New Studies Probe How LLMs Represent Psychology and Bias in Pain Care

Two new arXiv papers examine large language models: one tests whether they internally represent psychological constructs, and another audits GPT, Claude, Gemini, Llama, DeepSeek and Qwen for racial and sex bias in pain management recommendations.

Read the full story at arXiv cs.AI ↗

Timeline · 2 reports

1. 2026-11-13 05:00 · arXiv cs.CL
   Do Large Language Models Show Racial and Sex Bias in Pain Management? A Counterfactual Audit of GPT, Claude, Gemini, Llama, DeepSeek and Qwen
2. 2026-11-11 05:00 · arXiv cs.AI
   Tracing Psychometric Inference in Large Language Models