Scholar
Hejia Zhang
Google Scholar ID: h_0iAx4AAAAJ
Skild AI
Citations & Impact
Citations
399
H-index
11
i10-index
11
Publications
17
Co-authors
12
list available
Publications
Resume (English only)
Academic Achievements
- - Publications:
- * "ManipBench: Benchmarking Vision-Language Models for Low-Level Robot Manipulation", CoRL 2025
- * "Multi-Robot Geometric Task-and-Motion Planning for Collaborative Manipulation Tasks", AURO 2023
- * "PATO: Policy Assisted TeleOperation for Scalable Robot Data Collection", RSS 2023
- - Awards:
- * $2000 grant from Lambda's research grant program
- * $5000 grant from OpenAI Researcher Access Program
- - Projects:
- * Submitted to ICRA 2026: "Learning to Design and Control Hands with Human-Level Dexterity"
- * Accepted at ISRR 2024: "GPT-Fabric: Folding and Smoothing Fabric by Leveraging Pre-Trained Foundation Models"
- * Published in Journal of Industrial Safety: "A concept for integrating robotic systems in an underground roof support machine"
Research Experience
- - Researcher at Skild AI, focusing on embodied intelligence
- - Collaborated with Prof. Daniel Seita, Prof. Yue Wang, Prof. Joseph J. Lim, and Prof. Sven Koenig during Ph.D.
- - Main research on long-horizon, complex manipulation tasks in unstructured human environments
Education
- - Ph.D. in Computer Science, University of Southern California (USC), Advisor: Prof. Stefanos Nikolaidis
- - Master's degree, University of Southern California (USC), Advisors: Prof. Gaurav Sukhatme and Prof. Stefan Schaal
- - Bachelor's degree in Bioengineering, Zhejiang University (ZJU)
Background
- - Research Interests: learning and search, computational design, large-scale system evaluation
- - Field: Embodied intelligence
- - Brief Introduction: Currently a researcher at Skild AI, a startup co-founded by Prof. Deepak Pathak and Prof. Abhinav Gupta from CMU, focusing on building a general-purpose robot brain.
Miscellany
- - Personal Interest: Open to collaboration opportunities on enabling human-like dexterous robot manipulation
