Gender Bias in Emotion Recognition by Large Language Models

TL;DR

This paper investigates gender bias in emotion recognition by large language models and finds that training-based debiasing strategies are more effective than prompt engineering alone.

Contribution

It introduces and evaluates training-based debiasing methods to reduce gender bias in LLMs' emotion recognition tasks.

Findings

Training interventions significantly reduce gender bias.

Prompt engineering alone is insufficient for bias mitigation.

Debiasing improves fairness in emotion recognition by LLMs.

Abstract

The rapid advancement of large language models (LLMs) and their growing integration into daily life underscore the importance of evaluating and ensuring their fairness. In this work, we examine fairness within the domain of emotional theory of mind, investigating whether LLMs exhibit gender biases when presented with a description of a person and their environment and asked, ''How does this person feel?''. Furthermore, we propose and evaluate several debiasing strategies, demonstrating that achieving meaningful reductions in bias requires training based interventions rather than relying solely on inference-time prompt-based approaches such as prompt engineering, etc.