Ha Le
I build human-in-the-loop wearable sensing systems for personalized activity recognition, advised by Prof. Stephen Intille and Prof. Varun Mishra at Northeastern University.
About
I am a PhD candidate at Khoury College of Computer Sciences, Northeastern University, working at the intersection of Human-Computer Interaction, Ubiquitous/Wearable Computing, and Personal Health Informatics. My work has been recognized with Northeastern University's Outstanding PhD Student Award in Humanics and the Apple AI/ML PhD Fellowship 2026.
Most commercial activity-tracking systems rely on passive sensing and machine learning models trained on lab-collected datasets — and often fail to generalize to real-world settings and across individuals. I build personalized, human-in-the-loop tracking and measurement systems that (1) collect high-quality behavioral labels to capture each person's habits and support personalized recognition models, and (2) let users provide feedback with minimal effort.
I received my Bachelor's degree in Mathematics and Computer Science from Gustavus Adolphus College in 2022, where I did research under Prof. Louis Yu, Jillian Downey, and Jacob Siehler.
Research focus
Developing novel multimodal tracking systems using both in-situ methods (IMWUT'24, CHI'25, PervasiveHealth'24) and retrospective recall methods (IMWUT'25).
Studying users' uncertainty when self-reporting their behaviors, and how to incorporate that uncertainty into the tracking process (VIS'25, IMWUT'25).
Investigating how AI/LLMs can support the extraction of high-quality behavioral labels and enable low-effort tracking for health and wellbeing (Ubicomp'25, IMWUT'25).
Publications
Feasibility of using a multi-agent LLM system to correct annotations and support low-effort activity labeling
IMWUT'26
DAIMON: Designing AI-Augmented Research Dashboards to Enable Novel Human-AI Collaborative Workflows in Longitudinal Sensing Studies
IMWUT'26
Generating Personalized Games with Sensing Data for Longitudinal Data Collection
UIST'26
A Multi-Agent LLM Network for Suggesting and Correcting Human Activity and Posture Annotations
GenAI4HS Workshop @ Ubicomp'25
GLOSS: Group of LLMs for Open-Ended Sensemaking of Passive Sensing Data for Health and Wellbeing
IMWUT'25
An Evaluation of Temporal and Categorical Uncertainty on Timelines: A Case Study in Human Activity Recall Visualizations
IEEE VIS'25
Collecting Self-reported Physical Activity and Posture Data using Audio-based Ecological Momentary Assessment
IMWUT'24
Detecting Sleep Disruptions in Adolescents Using Context-sensitive Ecological Momentary Assessment: A Feasibility Study
PervasiveHealth'24
mHealth-Based Just-in-Time Adaptive Intervention to Improve the Physical Activity Levels of Individuals With Spinal Cord Injury: Protocol for a Randomized Controlled Trial
JMIR Research Protocols
A Feasibility Study on the Use of Audio-based Ecological Momentary Assessment with Persons with Aphasia
ASSETS'23
Relating Consistent Improvement to Overall Performance in a Calculus I Course that Utilizes Standards-Based Grading
PRIMUS