A Survey of Large Language Models in Medicine: Progress, Application, and Challenge

Large language models (LLMs), such as ChatGPT, have achieved substantial attention due to their impressive human language understanding and generation capabilities. Therefore, the application of LLMs in medicine to assist physicians and patient care emerges as a promising research direction in both artificial intelligence and clinical medicine. To this end, this survey provides a comprehensive overview of the current progress, applications, and challenges faced by LLMs in medicine. Specifically, we aim to address the following questions: 1) What are LLMs and how can medical LLMs be built? 2) What are the downstream performances of medical LLMs? 3) How can medical LLMs be utilized in real-world clinical practice? 4) What challenges arise from the use of medical LLMs? 5) How can we better construct and utilize medical LLMs? As a result, this survey aims to provide insights into the opportunities and challenges of LLMs in medicine and serve as a valuable resource for constructing practical and effective medical LLMs. A regularly updated list of practical guide resources of medical LLMs can be found at https://github.com/AI-in-Health/MedLLMsPracticalGuide.

Missing Information, Unresponsive Authors, Experimental Flaws: The Impossibility of Assessing the Reproducibility of Previous Human Evaluations in NLP

We report our efforts in identifying a set of previous human evaluations in NLP that would be suitable for a coordinated study examining what makes human evaluations in NLP more/less reproducible. We present our results and findings, which include that just 13\% of papers had (i) sufficiently low barriers to reproduction, and (ii) enough obtainable information, to be considered for reproduction, and that all but one of the experiments we selected for reproduction was discovered to have flaws that made the meaningfulness of conducting a reproduction questionable. As a result, we had to change our coordinated study design from a reproduce approach to a standardise-then-reproduce-twice approach. Our overall (negative) finding that the great majority of human evaluations in NLP is not repeatable and/or not reproducible and/or too flawed to justify reproduction, paints a dire picture, but presents an opportunity for a rethink about how to design and report human evaluations in NLP.