Fuyuan Lyu

McGill University

bio-photo-4.jpg

I’m currently a Ph.D. candidate at McGill University and Mila. I am fortunate to be advised by Prof Jin L.C. Guo and Prof Xue Liu.

Before joining McGill, I got my bachelor’s degree in Computer Science and Zhiyuan Honour Degree in Engineering from Shanghai Jiao Tong University(SJTU). I was advised by Prof Li Jiang for my Bachelor’s thesis and worked with Prof Xiaokang Yang. I also worked with Prof Weichen Liu as a visiting research student at Nanyang Technological University(NTU), who inspired me to conduct high-quality research.

I am passionate about building Data-Centric AI. Specificially, I am studying (i) how to model and select feature and (ii) how to provide high-quality labels for both deep learning and foundation models fully automaticially. I am eager to free humankind from repetitive, tedious work and enable us to focus on more creative tasks. I am actively working on its application in various domains, including advertisement, finance, and software engineering.

Email:

  • FIRST_NAME DOT LAST_NAME AT MAIL DOT MCGILL DOT CA
  • FIRST_NAME DOT LAST_NAME AT MILA DOT QUEBEC

news

May 16, 2025 Two papers got accepted by KDD 2025
May 15, 2025 One paper got accepted by ACL 2025, congs to Qiyuan and my other amazing collaborators!
May 10, 2025 I was invited to give a talk on test-time scaling at Multimodal Weekly at Twelve Lab. Thanks for the invite, James!
Jan 22, 2025 One paper got accepted by ICLR 2025
Oct 23, 2024 One paper got accepted by WSDM 2025

selected publications

2025

  1. KDD
    Timing is important: Risk-aware Fund Allocation based on Time-Series Forecasting
    Fuyuan Lyu, Linfeng Du, Yunpeng Weng, Qiufang Ying, Zhiyan Xu, Wen Zou, Haolun Wu, Xiuqiang He, and Xing Tang
    In KDD, 2025
    arXiv Code
  2. ACL
    Crowd Comparative Reasoning: Unlocking Comprehensive Evaluations for LLM-as-a-Judge
    Qiyuan Zhang, Yufei Wang, Yuxin Jiang, Liangyou Li, Chuhan Wu, Yasheng Wang, Xin Jiang, Lifeng Shang, Ruiming Tang, Fuyuan Lyu, and 1 more author
    In ACL, 2025
    arXiv
  3. CoRR
    A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?
    Qiyuan Zhang, Fuyuan Lyu, Zexu Sun, Lei Wang, Weixu Zhang, Wenyue Hua, Haolun Wu, Zhihan Guo, Yufei Wang, Niklas Muennighoff, and 3 more authors
    In CoRR, 2025
    arXiv Code Slides Video Website
  4. ICLR
    RevisEval: Improving LLM-as-a-Judge via Response-Adapted References
    Qiyuan Zhang, Yufei Wang, Tiezheng Yu, Yuxin Jiang, Chuhan Wu, Liangyou Li, Yasheng Wang, Xin Jiang, Lifeng Shang, Ruiming Tang, and 2 more authors
    In ICLR, 2025
    arXiv Proceeding

2024

  1. EMNLP
    Collaborative Performance Prediction for Large Language Models
    Qiyuan Zhang, Fuyuan Lyu, Xue Liu, and Chen Ma
    In EMNLP, 2024
    arXiv Code Proceeding Poster Slides Video

2023

  1. NeurIPS
    Towards Hybrid-grained Feature Interaction Selection for Deep Sparse Network
    Fuyuan Lyu, Xing Tang, Dugang Liu, Chen Ma, Weihong Luo, Xiuqiang He, and Xue Liu
    In NeurIPS, 2023
    arXiv Code Slides
  2. WWW
    Optimizing Feature Set for Click-Through Rate Prediction
    Fuyuan Lyu, Xing Tang, Dugang Liu, Chen Liang, Xiuqiang He, and Xue Liu
    In WWW, 2023
    arXiv Code Proceeding Slides Video

2022

  1. CIKM
    Learning Optimal Embedding Table for Click-through Rate Prediction
    Fuyuan Lyu, Xing Tang, Hong Zhu, Huifeng Guo, Yingxue Zhang, Ruiming Tang, and Xue Liu
    In CIKM, 2022
    arXiv Code Proceeding Slides
  2. ICDE
    Memorize, Factorize, or be Naive: Learning Optimal Feature Interaction Methods for CTR Prediction
    Fuyuan Lyu, Xing Tang, Huifeng Guo, Ruiming Tang, Xiuqiang He, Rui Zhang, and Xue Liu
    In ICDE, 2022
    arXiv Code Proceeding Poster Slides Video