Intent-AI Interaction

Intent-AI Interaction explores a paradigm shift in how humans interact with AI systems. Instead of requiring users to provide detailed, step-by-step instructions, this research investigates how AI systems can understand and respond to high-level user intents and conceptual goals.

Key contributions include:

• Direct Intent Manipulation — Drag-based interfaces for research ideation, evaluation, and evolution (UIST 2025)
• Design Space Mapping — Charting the design space of human-AI interactions through task creativity and complexity (IUI 2024 DC)
• Intent-Based UI — Frontend Diffusion for exploring intent-based user interfaces (VL/HCC 2025)

This line of research was featured in an invited talk at UMD HCIL in November 2024: “Intent-AI Interaction: Elevating Human-Computer Interaction to the Intent and Conceptual Level.”