👋 About Me

Hi👋！ I am Kaiyue Chen, an undergraduate student in Mechatronics and Robotic Systems at Xi’an Jiaotong-Liverpool University (XJTLU), Suzhou, China. My research focuses on Vision-Language-Action (VLA) and embodied manipulation.

At XJTLU, I work with Prof. Yaran Chen on a hierarchical Brain–Cerebellum–Supervisor framework. The brain outputs LLM quadruplets as a VLA-facing interface and is trained with RFT (beyond SFT) to support long-horizon and interruptible tasks.

At Tsinghua University, I collaborate with Prof. Rui Chen on ViTaMIn-B, a bimanual visuo-tactile data-collection system with DuoTact sensors and Meta Quest 3 tracking, and on integrating tactile sensing into VLA (pi0.5) for multimodal policies. I am also interested in deep reinforcement learning and LLM fine-tuning.

🔥 News

• 2026.06:  🥳 ManiSkill-ViTac Challenge 2026 was successfully completed.
• 2025.08:  🏆 National Finals, The 27th China Robot and Artificial Intelligence Competition — Baidu Service Robot track, coached by Prof. Eng Gee Lim (IEEE Fellow).
• 2025.05:  🎖 Received Xi’an Jiaotong-Liverpool University Excellent Student Cadre.
• 2024.09:  📚 Started BEng Mechatronics and Robotic Systems at XJTLU (expected 2028).

📝 Publications

Coming soon. 😁

🎖 Honors and Awards

• 2025.05, Excellent Student Cadre, Xi’an Jiaotong-Liverpool University.
• 2025.08, National Finals (team member), The 27th National Finals of China Robot and Artificial Intelligence Competition — Baidu Service Robot track.

💬 Research Experience

• 2026.04 - Present, Group Lead, Attribute-Aware Brain Module for Hierarchical VLA in Long-Horizon and Interrupted Manipulation Tasks, Xi’an Jiaotong-Liverpool University (Advisor: Prof. Yaran Chen).
  • Built an attribute-aware brain module that outputs LLM quadruplets directly for the VLA interface, and applied RFT (beyond SFT) within a Brain-Cerebellum-Supervisor framework to improve long-horizon execution and interruption recovery.
• 2026.03 - Present, Research Assistant, ViTaMIn-B: A Reliable and Efficient Visuo-Tactile Bimanual Manipulation Interface, Tsinghua University (Advisor: Prof. Rui Chen).
  • Contributed to a bimanual visuo-tactile data-collection system with DuoTact sensors, Meta Quest 3 pose tracking, and cross-modality synchronization; integrated tactile sensing into VLA (pi0.5) for multimodal policy learning.
• 2026.01 - 2026.03, Project Leader, Hierarchical Vision-Language-Action (VLA) Architecture for Long-Horizon Robotic Manipulation, Xi’an Jiaotong-Liverpool University.
  • Proposed a three-layer Brain-Cerebellum-Supervisor framework to decouple high-level planning, low-level control, and safety monitoring for long-horizon robot-arm tasks.
• 2025.03 - 2026.03, Research Assistant, Task Assignment and Navigation of AGV Clusters, Xi’an Jiaotong-Liverpool University (Advisor: Prof. Xinheng Wang).
  • Contributed to chassis design (SolidWorks), ROS2 + reinforcement learning integration, and Gazebo-based SLAM/navigation experiments.

💼 Internships

Sudo AI

2026.03 - Present · Intern

AI VLA Tactile

🧩 Projects

ManiSkill-ViTac Challenge 2026: Real-world Language-guided Bimanual Vision-Tactile Manipulation

CVPR 2026 Embodied AI Workshop · Student Organizer

CVPR 2026 TVLA

Project Page

💻 Competitions

• 2025.08, Baidu Service Robot Competition , Suzhou — National Finals of the 27th China Robot and Artificial Intelligence Competition; team coached by Prof. Eng Gee Lim.

📖 Educations

• 2024.09 - 2028.07 (expected), BEng Mechatronics and Robotic Systems, Xi’an Jiaotong-Liverpool University, Suzhou, China. GPA (Year 1): 3.80/4.0. Core courses include Linear Algebra, Calculus, C Programming and Software Engineering.