Xinyue Chen

In knowledge-intensive tasks such as analysis, writing, learning, and collaboration, people engage in ongoing sense-making—interpreting information and refining understanding as they work and learn. As AI becomes part of these cognitive processes, a central question arises: How can it assist without weakening the human sense-making they depend on?

My research addresses this question by designing and studying AI systems that scaffold human sense-making in both individual and collaborative contexts:

• In collaborative work, such as meetings and co-writing, I design AI-mediated systems that visualize and scaffold human–human common ground, making shared understanding visible and helping teams repair and sustain it as ideas evolve.
• In individual tasks, I study how AI can support human–AI alignment and preserve human agency during thinking processes. To this end, I propose the concept of intermediate AI design, which empirically and theoretically models how AI can augment human reasoning.

My work has been published in premier venues including: CHI, CSCW, DIS, and AIED , and has received several Best Paper Honorable Mention Awards. Together, my work contributes to a broader vision of AI systems that support human cognition, advance mechanisms for human–AI alignment and AI-mediated common ground, helping people and teams think, learn, and coordinate more effectively with and through AI.