Explain-Query-Test: Self-Evaluating LLMs Via Explanation and Comprehension Discrepancy

TL;DR

This paper introduces Explain-Query-Test (EQT), a self-evaluation pipeline for LLMs that assesses their comprehension by generating explanations, questions, and answers, revealing gaps between explanation quality and reasoning ability.

Contribution

EQT provides a novel self-evaluation method that predicts model performance and uncovers comprehension limitations without external data.

Findings

EQT accuracy correlates with benchmark performance like MMLU-Pro.

Models show a gap between explanation detail and reasoning ability.

EQT can rank models based on their comprehension without external datasets.

Abstract

Large language models (LLMs) have demonstrated remarkable proficiency in generating detailed and coherent explanations of complex concepts. However, the extent to which these models truly comprehend the concepts they articulate remains unclear. To assess the level of comprehension of a model relative to the content it generates, we implemented a self-evaluation pipeline where models: (i) given a topic generate an excerpt with information about the topic, (ii) given an excerpt generate question-answer pairs, and finally (iii) given a question generate an answer. We refer to this self-evaluation approach as Explain-Query-Test (EQT). Interestingly, the accuracy on generated questions resulting from running the EQT pipeline correlates strongly with the model performance as verified by typical benchmarks such as MMLU-Pro. In other words, EQT's performance is predictive of MMLU-Pro's, and EQT can be used to rank models without the need for any external source of evaluation data other than lists of topics of interest. Moreover, our results reveal a disparity between the models' ability to produce detailed explanations and their performance on questions related to those explanations. This gap highlights fundamental limitations in the internal knowledge representation and reasoning abilities of current LLMs. We release the code at https://github.com/asgsaeid/EQT.