Bias Analysis and Mitigation through Protected Attribute Detection and Regard Classification
Large language models (LLMs) acquire general linguistic knowledge from massive-scale pretraining. However, pretraining data mainly comprised of web-crawled texts contain undesirable social biases which can be perpetuated or even amplified by LLMs. In this study, we propose an efficient yet effective annotation pipeline to investigate social biases in the pretraining corpora. Our pipeline consists of protected attribute detection to identify diverse demographics, followed by regard classification to analyze the language polarity towards each attribute. Through our experiments, we demonstrate the effect of our bias analysis and mitigation measures, focusing on Common Crawl as the most representative pretraining corpus.
Code (0)
등록된 구현이 없습니다.
Tasks
AttributeSimilar Papers 제목 키워드 기반
Bias Mitigation Framework for Intersectional Subgroups in Neural Networks
We propose a fairness-aware learning framework that mitigates intersectional subgroup bias associated with protected attributes. Prior research has primarily focused on mitigating one kind of bias by incorporating comple…
FairnessScaling Fair Learning to Hundreds of Intersectional Groups
Bias mitigation algorithms aim to reduce the performance disparity between different protected groups. Existing techniques focus on settings where there is a small number of protected groups arising from a single protect…
AttributeFairnessKnowledge DistillationUnbiased Model Prediction Without Using Protected Attribute Information
The problem of bias persists in the deep learning community as models continue to provide disparate performance across different demographic subgroups. Therefore, several algorithms have been proposed to improve the fair…
Modular and On-demand Bias Mitigation with Attribute-Removal Subnetworks
Societal biases are reflected in large pre-trained language models and their fine-tuned versions on downstream tasks. Common in-processing bias mitigation approaches, such as adversarial training and mutual information r…
AttributeDisentanglementMeasuring and Mitigating Bias for Tabular Datasets with Multiple Protected Attributes
Motivated by the recital (67) of the current corrigendum of the AI Act in the European Union, we propose and present measures and mitigation strategies for discrimination in tabular datasets. We specifically focus on dat…
AttributeFairness