adjustment method for LLM reasoning that balances cost and accuracy, improving upon static approaches.

Findings

Reduces token usage in reasoning processes

Maintains high reasoning accuracy with budget adjustments

Offers a practical solution for cost-efficient LLM reasoning

Abstract

Reasoning is critical for large language models (LLMs) to excel in a wide range of tasks. While methods like Chain-of-Thought (CoT) reasoning and enhance LLM performance by decomposing problems into intermediate steps, they also incur significant overhead in token usage, leading to increased costs. We find that the reasoning process of current LLMs is unnecessarily lengthy and it can be compressed by including a reasonable token budget in the prompt, but the choice of token budget plays a crucial role in the actual compression effectiveness. We then propose a token-budget-aware LLM reasoning framework that dynamically adjusts the number of reasoning tokens based on the reasoning complexity of each problem. Experiments show that our method effectively reduces token costs in CoT reasoning with only a slight performance reduction, offering a practical solution to balance efficiency and accuracy in LLM reasoning. Code: https://github.com/GeniusHTX/TALE