Fuqiang Liu | AI Safety

I am Fuqiang, an about-to-graduate Ph.D. at McGill University, Canada, co-supervised by Prof. Lijun Sun and Prof. Luis Miranda-Moreno.

My research focuses on AI safety, specifically, to identify the potential risks hidden in powerful AI models and construct resilient solutions for practical intelligent systems. Ongoing research directions are:

1. Potential Risks in LLMs for Time Series Forecasting

2. Resilient Intelligent Transportion System (ITS)

Prior to my doctorate, I earned two Master of Science (MSc) degrees——one in Communication Engineering from Beijing Institute of Technology (BIT), China, and another in Electrical and Computer Engineering from Clarkson University, USA. I also hold a Bachelor’s degree in Electrical and Electronic Engineering from BIT, China.

News

Apr 22, 2025	I was glad to be chosen as the best reviewer award in AISTATS’25. see
Jan 22, 2025	The first-author paper, “Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting”, was accepted in AISTATS 2025.
Dec 15, 2024	I participated in NeuIPS’24 Safe Generative AI with a poster presentation.

Selected Publications

AISTATS’25

Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting

Fuqiang Liu, Sicong Jiang, Luis Miranda-Moreno, and 2 more authors

In International Conference on Artificial Intelligence and Statistics, 2025

Bib PDF Code Website

@inproceedings{liu2025adversarial,
  title = {Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting},
  author = {Liu, Fuqiang and Jiang, Sicong and Miranda-Moreno, Luis and Choi, Seongjin and Sun, Lijun},
  booktitle = {International Conference on Artificial Intelligence and Statistics},
  pages = {4672--4680},
  year = {2025},
  organization = {PMLR},
}

TNNLS’23

Adversarial danger identification on temporally dynamic graphs

Fuqiang Liu, Jingbo Tian, Luis Miranda-Moreno, and 1 more author

IEEE Transactions on Neural Networks and Learning Systems, 2023

Bib PDF

@article{liu2023adversarial,
  title = {Adversarial danger identification on temporally dynamic graphs},
  author = {Liu, Fuqiang and Tian, Jingbo and Miranda-Moreno, Luis and Sun, Lijun},
  journal = {IEEE Transactions on Neural Networks and Learning Systems},
  volume = {35},
  number = {4},
  pages = {4744--4755},
  year = {2023},
  publisher = {IEEE},
}