Xiang Shaowen

Xiang Shaowen

Ph.D. Student @ UMich ECE
DeepThink Lab & PixAIL

Xiang Shaowen

Personal Academic Page

Xiang Shaowen

Ph.D. Student @ University of Michigan ECE · Co-advised by Prof. Qing Qu & Prof. Joyce Wang

I am a Ph.D. student in Electrical and Computer Engineering at the University of Michigan, working at the intersection of AI for science, machine learning, intelligent optical networks, and medical AI reliability. I am co-advised by Prof. Qing Qu (DeepThink Lab) and Prof. Joyce Yan-Ran Wang (PixAIL).

Research Publications CV Email

About

I received my B.Eng. in Electronic Science & Technology from Shanghai Jiao Tong University (Zhiyuan Honors Program) in June 2026 (core GPA 3.85/4.3, rank 10/64), and was awarded the Shanghai Outstanding Undergraduate Graduate honor, SJTU Outstanding Bachelor Thesis, and the Zhiyuan Honors Bachelor Degree.

Before joining Michigan, I was a research intern at CogAI4Sci, National University of Singapore (advisor: Prof. Dianbo Liu), and collaborated with SJTU (Prof. Qunbi Zhuge) and HKUST (Prof. Xin Tong) on multi-agent optical network control and generative UI systems.

News

  • Fall 2026Started Ph.D. at UMich ECE; co-advised by Prof. Qing Qu (DeepThink Lab) and Prof. Joyce Wang (PixAIL).
  • Jun 2026Graduated from SJTU Zhiyuan Honors; Shanghai Outstanding Undergraduate Graduate (1%), Outstanding Bachelor Thesis (1%), Zhiyuan Honors Bachelor Degree.
  • 2026Two manuscripts under review at NeurIPS 2026; ACP 2026 and JLT manuscripts under review.
  • 2026Preprint: AI-generated data contamination erodes pathological variability and diagnostic reliability (arXiv:2601.12946).
  • 2025OptiMA accepted at ECOC 2025; DuetUI accepted at ACM CHI; ACP 2025 paper accepted.
  • Apr 2025Joined NUS CogAI4Sci as a research intern.
  • Sep 2024Received National Scholarship (Top 0.2%).
  • Sep 2022Started B.Eng. in Electronic Science & Technology at Shanghai Jiao Tong University (Zhiyuan Honors Program).

Research Interests

AI for Science & ML

Foundation models, representation learning, and machine learning methods for scientific discovery.

Intelligent Optical Networks

Multi-agent LLM frameworks, digital twins, and reliable QoT / EDFA modeling for optical systems.

Medical AI Reliability

Model collapse, synthetic data contamination, and trustworthy multi-modal healthcare AI.

Selected Publications

  1. S. Xiang, S. Wu, X. Liu, Q. Qiu, Y. Zhang, Y. Chen, and Q. Zhuge, “OptiMA: Collaborative Multi-Agent Framework for Modelling and Controlling Raman Amplifier in Intelligent Optical Networks,” ECOC 2025. (accepted) [IEEE Xplore]
  2. Y. Xu, S. Xiang, Y. Song, R. Sun, and X. Tong, “DuetUI: A Bidirectional Context Loop for Human-Agent Co-Generation of Task-Oriented Interfaces,” ACM CHI. (accepted) [arXiv]
  3. H. He, S. Xiang, et al., “AI-generated data contamination erodes pathological variability and diagnostic reliability,” arXiv preprint, 2026. [arXiv]

View all publications →

Contact