Side-by-side comparison of intent-equivalent SAE and AAVE tweets significantly exacerbates covert dialect bias in LMs compared to isolated evaluation, with explicit dialect labels worsening the effect further.
InProceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11328–11348, Toronto, Canada
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it