Summary of Position Paper on Diagnostic Uncertainty Estimation From Large Language Models: Next-word Probability Is Not Pre-test Probability, by Yanjun Gao et al.
Position Paper On Diagnostic Uncertainty Estimation from Large Language Models: Next-Word Probability Is Not Pre-test Probability
by Yanjun Gao, Skatje Myers, Shan Chen, Dmitriy Dligach, Timothy A Miller, Danielle Bitterman, Guanhua Chen, Anoop Mayampurath, Matthew Churpek, Majid Afshar
First submitted to arxiv on: 7 Nov 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Computation and Language (cs.CL)
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 paper evaluates the use of large language models (LLMs) for diagnostic decision support, specifically examining their ability to estimate pre-test probabilities crucial for clinical decision-making. The study compares two LLMs, Mistral-7B and Llama3-70B, using structured electronic health record data on three diagnosis tasks. It identifies limitations in current methods of extracting LLM probability estimations and highlights the need for improved techniques in estimating confidence. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at how big language models can help doctors make decisions. Right now, these models aren’t very good at guessing what’s likely to happen before a patient gets tested. The study tests two types of these models, Mistral-7B and Llama3-70B, using health records. It shows that the current ways we use these models have some problems and emphasizes the importance of finding better ways to do this. |
Keywords
» Artificial intelligence » Probability
