{"ID":2874212,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.05394","arxiv_id":"2509.05394","title":"Reverse Browser: Vector-Image-to-Code Generator","abstract":"Automating the conversion of user interface design into code (image-to-code or image-to-UI) is an active area of software engineering research. However, the state-of-the-art solutions do not achieve high fidelity to the original design, as evidenced by benchmarks. In this work, I approach the problem differently: I use vector images instead of bitmaps as model input. I create several large datasets for training machine learning models. I evaluate the available array of Image Quality Assessment (IQA) algorithms and introduce a new, multi-scale metric. I then train a large open-weights model and discuss its limitations.","short_abstract":"Automating the conversion of user interface design into code (image-to-code or image-to-UI) is an active area of software engineering research. However, the state-of-the-art solutions do not achieve high fidelity to the original design, as evidenced by benchmarks. In this work, I approach the problem differently: I use...","url_abs":"https://arxiv.org/abs/2509.05394","url_pdf":"https://arxiv.org/pdf/2509.05394v1","authors":"[\"Zoltan Toth-Czifra\"]","published":"2025-09-05T11:13:40Z","proceeding":"cs.SE","tasks":"[\"cs.SE\",\"cs.AI\"]","methods":"[]","has_code":false}