Explanations from Large Language Models Make Small Reasoners Better

TL;DR

This paper presents a method for training small language models to improve their reasoning and explanation capabilities by leveraging explanations generated by large models, outperforming larger models in accuracy.

Contribution

The paper introduces a multi-task learning framework that uses LLM-generated explanations to enhance small models' reasoning and explanation abilities, surpassing finetuning baselines.

Findings

Small models outperform finetuning baselines in reasoning tasks.

Method surpasses GPT-3 (175B) by up to 9.5% in accuracy.

Generated explanations are of high quality, supporting explainable AI.

Abstract

Integrating free-text explanations to in-context learning of large language models (LLM) is shown to elicit strong reasoning capabilities along with reasonable explanations. In this paper, we consider the problem of leveraging the explanations generated by LLM to improve the training of small reasoners, which are more favorable in real-production deployment due to their low cost. We systematically explore three explanation generation approaches from LLM and utilize a multi-task learning framework to facilitate small models to acquire strong reasoning power together with explanation generation capabilities. Experiments on multiple reasoning tasks show that our method can consistently and significantly outperform finetuning baselines across different settings, and even perform better than finetuning/prompting a 60x larger GPT-3 (175B) model by up to 9.5% in accuracy. As a side benefit, human evaluation further shows that our method can generate high-quality explanations to justify its predictions, moving towards the goal of explainable AI.