Evaluating ChatGPT on Medical Information Extraction Tasks: Performance, Explainability and Beyond
This study addresses the applicability of ChatGPT in medical NLP tasks for researchers and practitioners, highlighting limitations that may hinder its use in critical healthcare applications, and is incremental as it assesses an existing model on specific tasks.
The paper evaluated ChatGPT's performance on four medical information extraction tasks across six benchmark datasets, finding that it underperforms compared to fine-tuned baseline models and exhibits over-confidence in predictions, though it provides high-quality explanations and shows faithfulness to the original text.
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.