Recent works have shown considerable improvements in task-oriented dialogue (TOD) systems by utilizing pretrained large language models (LLMs) in an end-to-end manner. However, the biased behavior of each component in a TOD system and the error propagation issue in the end-to-end framework can lead to seriously biased TOD responses. Existing works of fairness only focus on the total bias of a system. In this paper, we propose a diagnosis method to attribute bias to each component of a TOD system. With the proposed attribution method, we can gain a deeper understanding of the sources of bias. Additionally, researchers can mitigate biased model behavior at a more granular level. We conduct experiments to attribute the TOD system's bias toward three demographic axes: gender, age, and race. Experimental results show that the bias of a TOD system usually comes from the response generation model.
@article{arxiv.2311.06513,
title = {Step by Step to Fairness: Attributing Societal Bias in Task-oriented Dialogue Systems},
author = {Hsuan Su and Rebecca Qian and Chinnadhurai Sankar and Shahin Shayandeh and Shang-Tse Chen and Hung-yi Lee and Daniel M. Bikel},
journal= {arXiv preprint arXiv:2311.06513},
year = {2023}
}