{"ID":2878983,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.17472","arxiv_id":"2508.17472","title":"T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation","abstract":"We propose T2I-ReasonBench, a benchmark evaluating reasoning capabilities of text-to-image (T2I) models. It consists of four dimensions: Idiom Interpretation, Textual Image Design, Entity-Reasoning and Scientific-Reasoning. We propose a two-stage evaluation protocol to assess the reasoning accuracy and image quality. We benchmark various T2I generation models, and provide comprehensive analysis on their performances.","short_abstract":"We propose T2I-ReasonBench, a benchmark evaluating reasoning capabilities of text-to-image (T2I) models. It consists of four dimensions: Idiom Interpretation, Textual Image Design, Entity-Reasoning and Scientific-Reasoning. We propose a two-stage evaluation protocol to assess the reasoning accuracy and image quality. W...","url_abs":"https://arxiv.org/abs/2508.17472","url_pdf":"https://arxiv.org/pdf/2508.17472v1","authors":"[\"Kaiyue Sun\",\"Rongyao Fang\",\"Chengqi Duan\",\"Xian Liu\",\"Xihui Liu\"]","published":"2025-08-24T17:59:38Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}