Intersectional Social Attribute

Intersectional social attribute research examines how multiple social identities (e.g., race, gender, socioeconomic status) interact to create unique experiences of bias and injustice, moving beyond the analysis of single attributes. Current research focuses on detecting and mitigating these intersectional biases in various domains, including healthcare, AI systems (like vision-language models and LLMs), and credit allocation, often employing causal inference methods, logistic regression, and fairness metrics to analyze data and model performance across intersecting subgroups. This work is crucial for developing fairer and more equitable systems across numerous sectors, highlighting the limitations of single-attribute analyses and promoting a more nuanced understanding of social inequality.