Context Matters: Auditing Gender Bias in T2I Generation through Risk-Tiered Use-Case Profiles

TL;DR

This paper presents a comprehensive, context-aware framework for auditing gender bias in text-to-image models, integrating risk profiles, evaluation metrics, and harm typologies aligned with deployment scenarios.

Contribution

It introduces a risk-tiered auditing framework, a consolidated metric catalog, and THUMB cards for systematic bias assessment tailored to use-case contexts.

Findings

Aligns gender bias metrics with EU AI Act risk categories.

Provides a structured approach for context-aware bias evaluation.

Introduces THUMB cards to guide systematic auditing.

Abstract

Text-to-image (T2I) generative models are increasingly used to produce content for education, media, and public-facing communication, and are starting to be integrated into higher-impact pipelines. Since generated images tend to reinforce stereotypes, producing representational erasure via "default" depictions and shaping perceptions of who belongs in certain roles, a growing body of work has proposed metrics to quantify gender bias in T2I outputs. Yet existing evaluations remain fragmented. Metrics are often reported without a shared view of what they measure, what assumptions they entail, or how their results should be interpreted under different deployment contexts. This limits the usefulness of gender bias measurement for both technical auditing and emerging governance discussions. We propose a risk-aligned auditing framework for gender bias in T2I models composed of three constituents that connects risk categories, evaluation metrics, and harms. First, we identify risk-tiered use-case profiles aligned with the EU AI Act's risk categories to motivate why auditing expectations may vary with deployment contexts and stakeholder exposure. Second, we construct a metric catalog that consolidates gender-bias evaluation methods and organizes them in three measurement categories: gender prediction, embedding similarity, and downstream task. Third, we introduce a harm typology that maps context-dependent harm categories (e.g., representational, quality-of-service) to specific risk-tired scenarios. Finally, we introduce THUMB cards (Text-to-image Harms-informed Use-case-aligned Metrics of gender Bias) that help formulate auditing systematically by the incorporation of context, scenario and bias manifestation, harm hypotheses, and audit strategy.