Rada Mihalcea

University of Michigan–Ann Arbor, Singapore University of Technology and Design

Papers

Total Citations

H-Index

About

Rada Mihalcea is a leading researcher at the intersection of computer vision, natural language processing, and human-robot interaction. Her work focuses on enabling machines to understand and interact with the world through multimodal learning, where visual and linguistic information are combined. A key contribution is her research on **semantic affordances**—the functional properties of objects that suggest possible actions. Her highly cited 2015 paper on mining these affordances from visual data (65 citations) laid groundwork for robots to recognize human activities and interact meaningfully with their environment. Mihalcea also advances **human-robot collaboration**, proposing a cyber-physical framework using reinforcement learning for personalized assistive training in manufacturing. More recently, she has explored **visual question answering (VQA)** , developing knowledge-enriched prompts to elevate multimodal language models. Her work bridges fundamental AI research with practical applications in robotics and assistive technologies, making her a notable figure in embodied AI and interactive systems.