Portrait of Kwan Man Nick Cheng

Kwan Man "Nick" Cheng

You can also call me 钧文

Email LinkedIn Google Scholar

About Me

I am an M.S. Robotics student at Carnegie Mellon University's Robotics Institute, advised by Prof. Fernando De la Torre.

My research focuses on vision-language learning and robotics. I am particularly interested in adapting pretrained foundation models to downstream tasks through post-training and improving agentic workflows.

I received my B.S. in Computer Science from the University of Wisconsin–Madison, where I was fortunate to be supervised by Prof. Yin Li. I began my undergraduate studies at Beijing Normal University and spent a year at UC Berkeley as an exchange student before transferring to UW–Madison.

Selected Publications

* Equal contribution

DRIFT: A Residual Flow Adapter for Decoding Continuous Outputs in Vision-Language Models

Zhuoming Liu*, Jinhong Lin*, Kwan Man Cheng*, Lin Zhang, Shayok Bagchi, Yin Li

Preprint, 2026

[Paper][Project]

SkillWeaver: Agentic Exploration over Neural Interaction Skills for Scalable Robot Data Generation

He Zhu, Lusen Zhao, Kwan Man Cheng, Su Li, Katerina Fragkiadaki

CoRL, 2026

[Paper][Project]

An Extended VIIRS-like Artificial Nighttime Light Data Reconstruction (1986–2024)

Yihe Tian*, Kwan Man Cheng*, Zhengbo Zhang, Tao Zhang, Junning Feng, Zhehao Ren, Suju Li, Dongmei Yan, Bing Xu

Scientific Data, 2026

[Paper][Data]

Modular-Cam: Modular Dynamic Camera-view Video Generation with LLM

Zirui Pan, Xin Wang, Yipeng Zhang, Hong Chen, Kwan Man Cheng, Yaofei Wu, Wenwu Zhu

AAAI, 2025

[Paper][Project]

Experience

Carnegie Mellon University · Robotics Institute

Summer 2026

Research Assistant · Prof. Katerina Fragkiadaki

Developed agentic simulation pipelines for SkillWeaver and hierarchical robot policies combining VLM reasoning with VLA control.

University of Wisconsin–Madison · Video Lab

Jan 2025 – May 2026

Research Assistant · Prof. Yin Li

Co-developed DRIFT for continuous robot control and visual grounding, and built a real-time ergonomic risk assessment system for General Motors.

Qualcomm

Sep – Dec 2025

Machine Learning Engineer · Capstone Project

Built a multimodal presentation coach and deployed it on a Snapdragon NPU using ONNX.

Tsinghua University

May 2024 – May 2025

Research Assistant · Prof. Wenwu Zhu and Prof. Bing Xu

Worked on controllable video generation with Modular-Cam and nighttime-light data reconstruction with EVAL.

Quwan Technology

Nov 2024 – Jan 2025

Algorithm Intern

Fine-tuned text-to-video models and built data pipelines and parallel inference infrastructure.