Social identity markers in medical questions degrade LLM accuracy and uncertainty calibration, producing a calibration crisis that is non-additive for intersectional cases.
arXiv preprint arXiv:2507.02799
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
verdicts
UNVERDICTED 3representative citing papers
Reasoning LLMs aggregate social biases through stereotype repetition and irrelevant information injection in their thinking processes, and a self-review prompt mitigates this on BBQ, StereoSet, and BOLD benchmarks.
State-of-the-art LLMs respond inconsistently to queries from protected-group personas, with some responses omitting key information that should be provided.
citing papers explorer
-
Calibrated? Not for Everyone: How Sexual Orientation and Religious Markers Distort LLM Accuracy and Confidence in Medical QA
Social identity markers in medical questions degrade LLM accuracy and uncertainty calibration, producing a calibration crisis that is non-additive for intersectional cases.
-
Investigating Thinking Behaviours of Reasoning-Based Language Models for Social Bias Mitigation
Reasoning LLMs aggregate social biases through stereotype repetition and irrelevant information injection in their thinking processes, and a self-review prompt mitigates this on BBQ, StereoSet, and BOLD benchmarks.
-
Discriminatory Compliance: How LLMs Answer Queries from Protected Groups
State-of-the-art LLMs respond inconsistently to queries from protected-group personas, with some responses omitting key information that should be provided.