Tag
This paper introduces a controlled audit of resume screening by LLMs, revealing that decisions can be unstable under presentation changes even when competence evidence is identical.
MUtE introduces a novel framework for optimal concept erasure and counterfactual interventions in text representations, enhancing algorithmic fairness and enabling counterfactual text generation in NLP models.
FairSelect is a toolkit for systematically evaluating fairness mitigation strategies across preprocessing, inprocessing, and postprocessing stages, supporting intersectional subgroup evaluation and comparison of fairness-utility tradeoffs.
This paper audits 14 large language models for hiring discrimination using a paired-resume methodology, finding that older models exhibit pro-White bias while newer models show null or pro-Black bias, indicating a reversal in algorithmic hiring bias across model generations.
This doctoral thesis critiques current fairness metrics in machine learning and proposes statistical hypothesis testing and structural analysis to address bias, emphasizing network and hierarchical contexts.
This paper proposes an α-Fair Individual Solvent Premium (α-FISP) framework for insurance pricing that balances actuarial fairness and solidarity fairness while ensuring solvency, using constrained optimization to yield a continuum of pricing solutions.
This paper investigates the gap between awareness and practice of algorithmic fairness in public health through a mixed-methods study, introducing the Fairness-to-Action framework to identify where translation stalls. The findings highlight weak institutionalization and system-level emphasis on accuracy over fairness.
This paper introduces the Explanation Fairness Taxonomy (EFT) to analyze disparities in how LLMs justify decisions across demographic groups, finding significant biases in explanation quality and tone despite balanced decisions.
This paper introduces a Probabilistic Graphical Model framework to causally audit LLM safety mechanisms, revealing that standard observational metrics overestimate demographic bias by ignoring context toxicity.