Portrait

Jiaqi Yin (殷家琦)

Undergraduate Researcher · Harbin Institute of Technology

I am an undergraduate student majoring in Artificial Intelligence at Harbin Institute of Technology. My research interests include embodied intelligence and robotics.

I am currently a visiting research student at Zhongguancun Academy, supervised by Prof. Ce Hao. During my gap year, I was a visiting student at SSR Group, Tsinghua SIGS (supervised by Prof. Wenbo Ding) and X-Institute (supervised by Prof. Wenzeng Zhang) in Shenzhen.

I am actively seeking PhD opportunities starting Fall 2027. If you think I could be a good fit for your group, please feel free to reach out! (Update: PhD advisor confirmed on 2026-05-14.)

Embodied Intelligence Robotics Artificial Intelligence

News


Contact

Email: yjqhit@gmail.com | 2078004110@qq.com


Research Projects

* Equally contributed, † corresponding author.

Teaser

PAIR-Former: Budgeted Relational MIL for miRNA Target Prediction

Venue/Year: arXiv 2026

Jiaqi Yin, Baiming Chen, Jia Fei, Mingjun Yang

SPARK Hand: Scooping-Pinching Adaptive Robotic Hand with Kempe Mechanism for Vertical Passive Grasp in Environmental Constraints

Venue/Year: IROS 2025

Jiaqi Yin, Tianyi Bi, Wenzeng Zhang

Teaser

Contactless Triboelectric Sensing for Real-Time 3D Motion Recognition in Human-Computer Interaction

Venue/Year: Advanced Electronic Materials 2025

Qinghao Xu, Junhao Gong, Jiayi Chen, Yimeng Zhang, Hongfa Zhao, Jiaqi Yin, Runze Zhao, Chuqiao Lyu, Wenbo Ding, Changsheng Wu

Teaser

LumiPane: Intelligent Interaction through Gesture Sensing and Ambient Light Communication

Venue/Year: IEEE GLOBECOM Workshops 2024

Liguang Ruan, Chenxin Liang, Jiaqi Yin, Jiarong Li, Xiaojun Liang, Wenbo Ding


Open Source

Octo UR5: Real-World Deployment of Octo on Universal Robots

A complete open-source pipeline for deploying the Octo embodied intelligence model on UR5/UR3 robots — simulation, teleoperation, data collection, finetuning, and distributed inference.

franka-control

franka-control: Franka Research 3 Control Wrapper for Embodied AI

A lightweight control library that decouples high-level policy (RL/IL/VLA) from 1 kHz FCI real-time control. Supports single-machine and dual-machine (GPU cluster + control PC) deployment via ZeroMQ. Provides a Gymnasium-style reset()/step()/close() interface, multiple interpolators (Ruckig, PT1, CubicSpline), joystick teleoperation, and independent gripper service.


Blog

See all posts →


Say Anything