CLAILGFeb 27, 2025

Self-Training Elicits Concise Reasoning in Large Language Models

arXiv:2502.20122v387 citationsh-index: 8Has CodeACL
AI Analysis

This addresses the issue of extraneous inference costs for users deploying LLMs in reasoning tasks, though it is incremental as it builds on existing fine-tuning and sampling techniques.

The paper tackles the problem of redundant tokens in chain-of-thought reasoning for large language models, proposing self-training methods that achieve a 30% reduction in output tokens while maintaining accuracy on GSM8K and MATH benchmarks.

Chain-of-thought (CoT) reasoning has enabled large language models (LLMs) to utilize additional computation through intermediate tokens to solve complex tasks. However, we posit that typical reasoning traces contain many redundant tokens, incurring extraneous inference costs. Upon examination of the output distribution of current LLMs, we find evidence on their latent ability to reason more concisely, relative to their default behavior. To elicit this capability, we propose simple fine-tuning methods which leverage self-generated concise reasoning paths obtained by best-of-N sampling and few-shot conditioning, in task-specific settings. Our combined method achieves a 30% reduction in output tokens on average, across five model families on GSM8K and MATH, while maintaining average accuracy. By exploiting the fundamental stochasticity and in-context learning capabilities of LLMs, our self-training approach robustly elicits concise reasoning on a wide range of models, including those with extensive post-training. Code is available at https://github.com/TergelMunkhbat/concise-reasoning

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes