How Interpretable are Reasoning Explanations from Prompting Large Language Models?

TL;DR

This paper evaluates the interpretability of reasoning explanations from large language models, considering multiple dimensions beyond faithfulness, and introduces a new alignment technique that significantly improves interpretability.

Contribution

It provides a comprehensive, multifaceted evaluation of reasoning explanations and proposes a novel Self-Entailment-Alignment Chain-of-Thought method that enhances interpretability.

Findings

Over 70% improvement in interpretability metrics

Evaluation across multiple reasoning benchmarks

Analysis of various prompting techniques

Abstract

Prompt Engineering has garnered significant attention for enhancing the performance of large language models across a multitude of tasks. Techniques such as the Chain-of-Thought not only bolster task performance but also delineate a clear trajectory of reasoning steps, offering a tangible form of explanation for the audience. Prior works on interpretability assess the reasoning chains yielded by Chain-of-Thought solely along a singular axis, namely faithfulness. We present a comprehensive and multifaceted evaluation of interpretability, examining not only faithfulness but also robustness and utility across multiple commonsense reasoning benchmarks. Likewise, our investigation is not confined to a single prompting technique; it expansively covers a multitude of prevalent prompting techniques employed in large language models, thereby ensuring a wide-ranging and exhaustive evaluation. In addition, we introduce a simple interpretability alignment technique, termed Self-Entailment-Alignment Chain-of-thought, that yields more than 70\% improvements across multiple dimensions of interpretability. Code is available at https://github.com/SenticNet/CoT_interpretability