๐ค AI Summary
This work addresses the challenge of efficiently transforming UI prototypes into executable React applicationsโa task particularly difficult for designers and non-professional developers. We propose an end-to-end React program synthesis method driven by user demonstrations, integrating sketch editing with runtime interaction examples. Our approach leverages event trajectory analysis, reactive dataflow inference, and enumerative program synthesis, augmented by large language models (LLMs) to enhance semantic understanding and code generation. It automatically converts static interface sketches into fully functional React code with correct state management and interactive behavior. Crucially, we pioneer example-driven synthesis in frontend UI generation, enabling unified modeling of simple to moderately complex interfaces while seamlessly embedding into existing prototyping workflows. Experimental evaluation across diverse UI scenarios demonstrates its effectiveness, significantly lowering the frontend development barrier and improving the prototype-to-executable-code translation efficiency.
๐ Abstract
ReDemon UI synthesizes React applications from user demonstrations, enabling designers and non-expert programmers to create UIs that integrate with standard UI prototyping workflows. Users provide a static mockup sketch with event handler holes and demonstrate desired runtime behaviors by interacting with the rendered mockup and editing the sketch. ReDemon UI identifies reactive data and synthesizes a React program with correct state update logic. We utilize enumerative synthesis for simple UIs and LLMs for more complex UIs.