ArticleJournal of the American Medical Informatics Association : JAMIA2025
Large language model uncertainty proxies: discrimination and calibration for medical diagnosis and treatment.
Article in Journal of the American Medical Informatics Association : JAMIA, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 54 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
54 citing papers in PubMed.
- From Innovation to Impact: The CREATE Framework as a Blueprint for Large Language Model Adoption in Opioid Treatment Programs.Journal of medical Internet research · 2026Article
- Navigating the Prescription Drug Information System.Clinical pharmacology and therapeutics · 2026Review
- On-premise medical AI agents for reliable clinical decision-making.Nature medicine · 2026Article
- The missing links in agentic AI autonomy.Nature medicine · 2026Article
- Conditional Perplexity Scoring for Large Language Model-Generated Differential Diagnoses in Case Reports: Preliminary Computational Evaluation.JMIR formative research · 2026Article
- Accuracy, Self-Reported Confidence, and Overconfidence of Large Language Models in Endodontics: An Evaluation Using National Specialty Examination Questions.Medical science monitor : international medical journal of experimental and clinical research · 2026Article
- Article
- The economics of accuracy for medical reasoning with large language models.PLOS digital health · 2026Article
- Navigating uncertainty matters: Evaluating large language models for drug-drug interaction identification.Journal of managed care & specialty pharmacy · 2026Article
- Artificial intelligence in hepatocellular carcinoma diagnosis and AI-assisted treatment support: Current landscape and future directions.Liver research (Beijing, China) · 2026Review
- Evaluation Methods for Inference-Time Retrieval-Augmented and Graph Retrieval-Augmented Large Language Models in Health Care: Scoping Review.Journal of medical Internet research · 2026Article
- A framework for human-artificial intelligence co-learning for disease activity labeling using electronic health records.medRxiv : the preprint server for health sciences · 2026Article
- Benchmarking Multimodal Large Language Models for Cardiopulmonary Findings on Chest Radiographs: Sex-Stratified Discrimination and Operating Characteristics.Diagnostics (Basel, Switzerland) · 2026Article
- Large language models in respiratory care: safety and clinical integration.ERJ open research · 2026Article
- A pilot study on the impact of large language model assistance on the evaluation of complex medical living kidney donor candidates.Clinical transplantation and research · 2026Article
- Structured reasoning failures compromise LLM interpretation of clinical oncology notes.NPJ digital medicine · 2026Article
- Calibration of Self-Reported Confidence and Accuracy of Large Language Models in Medical Question Answering.Journal of medical systems · 2026Article
- Confidence Measurement Metrics in Multimodal Large Language Models for Ultrasound-Based Radiology Cases: Comparative Evaluation Study of Self-Reported, Consistency-Based, and Hybrid Methods.Journal of medical Internet research · 2026Article
- Benchmarking reliability and calibration of LLMs for multi-cancer early detection test communication.JAMIA open · 2026Article
- Development and External Validation of a Large Language Model-Based Clinical Decision-Support System for Colonoscopy Surveillance Intervals.Clinical gastroenterology and hepatology : the official clinical practice journal of the American Gastroenterological Association · 2026Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
8 authors.
Funding
Abstract
introductionThe inability of large language models (LLMs) to communicate uncertainty is a significant barrier to their use in medicine. Before LLMs can be integrated into patient care, the field must assess methods to estimate uncertainty in ways that are useful to physician-users.
objectiveEvaluate the ability for uncertainty proxies to quantify LLM confidence when performing diagnosis and treatment selection tasks by assessing the properties of discrimination and calibration.
methodsWe examined confidence elicitation (CE), token-level probability (TLP), and sample consistency (SC) proxies across GPT3.5, GPT4, Llama2, and Llama3. Uncertainty proxies were evaluated against 3 datasets of open-ended patient scenarios.
resultsSC discrimination outperformed TLP and CE methods. SC by sentence embedding achieved the highest discriminative performance (ROC AUC 0.68-0.79), yet with poor calibration. SC by GPT annotation achieved the second-best discrimination (ROC AUC 0.66-0.74) with accurate calibration. Verbalized confidence (CE) was found to consistently overestimate model confidence. DISCUSSION AND
conclusionsSC is the most effective method for estimating LLM uncertainty of the proxies evaluated. SC by sentence embedding can effectively estimate uncertainty if the user has a set of reference cases with which to re-calibrate their results, while SC by GPT annotation is the more effective method if the user does not have reference cases and requires accurate raw calibration. Our results confirm LLMs are consistently over-confident when verbalizing their confidence (CE).
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.