Scholar
Parv Kapoor
Google Scholar ID: OpZzffgAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
82
H-index
4
i10-index
3
Publications
16
Co-authors
12
list available
Publications
Resume (English only)
Academic Achievements
- Publications:
- - STLCG++ accepted to IEEE Robotics and Automation Letters 2025, to be presented at ICRA 2026.
- - Vision-based safety for aerial robotics accepted to Robotics Science and Systems (RSS) 2025.
- - Learning behavior from system trajectories accepted to International Conference on Software Engineering (ICSE) 2025.
- - Logically Constrained Robotics Transformers for Enhanced Perception-Action Planning accepted to Robotics Science and Systems: Towards Safe Autonomy.
Research Experience
- Interned at Microsoft Research, working with Dr. John Langford on scaling long-horizon reasoning in large language models using Matryoshka representations; researcher with General Robotics, working with Dr. Ashish Kapoor on developing safe robot foundation models; worked with Dr. Jyotirmoy Deshmukh at CPS VIDA lab, University of Southern California; conducted research at Université Grenoble Alpes and Cardiff University.
Education
- Ph.D. student in Software Engineering at Carnegie Mellon University, School of Computer Science, Software and Societal Systems Department, advised by Dr. Eunsuk Kang and Dr. Sebastian Scherer.
Background
- Ph.D. student in Software Engineering, with research interests in safe and robust robot learning, leveraging diverse methodologies such as constrained robot foundation models, reinforcement learning, control barrier functions, and imitation learning. Recently, the focus has shifted towards large action models and embodied AI.
Miscellany
- Other:
- - Developed a custom deep learning library with implicit differentiation layers from scratch.
- - Conducted pilot user studies for trustworthy human-robot interaction.
- - Contributed to the simulation and the foundation model workstreams for the Scaled Foundations GRID platform.
- - Currently a part of AirLab and SoDA.
- - Gave a talk at NASA Ames Research Center and Toyota Infotech Labs.
