H-SAL erases latent concepts from text profiles using self-descriptions as implicit debiasing signals and shows competitive performance on a new multi-domain Stack Exchange helpfulness benchmark.
Fairness in large language models: A taxonomic survey
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 2years
2026 2roles
background 1polarities
background 1representative citing papers
LLMs produce explanations with significant disparities in verbosity, sentiment, hedging, faithfulness, and lexical complexity across demographic groups, varying by model and only partially mitigated by prompting.
citing papers explorer
-
Debiasing Without Protected Attributes: Latent Concept Erasure from Textual Profiles
H-SAL erases latent concepts from text profiles using self-descriptions as implicit debiasing signals and shows competitive performance on a new multi-domain Stack Exchange helpfulness benchmark.
-
Explanation Fairness in Large Language Models: An Empirical Analysis of Disparities in How LLMs Justify Decisions Across Demographic Groups
LLMs produce explanations with significant disparities in verbosity, sentiment, hedging, faithfulness, and lexical complexity across demographic groups, varying by model and only partially mitigated by prompting.