About

Joyce Chai is a pioneering researcher at the intersection of natural language processing, human-robot interaction, and grounded language understanding. Her work has fundamentally shaped how robots comprehend and respond to human language in real-world, physically situated environments — a challenge that sits at the heart of making robots genuinely useful collaborators. Chai's most celebrated contributions center on situated human-robot dialogue, particularly the problem of grounding language to a robot's internal representation of its physical surroundings. Her influential 2014 trilogy of papers — exploring action learning, collaborative effort toward common ground, and teaching robots through natural language — collectively reflect her insight that humans and robots must actively negotiate shared understanding rather than assume it. These works, accumulating over 200 combined citations, established foundational frameworks still referenced today. Her research has also tackled referring expression generation, grounded verb semantics, and more recently, leveraging large language models for 3D visual grounding, with her LLM-Grounder paper (2024) already garnering 60 citations. A co-author on a landmark 2021 roadmap for spoken language interaction with robots, Chai has helped define the broader field's research agenda. Her body of work equips the next generation of researchers with both theoretical grounding and practical tools for building robots that truly understand us.

Research Focus

Key Achievements

16
H-Index
30
Papers
914
Total Citations
30
Avg Citations/Paper
🏆 Most Cited Paper
Spoken language interaction with robots: Recommendations for future research
106 citations · 2021
📈 Most Prolific Year: 2014 (7 Papers)
🤝 Key Collaborators: 71
🏛 Institutions: University of Michigan–Ann Arbor, Michigan State University, Microsoft (United States), Universitas Dharmas Indonesia

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago