Papers
46
Total Citations
1,724
H-Index
19
About
Harold Soh is an Assistant Professor whose research sits at the intersection of human-robot interaction, machine learning, and robot perception. He is perhaps best known for his pioneering work on trust in robotics, exploring how humans form, calibrate, and transfer trust toward autonomous systems. His computational models of trust — notably those grounded in partially observable Markov decision processes — have fundamentally advanced how robots can make trust-aware decisions during collaboration, earning him over 400 citations across related works alone. Beyond trust dynamics, Soh has made notable contributions to tactile sensing, developing the neuromorphic NeuTouch sensor and event-driven visual-tactile learning frameworks that push the boundaries of robot perception. His more recent work leverages large language models as zero-shot human models for interaction, reflecting his keen interest in bridging AI advances with real-world robotics. Notably, his social science-informed research — examining how perceived robot feelings affect forgiveness after failures, and how robot supervisors trigger worker spite — has attracted significant interdisciplinary attention, with one paper alone amassing over 300 citations. Soh's body of work reflects a rare and valuable blend of rigorous engineering and human-centered thinking.
Research Focus
Key Achievements
Top Papers
- 1
- 2Trust in Robots: Challenges and Opportunities203 citations · 2020
- 3Trust-Aware Decision Making for Human-Robot Collaboration137 citations · 2020
- 4Event-Driven Visual-Tactile Sensing and Learning for Robots118 citations · 2020
- 5
- 6Multi-task trust transfer for human–robot interaction83 citations · 2019
- 7Large Language Models as Zero-Shot Human Models for Human-Robot Interaction73 citations · 2023
- 8Planning with Trust for Human-Robot Collaboration60 citations · 2018
- 9
- 10Robot Capability and Intention in Trust-Based Decisions Across Tasks55 citations · 2019