Hugging Face 推出 WebSight 數據集：解鎖網頁截圖直接轉換為 HTML 程式碼的能力
Original: Unlocking the conversion of Web Screenshots into HTML Code with the WebSight Dataset
The Hugging Face official blog has published a post introducing WebSight, a brand-new open-source dataset designed to address the…
Hugging Face 宣布推出 WebSight 數據集，專為訓練視覺語言模型（VLM）進行「截圖轉網頁程式碼（Screenshot-to-Code）」而設計。該數據集包含約 200 萬個由合成技術產生的網頁截圖及其對應的乾淨 HTML/CSS 程式碼。透過 WebSight，開發者與研究人員能更有效率地微調多模態模型，加速自動化前端開發與設計稿轉程式碼的技術落地。
The Hugging Face official blog has published a post introducing WebSight, a brand-new open-source dataset designed to address the bottleneck that multimodal vision-language models (VLMs) face in the task of converting webpage screenshots into HTML/CSS code. In the past, enabling AI to accurately translate a webpage design image or screenshot into structurally complete, correctly styled web code has been a major challenge, primarily due to the lack of high-quality, license-clear, and uniformly formatted image-to-code alignment pairs.
Free shows the 3-line summary; Pro unlocks the full deep summary (~300 words) so you never have to click through.
See Pro plans →Want the original English / full article?
Read on Hugging Face Blog →Summaries are AI-generated; the original article is authoritative.