This paper is in the following e-collection/theme issue:

Chatbots and Conversational Agents (754) AI Language Models in Health Care (215) Generative Language Models Including ChatGPT (745)

Published on in Vol 27 (2025)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/64348, first published .
Token Probabilities to Mitigate Large Language Models Overconfidence in Answering Medical Questions: Quantitative Study

Token Probabilities to Mitigate Large Language Models Overconfidence in Answering Medical Questions: Quantitative Study

Token Probabilities to Mitigate Large Language Models Overconfidence in Answering Medical Questions: Quantitative Study

Authors of this article:

Raphaël Bentegeac1, 2 Author Orcid Image ;   Bastien Le Guellec3, 4 Author Orcid Image ;   Grégory Kuchcinski3, 4 Author Orcid Image ;   Philippe Amouyel1, 2, 5 Author Orcid Image ;   Aghiles Hamroun1, 2, 5 Author Orcid Image
Article Authors Cited by (3) Tweetations (2) Metrics

Journals

  1. Chan J, Kwek R. Uncovering bias and variability in how large language models attribute cardiovascular risk. Frontiers in Digital Health 2025;7 View
  2. Callens S. Effective prompt design for large language models in clinical practice. Acta Clinica Belgica 2026:1 View
  3. Small W, Crowley R, Pariente C, Zhang J, Eaton K, Jiang L, Oermann E, Aphinyanaphongs Y. Enhancing the prediction of hospital discharge disposition with extraction-based language model classification. npj Health Systems 2026;3(1) View