Iaroslav V. Ponomarenko

I specialize in embodied AI and robotics, conducting research across two leading institutions. At Peking University's Center on Frontiers of Computing Studies, I am a second-year master's student in Computer Science working under Professor Hao Dong. In parallel, I serve as a visiting student researcher at the Department of Robotics at Mohamed Bin Zayed University of Artificial Intelligence (MBZUAI), mentored by Professor Yoshihiko Nakamura.

Previously, I obtained two earlier technical degrees: an Engineering degree in Information Systems and Technologies from the Voronezh Institute of High Technologies, and a Technician degree in Automated Information Processing and Control Systems from Borisoglebsk College of Informatics and Computer Engineering.

Google Scholar  /  LinkedIn  /  GitHub


News
  • — ManipGPT [P5] has been accepted for publication at IROS 2025.
  • — Started as Visiting Student in the Department of Robotics at MBZUAI (Abu Dhabi, UAE).
  • — Concluded a 1-year Research Internship at AGIBot (Beijing, China).
  • — CrayonRobo [P4] has been accepted for publication at CVPR 2025.
  • — SpatialBot [P3] has been accepted for publication at ICRA 2025.
  • — Presented ManipVQA [P2] at IROS 2024, Certificate of Attendance (Abu Dhabi, UAE).
  • — Presented ManipVQA [P2] at Microsoft Research Asia Summer Tech Fest (Beijing, China).
  • — ManipVQA [P2] has been accepted for publication at IROS 2024.
  • — Began a Research Internship at AGIBot (Beijing, China).
  • — Commenced Master's studies in Computer Science at Peking University (Beijing, China).
  • — Concluded 3-month Visiting Student appointment at Peking University (Beijing, China).
  • — Started as Visiting Student at the School of Computer Science, Peking University (Beijing, China).

Research Focus

My research centers on embodied AI, visual perception, reasoning, and robotic control. I explore how embodied agents can acquire environmental awareness through vision, with a focus on affordance understanding [P1, P2, p1, P5] and spatial reasoning [P3], enabling them to perform complex manipulation tasks.

Currently, I'm inspired by questions at the intersection of cognitive science and embodied intelligence—how can we build embodied agents that not only see and act, but also understand? I aspire to develop systems that grasp the flow of time, the cause and effect behind their actions, and the intentions of others—bringing machines closer to human-like perception and reasoning. Through these lenses, my goal is to push the boundaries of what embodied agents can do in dynamic, real-world environments.

Service
  • Reviewer:
    • Conference on Robotics: Science and Systems (RSS 2025).
    • IEEE International Conference on Robotics and Automation (ICRA 2025).
  • Student Member, Chinese Association for Artificial Intelligence (CAAI 2024–2029), Certificate.
  • Teaching Assistant, Fundamentals of Artificial Intelligence, Peking University (Spring 2024).

Publications

(*) indicates equal contribution, (†) denotes the corresponding author, and `highlighted` papers are marked with special recognition

[P5]
Title: ManipGPT – Is Affordance Segmentation by Large Vision Models Enough for Articulated Object Manipulation?

Authors: Taewhan Kim, Hojin Bae, Zeming Li, Xiaoqi Li, Iaroslav Ponomarenko, Ruihai Wu,
Hao Dong†

Conference: Conference on Intelligent Robots and Systems (IROS), 2025
Links: View on arXiv

[P4]
Title: CrayonRobo – Object-Centric Prompt-Driven Vision-Language-Action Model for Robotic Manipulation

Authors: Xiaoqi Li, Lingyun Xu, Mingxu Zhang, Jiaming Liu, Yan Shen, Iaroslav Ponomarenko,
Jiahui Xu, Liang Heng, Siyuan Huang, Shanghang Zhang, Hao Dong†

Conference: Conference on Computer Vision and Pattern Recognition (CVPR), 2025
Links: View on arXiv

[P3]
Title: SpatialBot – Precise Spatial Understanding with Vision Language Models

Authors: Wenxiao Cai*, Iaroslav Ponomarenko*, Jianhao Yuan, Xiaoqi Li, Wankou Yang,
Hao Dong, Bo Zhao†

Conference: International Conference on Robotics and Automation (ICRA), 2025
Links: Paper | GitHub GitHub Repo stars

[P2]
Title: ManipVQA – Injecting Robotic Affordance and Physically Grounded Information into Multi-Modal Large Language Models

Authors: Siyuan Huang*, Iaroslav Ponomarenko*, Zhengkai Jiang, Xiaoqi Li, Xiaobin Hu,
Peng Gao, Hongsheng Li, Hao Dong†

Featured: [Oral Pitch and Interactive Presentation]
Conference: Conference on Intelligent Robots and Systems (IROS), 2024
Links: View on arXiv | Oral Pitch (YouTube) | Slides | Poster | GitHub GitHub Repo stars

[P1]
Title: Learning Part-Aware Visual Actionable Affordance for 3D Articulated Object Manipulation

Authors: Yuanchen Ju*, Haoran Geng*, Ming Yang*, Yiran Geng, Yaroslav Ponomarenko,
Taewhan Kim, He Wang, Hao Dong†

Featured: [Spotlight Presentation]
Conference: Workshop at the Conference on Computer Vision and Pattern Recognition (CVPR @ 3DVR), 2023
Links: Paper

Preprints
(*) indicates equal contribution, and (†) denotes the corresponding author

[p1]
Title: ImageManip – Image-Based Robotic Manipulation with Affordance-Guided Next View Selection

Authors: Xiaoqi Li, Yanzi Wang, Yan Shen, Iaroslav Ponomarenko, Haoran Lu, Qianxu Wang,
Boshi An, Jiaming Liu, Hao Dong†

Release: Preprint, 2023
Links: View on arXiv

Early Research & Publications
(†) indicates the corresponding author

[EP4]

 Sukhanov Alexey, Iaroslav Ponomarenko. “Application of Block Periodization in the Design of Health-Prolonging Training Cycles” (2017) Proceedings of Students and Young Scientists of Russian State University of Physical Education, Sport, Youth and Tourism, 354: 275-279. Paper.

[EP3]

 Sukhanov Alexey, Iaroslav Ponomarenko, Rubin Vladimir†. “The Potential of Instrumental Methods for Medical Soft Tissues Diagnostics in Physical Education and Health-Improving Training” (2016) Fitness-Aerobics-2016, 226: 97-98. Paper.

[EP2]

 Sukhanov Alexey, Iaroslav Ponomarenko, Rubin Vladimir†. “The Study of the Relationships of Methodological Methods Aimed at Inter-Muscular Coordination and Strength Abilities Development Among Women in the First Maturity Period in Health-Improving Training” (2016) Fitness-Aerobics-2016, 226: 98-100. Paper.

[EP1]

 Sukhanov Alexey, Iaroslav Ponomarenko. “Assessment of the Muscle Condition as One of the Physical Health Indicators in the Framework of Physical Education and Health-Improving Training” (2016) Proceedings of Students and Young Scientists of Russian State University of Physical Education, Sport, Youth and Tourism, 279: 78-80. Paper.
Note: Prior research in sports science on training methodologies and physiological diagnostics provides context for my background but has limited direct relevance to my current work in embodied AI.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Last updated on Monday, June 16, 2025, at 02:30:11 PM.

Design inspired by Jon Barron's website.