{"ID":2862153,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.01010","arxiv_id":"2510.01010","title":"ImageDoctor: Diagnosing Text-to-Image Generation via Grounded Image Reasoning","abstract":"The rapid advancement of text-to-image (T2I) models has increased the need for reliable human preference modeling, a demand further amplified by recent progress in reinforcement learning for preference alignment. However, existing approaches typically quantify the quality of a generated image using a single scalar, limiting their ability to provide comprehensive and interpretable feedback on image quality. To address this, we introduce ImageDoctor, a unified multi-aspect T2I model evaluation framework that assesses image quality across four complementary dimensions: plausibility, semantic alignment, aesthetics, and overall quality. ImageDoctor also provides pixel-level flaw indicators in the form of heatmaps, which highlight misaligned or implausible regions, and can be used as a dense reward for T2I model preference alignment. Inspired by the diagnostic process, we improve the detail sensitivity and reasoning capability of ImageDoctor by introducing a \"look-think-predict\" paradigm, where the model first localizes potential flaws, then generates reasoning, and finally concludes the evaluation with quantitative scores. Built on top of a vision-language model and trained through a combination of supervised fine-tuning and reinforcement learning, ImageDoctor demonstrates strong alignment with human preference across multiple datasets, establishing its effectiveness as an evaluation metric. Furthermore, when used as a reward model for preference tuning, ImageDoctor significantly improves generation quality -- achieving an improvement of 10% over scalar-based reward models.","short_abstract":"The rapid advancement of text-to-image (T2I) models has increased the need for reliable human preference modeling, a demand further amplified by recent progress in reinforcement learning for preference alignment. However, existing approaches typically quantify the quality of a generated image using a single scalar, lim...","url_abs":"https://arxiv.org/abs/2510.01010","url_pdf":"https://arxiv.org/pdf/2510.01010v1","authors":"[\"Yuxiang Guo\",\"Jiang Liu\",\"Ze Wang\",\"Hao Chen\",\"Ximeng Sun\",\"Yang Zhao\",\"Jialian Wu\",\"Xiaodong Yu\",\"Zicheng Liu\",\"Emad Barsoum\"]","published":"2025-10-01T15:15:55Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Reinforcement Learning\",\"Language Model\"]","has_code":false}