Investigating Intersectional Bias in Large Language Models using Confidence Disparities in Coreference Resolution
- URL: http://arxiv.org/abs/2508.07111v1
- Date: Sat, 09 Aug 2025 22:24:40 GMT
- Title: Investigating Intersectional Bias in Large Language Models using Confidence Disparities in Coreference Resolution
- Authors: Falaah Arif Khan, Nivedha Sivakumar, Yinong Oliver Wang, Katherine Metcalf, Cezanne Camacho, Barry-John Theobald, Luca Zappella, Nicholas Apostoloff,
- Abstract summary: Large language models (LLMs) have achieved impressive performance, leading to their widespread adoption as decision-support tools in resource-constrained contexts like hiring and admissions.<n>There is, however, scientific consensus that AI systems can reflect and exacerbate societal biases, raising concerns about identity-based harm when used in critical social contexts.<n>In this work, we extend single-axis fairness evaluations to examine intersectional bias, recognizing that when multiple axes of discrimination intersect, they create distinct patterns of disadvantage.
- Score: 5.061421107401101
- License: http://creativecommons.org/licenses/by-nc-nd/4.0/
- Abstract: Large language models (LLMs) have achieved impressive performance, leading to their widespread adoption as decision-support tools in resource-constrained contexts like hiring and admissions. There is, however, scientific consensus that AI systems can reflect and exacerbate societal biases, raising concerns about identity-based harm when used in critical social contexts. Prior work has laid a solid foundation for assessing bias in LLMs by evaluating demographic disparities in different language reasoning tasks. In this work, we extend single-axis fairness evaluations to examine intersectional bias, recognizing that when multiple axes of discrimination intersect, they create distinct patterns of disadvantage. We create a new benchmark called WinoIdentity by augmenting the WinoBias dataset with 25 demographic markers across 10 attributes, including age, nationality, and race, intersected with binary gender, yielding 245,700 prompts to evaluate 50 distinct bias patterns. Focusing on harms of omission due to underrepresentation, we investigate bias through the lens of uncertainty and propose a group (un)fairness metric called Coreference Confidence Disparity which measures whether models are more or less confident for some intersectional identities than others. We evaluate five recently published LLMs and find confidence disparities as high as 40% along various demographic attributes including body type, sexual orientation and socio-economic status, with models being most uncertain about doubly-disadvantaged identities in anti-stereotypical settings. Surprisingly, coreference confidence decreases even for hegemonic or privileged markers, indicating that the recent impressive performance of LLMs is more likely due to memorization than logical reasoning. Notably, these are two independent failures in value alignment and validity that can compound to cause social harm.
Related papers
- Who's Asking? Investigating Bias Through the Lens of Disability Framed Queries in LLMs [2.722784054643991]
Large Language Models (LLMs) routinely infer users demographic traits from phrasing alone.<n>Disability cues in shaping these inferences remains largely uncharted.<n>We present the first systematic audit of disability-conditioned demographic bias across eight state-of-the-art instruction-tuned LLMs.
arXiv Detail & Related papers (2025-08-18T21:03:09Z) - Mitigating Subgroup Disparities in Multi-Label Speech Emotion Recognition: A Pseudo-Labeling and Unsupervised Learning Approach [53.824673312331626]
Implicit Demography Inference (IDI) module uses k-means clustering to mitigate bias in Speech Emotion Recognition (SER)<n>Experiments show that pseudo-labeling IDI reduces subgroup disparities, improving fairness metrics by over 28%.<n>Unsupervised IDI yields more than a 4.6% improvement in fairness metrics with a drop of less than 3.6% in SER performance.
arXiv Detail & Related papers (2025-05-20T14:50:44Z) - Fairness Mediator: Neutralize Stereotype Associations to Mitigate Bias in Large Language Models [66.5536396328527]
LLMs inadvertently absorb spurious correlations from training data, leading to stereotype associations between biased concepts and specific social groups.<n>We propose Fairness Mediator (FairMed), a bias mitigation framework that neutralizes stereotype associations.<n>Our framework comprises two main components: a stereotype association prober and an adversarial debiasing neutralizer.
arXiv Detail & Related papers (2025-04-10T14:23:06Z) - The Root Shapes the Fruit: On the Persistence of Gender-Exclusive Harms in Aligned Language Models [91.86718720024825]
We center transgender, nonbinary, and other gender-diverse identities to investigate how alignment procedures interact with pre-existing gender-diverse bias.<n>Our findings reveal that DPO-aligned models are particularly sensitive to supervised finetuning.<n>We conclude with recommendations tailored to DPO and broader alignment practices.
arXiv Detail & Related papers (2024-11-06T06:50:50Z) - GenderCARE: A Comprehensive Framework for Assessing and Reducing Gender Bias in Large Language Models [73.23743278545321]
Large language models (LLMs) have exhibited remarkable capabilities in natural language generation, but have also been observed to magnify societal biases.<n>GenderCARE is a comprehensive framework that encompasses innovative Criteria, bias Assessment, Reduction techniques, and Evaluation metrics.
arXiv Detail & Related papers (2024-08-22T15:35:46Z) - VLBiasBench: A Comprehensive Benchmark for Evaluating Bias in Large Vision-Language Model [72.13121434085116]
We introduce VLBiasBench, a benchmark to evaluate biases in Large Vision-Language Models (LVLMs)<n>VLBiasBench features a dataset that covers nine distinct categories of social biases, including age, disability status, gender, nationality, physical appearance, race, religion, profession, social economic status, as well as two intersectional bias categories: race x gender and race x social economic status.<n>We conduct extensive evaluations on 15 open-source models as well as two advanced closed-source models, yielding new insights into the biases present in these models.
arXiv Detail & Related papers (2024-06-20T10:56:59Z) - Evaluating Implicit Bias in Large Language Models by Attacking From a Psychometric Perspective [66.34066553400108]
We conduct a rigorous evaluation of large language models' implicit bias towards certain demographics.<n>Inspired by psychometric principles, we propose three attack approaches, i.e., Disguise, Deception, and Teaching.<n>Our methods can elicit LLMs' inner bias more effectively than competitive baselines.
arXiv Detail & Related papers (2024-06-20T06:42:08Z) - Sociodemographic Prompting is Not Yet an Effective Approach for Simulating Subjective Judgments with LLMs [13.744746481528711]
Large Language Models (LLMs) are widely used to simulate human responses across diverse contexts.<n>We evaluate nine popular LLMs on their ability to understand demographic differences in two subjective judgment tasks: politeness and offensiveness.<n>We find that in zero-shot settings, most models' predictions for both tasks align more closely with labels from White participants than those from Asian or Black participants.
arXiv Detail & Related papers (2023-11-16T10:02:24Z) - Social Bias Probing: Fairness Benchmarking for Language Models [38.180696489079985]
This paper proposes a novel framework for probing language models for social biases by assessing disparate treatment.
We curate SoFa, a large-scale benchmark designed to address the limitations of existing fairness collections.
We show that biases within language models are more nuanced than acknowledged, indicating a broader scope of encoded biases than previously recognized.
arXiv Detail & Related papers (2023-11-15T16:35:59Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.