Hi there 👋🏼👩🏻💻
Master Student in HCI, University of Washington
I am currently pursuing a Master of Science in Technology Innovation (Connected Devices) at the University of Washington, Seattle. I received my Bachelor's degree in Communication with minors in Informatics and Data Science from the University of Washington, where I was a three-time Annual Dean's List recipient.
Currently, I am a Research Intern at the AI4SG Lab at the National University of Singapore (NUS), supervised by Dr. Yi-Chieh (EJ) Lee and Tianqi (Holly) Song. I also serve as a Research Assistant at the ACE Lab (University of Washington), advised by Dr. Jacob O. Wobbrock, Dr. James A. Fogarty, and Mingyuan Zhong. Previously, I have research experiences at the Department of Human Centered Design and Engineering (HCDE), and gained industry experience as a User Research Intern at Xiaomi Inc.
My research interests include human-computer interaction related to accessibility and responsible AI. I am passionate about combining empirical studies of human behavior with the design of inclusive, human-centered technologies to lower barriers to participation and address overlooked social and emotional needs. Additionally, I am interested in how digital platforms and social media influence collective meaning-making and public perception.
I enjoy engaging in interdisciplinary collaborations that bridge the gap between social computing and physical prototyping. Feel free to email me if you are interested in collaborating, discussing research ideas, or just making friends! :)
The core of my work centers on three research directions. Click each card for an expanded page with project context and related papers.
I study how interactive systems can lower physical, social, and emotional barriers, from accessible educational games to support-oriented AI experiences for older adults.
I examine how safety mechanisms, user agency, and social context shape emotionally supportive uses of conversational AI, especially for older adults.
I explore how conversational agents and educational tools can support older adults in learning about AI, and how computing education research can inform better learning design.
Selected publications and manuscripts in progress.
[IJHCS - FULL PAPER]
TL;DR: This work examines how conversational agents can support older adults in learning about AI. Across interviews and a comparative study, the findings suggest that peer-like agents can better foster engagement and support confidence, pointing to design opportunities for AI literacy tools tailored to older adults.
[UIST 2026 - POSTER]
TL;DR: Large language models (LLMs) show promise for web accessibility evaluation, but they still produce many false positives and false negatives on W3C's ACT rules, mainly due to missing evidence, weak criterion support, and ungrounded semantic judgments. We propose interactive tools that gather evidence like focus and error messages, paired with atomic rubrics that guide LLM agents on WCAG criteria. This approach achieves much higher recall (99.1% vs. 68.7%) and precision (91.2% vs. 68.1%) than the strongest baseline.
[DIS 2026 - FULL PAPER]
TL;DR: Interviews with 18 older adults show that they often turn to conversational AI for emotional support when human support feels inaccessible, but safety interventions such as redirection or interruption can feel abrupt, undermine emotional pacing, and reduce users' sense of agency. The work argues for emotionally supportive AI systems whose safety mechanisms better align with older adults' social contexts and lived experiences.
[CHI 2026 - POSTER]
TL;DR: Older adults sometimes prefer AI over friends/family for emotional support, not out of general preference, but in specific moments: when loved ones are unavailable, to avoid feeling like a burden, or to protect their dignity. Interviews with 18 older adults reveal how concerns around independence and self-image drive these situational choices, with implications for designing more context-aware AI companions.
[CHI 2026 WORKSHOP - POSITION PAPER]
TL;DR: This position paper frames emotionally supportive AI as a relational, co-adaptive process rather than a one-sided interaction, highlighting tensions around authenticity, emotional reciprocity, and how users' expectations evolve over time.
[ICER 2023 - FULL PAPER]
TL;DR: This work synthesizes the concept inventory literature in computing education through a review of 65 papers and large-scale dataset analysis, surfacing major trends, gaps, and methodological patterns in how conceptual understanding is assessed.
Selected research roles and projects.
AI4SG Lab
National University of Singapore
Advisors Yi-Chieh (EJ) Lee, Tianqi (Holly) Song
ACE Lab
University of Washington
Advisors Jacob O. Wobbrock, James A. Fogarty, Mingyuan Zhong
Department of Human Centered Design & Engineering
University of Washington
Advisors Sayamindu Dasgupta, Murtaza Ali
Academic training and degree milestones.
09/2025 - 03/2027
09/2021 - 06/2025
Selected teaching and academic service activities.
A small corner for side projects and everyday interests.