Publications
Peer-reviewed publications in human-computer interaction and explainable AI.
Designing Effective Training Dataset Explanations: The Impact of Information Depth and Progressive Disclosure
Proceedings of the 31st International Conference on Intelligent User Interfaces (IUI '26), ACM.
Investigates how information depth and progressive disclosure in training dataset explanations affect user understanding, system assessments, and cognitive load. Findings show that users consistently favored comprehensive information over brevity, even at the cost of higher cognitive load.
doi.org/10.1145/3742413.3789087The Landscape of Digital Tech Disengagement Solutions for Early Adolescents: Insights from a Systematic Scoping Review and App Analysis
ACM Conference on Computer-Supported Cooperative Work and Social Computing (CSCW '25), ACM.
Presents a systematic scoping review of academic literature and an analysis of 47 apps to characterize existing tech-mediated digital disengagement solutions for early adolescents. Findings reveal that existing apps tend to prioritize restrictive measures, overlooking self-regulation and parental engagement, and highlight misalignments between current interventions and research recommendations.
dl.acm.org/doi/abs/10.1145/3757674Supporting User Critiques of AI Systems via Training Dataset Explanations: Investigating Critique Properties and the Impact of Presentation Style
2024 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC).
Examines how presentation style of training dataset explanations shapes the properties of user critiques of AI systems. Findings show that the same training data content, packaged differently, produces measurably different quality of user critiques, including their coverage, accuracy, and identification of AI bias.
doi.org/10.1109/VL/HCC60511.2024.00024Diversity Challenges in Recruiting for Human-Centered Explainable AI Studies
2024 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC).
Reflects on recruitment experiences across multiple user studies in a human-centered XAI project, examining challenges around participant diversity in terms of AI knowledge and experience. Discusses tradeoffs and strategies for recruiting diverse participants for both in-person and remote study contexts.
doi.org/10.1109/VL/HCC60511.2024.00067Data-Centric Explanations: Explaining Training Data of Machine Learning Systems to Promote Transparency
Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems (CHI '21), ACM.
Introduces data-centric explanations as a new type of AI transparency mechanism that surfaces training data provenance rather than model behavior. Findings establish training data explanations as a viable and effective approach to promoting transparency and supporting informed judgment of AI systems.
doi.org/10.1145/3411764.3445736Activity Recognition of a Badminton Game Through Accelerometer and Gyroscope
2016 19th International Conference on Computer and Information Technology (ICCIT), IEEE.
Proposes a sensor-based approach using accelerometer and gyroscope data with k-NN and SVM classifiers to recognize badminton game activities such as serve, smash, and forehand. Demonstrates a low-cost, faster alternative to vision-based approaches with a decent recognition rate.
doi.org/10.1109/ICCITECHN.2016.7860197Developing a Psychometrically Validated Dual-format Scale for Measuring AI Literacy
Manuscript under review.
Develops and validates a dual-format psychometric scale that separately measures self-assessed and demonstrated AI literacy. Findings reveal a systematic calibration gap between the two.