Assessing the Reliability of LLMs Annotations in the Context of Demographic Bias and Model Explanation
This work addresses fairness in NLP systems for sexism detection, but it is incremental as it builds on existing methods to quantify and mitigate demographic bias.
The study investigated how annotator demographics affect labeling decisions in sexism detection, finding that demographic factors accounted for only 8% of variance, with tweet content being dominant. It also assessed Generative AI models as annotators, showing that simplistic persona prompting often fails to improve or degrades performance compared to baselines.
Understanding the sources of variability in annotations is crucial for developing fair NLP systems, especially for tasks like sexism detection where demographic bias is a concern. This study investigates the extent to which annotator demographic features influence labeling decisions compared to text content. Using a Generalized Linear Mixed Model, we quantify this inf luence, finding that while statistically present, demographic factors account for a minor fraction ( 8%) of the observed variance, with tweet content being the dominant factor. We then assess the reliability of Generative AI (GenAI) models as annotators, specifically evaluating if guiding them with demographic personas improves alignment with human judgments. Our results indicate that simplistic persona prompting often fails to enhance, and sometimes degrades, performance compared to baseline models. Furthermore, explainable AI (XAI) techniques reveal that model predictions rely heavily on content-specific tokens related to sexism, rather than correlates of demographic characteristics. We argue that focusing on content-driven explanations and robust annotation protocols offers a more reliable path towards fairness than potentially persona simulation.