Evaluating ChatGPT on Medical Information Extraction Tasks: Performance, Explainability and Beyond

TL;DR

This study evaluates ChatGPT's performance, explainability, confidence, and faithfulness in medical information extraction tasks, revealing strengths in explainability but limitations in accuracy and over-confidence compared to specialized models.

Contribution

It provides a comprehensive analysis of ChatGPT's capabilities and limitations in medical information extraction, highlighting areas for improvement and potential challenges in clinical applications.

Findings

ChatGPT underperforms compared to fine-tuned models in MedIE tasks.

ChatGPT offers high-quality explanations but is often over-confident.

Uncertainty in generation affects extraction reliability.

Abstract

Large Language Models (LLMs) like ChatGPT have demonstrated amazing capabilities in comprehending user intents and generate reasonable and useful responses. Beside their ability to chat, their capabilities in various natural language processing (NLP) tasks are of interest to the research community. In this paper, we focus on assessing the overall ability of ChatGPT in 4 different medical information extraction (MedIE) tasks across 6 benchmark datasets. We present the systematically analysis by measuring ChatGPT's performance, explainability, confidence, faithfulness, and uncertainty. Our experiments reveal that: (a) ChatGPT's performance scores on MedIE tasks fall behind those of the fine-tuned baseline models. (b) ChatGPT can provide high-quality explanations for its decisions, however, ChatGPT is over-confident in its predcitions. (c) ChatGPT demonstrates a high level of faithfulness to the original text in the majority of cases. (d) The uncertainty in generation causes uncertainty in information extraction results, thus may hinder its applications in MedIE tasks.