Spoken Stereoset: on Evaluating Social Bias Toward Speaker in Speech Large Language Models
Part Of
Proceedings of 2024 IEEE Spoken Language Technology Workshop, SLT 2024
Start Page
871
End Page
878
ISBN
979-835039225-8
Date Issued
2024-12-02
Author(s)
DOI
10.1109/SLT61566.2024.10832259
Abstract
Warning: This paper may contain texts with uncomfortable content.Large Language Models (LLMs) have achieved remarkable performance in various tasks, including those involving multimodal data like speech. However, these models often exhibit biases due to the nature of their training data. Recently, more Speech Large Language Models (SLLMs) have emerged, underscoring the urgent need to address these biases. This study introduces Spoken Stereoset, a dataset specifically designed to evaluate social biases in SLLMs. By examining how different models respond to speech from diverse demographic groups, we aim to identify these biases. Our experiments reveal significant insights into their performance and bias levels. The findings indicate that while most models show minimal bias, some still exhibit slightly stereotypical or anti-stereotypical tendencies.
Event(s)
2024 IEEE Spoken Language Technology Workshop, SLT 2024
SDGs
Publisher
IEEE
Type
conference paper