Article

Large language models lose diabetes stigma signals when patient culture shifts

2026-06-12

Abstract excerpt

<title>Abstract</title> <p>A disease label does not always define a stable language model task. We test this problem in diabetes, a common and stigmatized condition where Type 1 and Type 2 communities share a diagnosis but use different cultural language for daily care and public judgment. We built a patient discourse corpus with a manually annotated reference set of 1526 posts from condition specific Type 1 and...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: cfabb683-5a0b-5c9c-ad54-6f171ad5f0fb
DOI: 10.21203/rs.3.rs-9931021/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Large language models lose diabetes stigma signals when patient culture shiftsDOI 10.21203/rs.3.rs-9931021/v1Read

Large Language Models Generate Stigmatizing Language During Reasoning Over Real-World Clinical DataSemantic proximity Read

Evaluating anti-LGBTQIA+ medical bias in large language modelsSemantic proximity Read

Using Large Language Models for sentiment analysis of health-related social media data: empirical evaluation and practical tipsSemantic proximity Read

Digital shadows in mental health map how LLMs simulate depression, anxiety, and stress through language and psychometricsSemantic proximity Read

Digital shadows in mental health map how LLMs simulate depression, anxiety, and stress through language and psychometricsSemantic proximity Read

Digital shadows in mental health map how LLMs simulate depression, anxiety, and stress through language and psychometricsSemantic proximity Read

Detecting Stigmatizing Language in Clinical Notes with Large Language Models for Addiction CareSemantic proximity Read

Trust in large language model-based solutions in healthcare among people with and without diabetes: a cross-sectional survey from the Health in Central Denmark cohortSemantic proximity Read

ChatGPT- versus human-generated answers to frequently asked questions about diabetes: a Turing test-inspired survey among employees of a Danish diabetes centerSemantic proximity Read

A Multi-Model, Multi-Domain Benchmark of Large Language Model Agreement with Humans and with Each Other in Sentiment ClassificationSemantic proximity Read

Leveraging Pretrained Large Language Model for Prognosis of Type 2 Diabetes Using Longitudinal Medical RecordsSemantic proximity Read

LLM-Based Annotation and Token-Augmented Modeling for Emotional Tone Classification in Online Cancer Peer-Support PostsSemantic proximity Read