Attribute Structuring Improves LLM-Based Evaluation of Clinical Text Summaries

TL;DR

This paper introduces Attribute Structuring (AS), a framework that improves the evaluation of clinical text summaries by decomposing the process, leading to better alignment with human judgments and enabling efficient auditing.

Contribution

The paper proposes Attribute Structuring, a novel evaluation framework that enhances the accuracy and interpretability of LLM-based clinical summary assessments.

Findings

AS improves correlation with human annotations

AS provides interpretable evaluation spans

Framework enhances trustworthy clinical summary evaluation

Abstract

Summarizing clinical text is crucial in health decision-support and clinical research. Large language models (LLMs) have shown the potential to generate accurate clinical text summaries, but still struggle with issues regarding grounding and evaluation, especially in safety-critical domains such as health. Holistically evaluating text summaries is challenging because they may contain unsubstantiated information. Here, we explore a general mitigation framework using Attribute Structuring (AS), which structures the summary evaluation process. It decomposes the evaluation process into a grounded procedure that uses an LLM for relatively simple structuring and scoring tasks, rather than the full task of holistic summary evaluation. Experiments show that AS consistently improves the correspondence between human annotations and automated metrics in clinical text summarization. Additionally, AS yields interpretations in the form of a short text span corresponding to each output, which enables efficient human auditing, paving the way towards trustworthy evaluation of clinical information in resource-constrained scenarios. We release our code, prompts, and an open-source benchmark at https://github.com/microsoft/attribute-structuring.