Latest News

A busy and exciting summer for the COCOA Lab

This summer brought new faces, fresh ideas, and plenty of collaboration. We welcomed several undergraduate research assistants, shared our work, and celebrated two new publications:

Together, these papers reflect our work in human-AI collaboration, AI-supported creative work, and thoughtful, human-centered design.

We also got out into the community and hosted activities that made room for learning and connection:

CG Connect

The lab headed to CG Connect, hosted by Autodesk Research in Toronto, to meet the wider graphics and interactive systems community. Read more here

GI Conference

Pengqi Wang presented a demo and position paper on AI proactivity at the GI conference in Waterloo. Read more here

Lunch & Learn

The lab hosted a Lunch & Learn to share ideas and connect with Dr. Sharan from the University of Hamburg. Read more here

CAIS Conference

Many of our undergraduate research assistants attended their first conference and presented posters at CAIS, learning from others in AI and HCI and connecting with the research community. Read more here

Bringing STEM to Life Symposium

Dr. Kuang joined the Bringing STEM to Life Symposium to see high school students’ projects addressing her research challenge: using AI to make education more accessible. Read more here

Thanks to everyone who made it such a lively and productive season!

Two new publications in Fall 2025!

Our lab has two new publications this semester:

Evaluating Usability Challenges in VR Games for Older Adults: A Comparison With and Without AI Assistance

In Adjunct Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology. Key highlights:

  • Older adults encountered several usability problems in VR, including unclear instructions and goals, confusing game mechanics, and difficult-to-navigate menus.
  • AI-generated suggestions helped UX evaluators to identify a greater number of usability problems.
LLM-powered assistant with electrotactile feedback to assist blind and low vision people with maps and routes preview

In the International Journal of Human-Computer Studies. Key highlights:

  • An LLM-powered assistant that supports natural language queries about map elements, route conditions, and trajectories.
  • A wearable electrotactile feedback device that conveys map features via haptic signals.
  • A user study with blind and low vision participants demonstrating improved efficiency and spatial understanding of route previews.