Xinyue Chen awarded Rackham Barbour Scholarship

Awarded to women of high academic achievement from Asian countries, the Barbour Scholarship will support Chen’s research on the use of AI in human-human communication.
Xinyue Chen
Xinyue Chen

CSE PhD student Xinyue Chen has been selected to receive the prestigious Barbour Scholarship as part of the Rackham Predoctoral Fellowship program. Endowed in 1917 by Levi Lewis Barbour, the Barbour Scholarship program is intended to provide support for “women of the highest academic and professional caliber” from countries in Asia who are pursuing PhDs in STEM disciplines. 

With the support of this scholarship, Chen will continue her dissertation research exploring the role of artificial intelligence (AI) in human-human communication and collaboration. Specifically, her work looks at new ways to incorporate AI assistance and proposes novel human-AI interaction methods to support synchronous collaboration and communication.

Recent technological advancements, hastened in part by the Covid-19 pandemic, have led to an increasing use of synchronous communication platforms, such as FaceTime and Zoom, especially at the workplace and in educational settings. While these platforms have developed rapidly, they have led to increased misunderstanding and suboptimal collaboration outcomes due to insufficient grounding among conversation partners.  

Extraordinary leaps in AI have led to new tools for enabling greater understanding in synchronous meetings; AI applications have shown great promise in generating meeting minutes and assisting in brainstorming, to give two examples. But the implementation of these AI-mediated communication tools has not been without challenges, as overreliance on AI risks compromising human creativity and engagement, as well as creating opportunities for potential bias and error.

To gain a better understanding of AI’s role in human-human communication in synchronous meetings and other channels, Chen is working first to investigate and define the role of AI in human communication. She has also developed AI-assisted interactive systems to promote  collaboration and understanding in synchronous meetings.

For example, Chen developed MeetScript, a system that provides parallel participation channels through real-time interactive transcripts, leveraging social annotation techniques and information filtering mechanisms. Following this work, she developed MeetMap, a system using LLMs to create dialogue maps in real-time to visually structure and connect ideas to support collaborative sense-making. The studies of MeetScript and MeetMap show that the use of AI to enhance human-human communication requires carefully designed human-AI interaction methods, where users can easily express their intents, AI provides just the right amount of support, and AI does not distract users or replace their valuable cognitive efforts. 

“Xinyue does not go after easy problems,” said Prof. Xu Wang, Chen’s advisor and nominator. “Her dissertation topic on using AI to support human-human communication is uniquely challenging in that it connects several disciplines including communication, human-computer interaction, and cognitive science. Moreover, human collaboration processes are uniquely nuanced and complex. She has grown tremendously as a researcher in the past three years and I’m keen to see what she’ll come up with next.”

Chen’s research has resulted in several published papers, including one that was an honorable mention for Best Paper Award at CSCW, a top conference in human-computer interaction. In addition to her research, Chen is a dedicated mentor for undergraduate students and is committed to promoting diversity, equity, and inclusion in computer science and beyond.