Soft-prompt Tuning for Large Language Models to Evaluate Bias
This work addresses the critical issue of bias in LLMs for industry applications, though it is incremental as it applies an existing soft-prompt tuning method to a new evaluation context.
The authors tackled the problem of quantifying biases in large language models (LLMs) like OPT and Galactica by using soft-prompt tuning on sentiment classification tasks, identifying bias patterns across sensitive attributes with group fairness metrics. They open-sourced their pipeline to help industry researchers adapt the method for practical use.
Prompting large language models has gained immense popularity in recent years due to the advantage of producing good results even without the need for labelled data. However, this requires prompt tuning to get optimal prompts that lead to better model performances. In this paper, we explore the use of soft-prompt tuning on sentiment classification task to quantify the biases of large language models (LLMs) such as Open Pre-trained Transformers (OPT) and Galactica language model. Since these models are trained on real-world data that could be prone to bias toward certain groups of populations, it is important to identify these underlying issues. Using soft-prompts to evaluate bias gives us the extra advantage of avoiding the human-bias injection that can be caused by manually designed prompts. We check the model biases on different sensitive attributes using the group fairness (bias) and find interesting bias patterns. Since LLMs have been used in the industry in various applications, it is crucial to identify the biases before deploying these models in practice. We open-source our pipeline and encourage industry researchers to adapt our work to their use cases.