Beyond Overcorrection: Evaluating Diversity in T2I Models with DIVBENCH
Journal:
arXiv
Published Date:
Jul 2, 2025
Abstract
Current diversification strategies for text-to-image (T2I) models often
ignore contextual appropriateness, leading to over-diversification where
demographic attributes are modified even when explicitly specified in prompts.
This paper introduces DIVBENCH, a benchmark and evaluation framework for
measuring both under- and over-diversification in T2I generation. Through
systematic evaluation of state-of-the-art T2I models, we find that while most
models exhibit limited diversity, many diversification approaches overcorrect
by inappropriately altering contextually-specified attributes. We demonstrate
that context-aware methods, particularly LLM-guided FairDiffusion and prompt
rewriting, can already effectively address under-diversity while avoiding
over-diversification, achieving a better balance between representation and
semantic fidelity.
