{"ID":2899180,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.03015","arxiv_id":"2507.03015","title":"Beyond Overcorrection: Evaluating Diversity in T2I Models with DivBench","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.","short_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- a...","url_abs":"https://arxiv.org/abs/2507.03015","url_pdf":"https://arxiv.org/pdf/2507.03015v2","authors":"[\"Felix Friedrich\",\"Thiemo Ganesha Welsch\",\"Manuel Brack\",\"Patrick Schramowski\",\"Kristian Kersting\"]","published":"2025-07-02T13:14:42Z","proceeding":"cs.CL","tasks":"[\"cs.CL\",\"cs.CY\",\"cs.LG\"]","methods":"[\"Diffusion Model\",\"Large Language Model\"]","has_code":false}