About

I am now a visiting researcher at the University of Cambridge under the supervision of Prof. Nicholas Lane. I will be the Associate Chair of FedKDD 2025 (Federated Learning Workshop of KDD 2025). I am a Ph.D. candidate in Artificial Intelligence at the College of Computer Science and Technology, Zhejiang University (ZJU), China, supervised by Prof. Chao Wu. I am honored to work with Prof. Tao Lin at Westlake University/EPFL.

During my Ph.D., I also earned a minor degree in Technology-based (BEST) MBA from the School of Management, ZJU. Before Ph.D., I received my bachelor’s degree in Engineering from ZJU; meanwhile, I obtained an honorary undergraduate degree from Chu Kochen Honors College of ZJU and a minor degree in Public Affairs (social governance).

My main research interests lie in the following aspects. Large Language models and foundation models. i. Model editing and memory management for large language models (LLMs). ii. LLM agent and reasoning. iii. Parametric understanding of LLMs (localization, merging, scaling, pruning, stitching, unlearning, editing, and etc.) iv. Text-to-model generation by diffusion transformers (DiTs). v. Vision-Language (V-L) representation and understanding of multi-modal foundation models. Trustworthy deep learning. i. Privacy-preserving federated learning (FL), efficient & robust algorithm design, and generalization, personalization & training dynamics understanding. ii. Mechanistic interpretability of neural networks, weight decay, loss landscape, permutation invariance, linear mode connectivity, and etc. iii. Socio-technical issues brought by collaborative learning. iv. Responsible and trustworthy AI.

Contact

Email: zexi.li[at]zju.edu.cn / zl614[at]cam.ac.uk / tomleeze[at]gmail.com

Wechat and Phone: (+86) 18868104540

Selected Publications

• [NeurIPS 2024] WISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Models
  Peng Wang*, Zexi Li*, Ningyu Zhang#, Ziwen Xu, Yunzhi Yao, Yong Jiang, Pengjun Xie, Fei Huang, and Huajun Chen#
• [Preprint] Text-to-Model: Text-Conditioned Neural Network Diffusion for Train-Once-for-All Personalization
  Zexi Li*, Lingzhi Gao*, and Chao Wu#
• [ICML 2023] Revisiting Weighted Aggregation in Federated Learning with Neural Networks
  Zexi Li, Tao Lin#, Xinyi Shang, and Chao Wu#
• [Patterns, Cell Press] Can We Share Models If Sharing Data Is Not an Option?
  Zexi Li, Feng Mao#, and Chao Wu#
• [ICCV 2023] No Fear of Classifier Biases: Neural Collapse Inspired Federated Learning with Synthetic and Fixed Classifier
  Zexi Li, Xinyi Shang, Rui He, Tao Lin#, and Chao Wu#
• [ACM SIGIR 2023] Edge-cloud Collaborative Learning with Federated and Centralized Features
  Zexi Li*, Qunwei Li*, Yi Zhou, Wenliang Zhong#, Guannan Zhang, and Chao Wu#
• [IEEE Transactions on Big Data] Towards Effective Clustered Federated Learning: A Peer-to-peer Framework with Adaptive Neighbor Matching
  Zexi Li, Jiaxun Lu, Shuang Luo, Didi Zhu, Yunfeng Shao, Yinchuan Li, Zhimeng Zhang, Yongheng Wang#, and Chao Wu#

Recent News

• [2025.03] I will be the Associate Chair of FedKDD 2025, International Joint Workshop on Federated Learning for Data Mining and Graph Analytics, Co-located with the 31st ACM SIGKDD Conference (KDD 2025). Submissions are welcome!
• [2024.09] I am happy to share our WISE, a model editor for large language models’ lifelong model editing, is accepted to NeurIPS 2024!
• [2024.05] Two papers are accepted by KDD 2024, and one paper is accepted by ICML 2024.
• [2023.11] I am presented with Guoqiang Scholarship of Zhejiang University (Top 0.5%) and First Prize of Chunxiao Scholarship of BEST-MBA (Top 3%) by Zhejiang University.
• [2023.07] Two papers are accepted by International Conference on Computer Vision (ICCV 2023)! One is about federated learning and the other is about domain adaptation. Many thanks to my coauthors!
• [2023.06] We will have an oral presentation of our paper FedCNI at ICME 2023 on July 11th.
• [2023.04] Our paper “Revisiting Weighted Aggregation in Federated Learning with Neural Networks” is accepted by The Fortieth International Conference on Machine Learning (ICML 2023)! Many thanks to my coauthors!
• [2023.04] Our paper “Edge-cloud Collaborative Learning with Federated and Centralized Features” is accepted by The 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2023)! Thanks a lot to Ant Group’s coauthors!
• [2023.03] Our paper “Learning Cautiously in Federated Learning with Noisy and Heterogeneous Clients” is accepted by IEEE International Conference on Multimedia and Expo 2023 (ICME 2023)! Congratulations to me and Chenrui!
• [2023.03] I have joined the BEST-MBA (Business, Engineering, Science, and Technology based MBA for on-campus postgradutes) program of School of Management, Zhejiang University. <!–* [2022.11] Our paper “Towards Effective Clustered Federated Learning: A Peer-to-peer Framework with Adaptive Neighbor Matching”[arxiv] is online now.
• [2022.11] Our paper “Can we share models if sharing data is not an option?” is online now.
• [2022.11] Our paper “Towards Effective Clustered Federated Learning: A Peer-to-peer Framework with Adaptive Neighbor Matching” has been accepted by IEEE Transactions on Big Data! This paper had been through a dark time in the last year, and the acceptance means a lot to me. Congratulations to myself!
• [2022.10] I have passed the Mid-term Assessment of Doctoral Program.
• [2022.10] I am presented with Outstanding Postgraduate Student Award on 2021-2022 by Zhejiang University.
• [2022.09] Our new perspective paper “Can we share models if sharing data is not an option?” is accepted by Patterns, Cell Press. Patterns is a child-journal about data science of Cell Press, and it will have a first strong impact factor in 2023.–>

Academic Service

• Invited Reviewers: TKDE, IJCV, TMM, Machine Learning, AISTATS 2024, CVPR 2024, ICML 2024 2025, NeurIPS 2024 2025, ICLR 2024 2025, KDD 2025, ICCV 2025.
• Associate Chair of FedKDD 2025, International Joint Workshop on Federated Learning for Data Mining and Graph Analytics, Co-located with the 31st ACM SIGKDD Conference (KDD 2025).