Six state-of-the-art LLMs systematically prefer Standard American English over AAE continuations, and a training-free activation steering method reduces this bias 5-20x more than prompting while preserving fluency.
Don’t Forget the Teachers
2 Pith papers cite this work, alongside 41 external citations. Polarity classification is still indexing.
2
Pith papers citing it
41
external citations · OpenAlex
fields
cs.CL 2years
2026 2representative citing papers
Model developers must address human concerns, preferences, values, and goals with rigor at every stage of the LLM pipeline rather than only in post-training.
citing papers explorer
-
LLMs Silently Correct African American English: Auditing and Mitigating Dialect Bias via Activation Steering
Six state-of-the-art LLMs systematically prefer Standard American English over AAE continuations, and a training-free activation steering method reduces this bias 5-20x more than prompting while preserving fluency.
-
Reflections and New Directions for Human-Centered Large Language Models
Model developers must address human concerns, preferences, values, and goals with rigor at every stage of the LLM pipeline rather than only in post-training.