Few-shot Instruction Prompts for Pretrained Language Models to Detect Social Biases
This addresses the problem of building bias detectors efficiently for researchers and practitioners, though it is incremental as it adapts existing few-shot prompting techniques to a specific domain.
The paper tackled the challenge of detecting social bias in text by proposing a few-shot instruction-based method using pre-trained language models, achieving similar or superior accuracy to fine-tuned models with a 13% improvement in AUC for the largest model and maintaining high performance with as few as 100 labeled samples.
Detecting social bias in text is challenging due to nuance, subjectivity, and difficulty in obtaining good quality labeled datasets at scale, especially given the evolving nature of social biases and society. To address these challenges, we propose a few-shot instruction-based method for prompting pre-trained language models (LMs). We select a few class-balanced exemplars from a small support repository that are closest to the query to be labeled in the embedding space. We then provide the LM with instruction that consists of this subset of labeled exemplars, the query text to be classified, a definition of bias, and prompt it to make a decision. We demonstrate that large LMs used in a few-shot context can detect different types of fine-grained biases with similar and sometimes superior accuracy to fine-tuned models. We observe that the largest 530B parameter model is significantly more effective in detecting social bias compared to smaller models (achieving at least 13% improvement in AUC metric compared to other models). It also maintains a high AUC (dropping less than 2%) when the labeled repository is reduced to as few as $100$ samples. Large pretrained language models thus make it easier and quicker to build new bias detectors.