Tracing the Latent Threads: A Mechanistic Study of How LLMs Represent and Operationalize Race and Ethnicity Cues
cs.CL, cs.LG
Submitted: 2026-01-19
Updated: 2026-08-31
Terminology
Sources
- The Llama 3 Herd of Models
- Elucidating Mechanisms of Demographic Bias in LLMs for Healthcare
- Mistral 7B
- Can SAEs reveal and mitigate racial biases of LLMs in healthcare?
- Robustly Improving LLM Fairness in Realistic Settings via Interpretability
- Assessing Bias in Metric Models for LLM Open-Ended Generation Bias Benchmarks
- ChatGPT Exhibits Gender and Racial Biases in Acute Coronary Syndrome Management
- FairSteer: Inference Time Debiasing for LLMs with Dynamic Activation Steering
- Fairness Evaluation and Inference Level Mitigation in LLMs
- Bias patterns in the application of LLMs for clinical decision support: A comprehensive study
- GLU Variants Improve Transformer
- Understanding and Mitigating Gender Bias in LLMs via Interpretable Neuron Editing
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering