Inducing Group Fairness in LLM-Based Decisions

arXiv:2406.16738v1 Announce Type: new
Abstract: Prompting Large Language Models (LLMs) has created new and interesting means for classifying textual data. While evaluating and remediating group fairness is a well-studied problem in classifier fairness literature, some classical approaches (e.g., regularization) do not carry over, and some new opportunities arise (e.g., prompt-based remediation). We measure fairness of LLM-based classifiers on a toxicity classification task, and empirically show that prompt-based classifiers may lead to unfair decisions. We introduce several remediation techniques and benchmark …

abstract arxiv classifier cs.ai cs.cy cs.lg data decisions fairness language language models large language large language models literature llm llms opportunities problem prompt prompting regularization textual type while