@inproceedings{zhang-etal-2025-strict,
    title = "{STRICT}: Stress-Test of Rendering Image Containing Text",
    author = "Zhang, Tianyu  and
      Wang, Xinyu  and
      Li, Lu  and
      Tai, Zhenghan  and
      Chi, Jijun  and
      Tian, Jingrui  and
      He, Hailin  and
      Wang, Suyuchen",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.1070/",
    pages = "21148--21161",
    ISBN = "979-8-89176-332-6",
    abstract = "While diffusion models have revolutionized text-to-image generation with their ability to synthesize realistic and diverse scenes, they continue to struggle with generating consistent and legible text within images. This shortcoming is commonly attributed to the locality bias inherent in diffusion-based generation, which limits their capacity to model long-range spatial dependencies. In this paper, we introduce \textbf{STRICT}, a benchmark designed to systematically stress-test the ability of diffusion models to render coherent and instruction-aligned text in images. Our benchmark evaluates models across multiple dimensions: (1) the maximum length of readable text that can be generated and (2) the correctness and legibility of the generated text. We assess several state-of-the-art models, including proprietary and open-source variants, and reveal persistent limitations in long-range consistency and instruction-following capabilities. Our findings provide insights into architectural bottlenecks and motivate future research directions in multimodal generative modeling."
}