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.
Bias in Language Models: Beyond Trick Tests and Towards RUTE d Evaluation
2 Pith papers cite this work, alongside 2 external citations. Polarity classification is still indexing.
2
Pith papers citing it
2
external citations · OpenAlex
years
2026 2representative citing papers
Generative AI evaluation must shift from static benchmark scores to measuring sustained improvements in human capabilities within specific deployment contexts.
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.
-
Benchmarked Yet Not Measured -- Generative AI Should be Evaluated Against Real-World Utility
Generative AI evaluation must shift from static benchmark scores to measuring sustained improvements in human capabilities within specific deployment contexts.