Summary of Quality Of Answers Of Generative Large Language Models Vs Peer Patients For Interpreting Lab Test Results For Lay Patients: Evaluation Study, by Zhe He et al.
Quality of Answers of Generative Large Language Models vs Peer Patients for Interpreting Lab Test Results for Lay Patients: Evaluation Study
by Zhe He, Balu Bhasuran, Qiao Jin, Shubo Tian, Karim Hanna, Cindy Shavor, Lisbeth Garcia Arguello, Patrick Murray, Zhiyong Lu
First submitted to arxiv on: 23 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The abstract discusses the feasibility of using Large Language Models (LLMs) like ChatGPT to generate accurate, helpful, and safe responses to lab test-related questions asked by patients. The researchers collected QA data from Yahoo! Answers, selected 53 pairs for study, and used LLMs GPT-4, Meta LLaMA 2, MedAlpaca, and ORCA_mini to generate responses. They evaluated the similarity of answers using standard metrics like ROUGE, BLEU, METEOR, and BERTScore, and also utilized an LLM-based evaluator for relevance, correctness, helpfulness, and safety. Manual evaluation by medical experts showed that GPT-4’s responses achieved better scores than other LLMs and human responses on all four aspects. However, LLM responses may lack interpretation in a patient’s medical context, contain incorrect statements, or lack references. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at using special computer models called Large Language Models (LLMs) to help patients understand lab test results. They collected questions and answers from the internet and had four different LLMs generate responses. The researchers compared these responses to each other and to human answers, looking for accuracy, helpfulness, and safety. They found that one of the LLMs, called GPT-4, did a better job than others on most aspects. However, there are still some problems with using LLMs in this way. |
Keywords
» Artificial intelligence » Bleu » Gpt » Llama » Rouge
