Xiaokun Yang

Large language models (LLMs) for code generation have achieved remarkable progress in synthesizing functional code from natural language instructions. However, a critical challenge persists in generating visually accurate and structurally sound front-end code that faithfully renders user-intended...