Skip to main content

Spring Deadline: Sunday, March 1 @ 11:59pm PT. Click here to apply.

Back to Research
Accepted to SynthData @ ICLR 2025

Deconstructing Bias: A Multifaceted Framework for Diagnosing Cultural and Compositional Inequities in Text-to-Image Generative Models

Muna Said, Aarib Haider, Rabia Usman, Sonia Okon

Abstract

The transformative potential of text-to-image (T2I) models hinges on their ability to synthesize culturally diverse, photorealistic images from textual prompts. However, these models often perpetuate cultural biases embedded within their training data, leading to systemic misrepresentations. This paper benchmarks the Component Inclusion Score (CIS), a metric designed to evaluate the fidelity of image generation across cultural contexts. Through extensive analysis involving 2,400 images, we quantify biases in terms of compositional fragility and contextual misalignment, revealing significant performance gaps between Western and non-Western cultural prompts.

Citation

Muna Said, Aarib Haider, Rabia Usman, Sonia Okon. "Deconstructing Bias: A Multifaceted Framework for Diagnosing Cultural and Compositional Inequities in Text-to-Image Generative Models". Accepted to SynthData @ ICLR 2025.

Details

Conference
Accepted to SynthData @ ICLR 2025
Authors
4 authors

Publish Your Research

Join Algoverse and work with world-class mentors to publish at top AI conferences.

Start Your Application