Wanru (Renee) Zhao

William Gates Building
15 JJ Thomson Ave
Cambridge CB3 0FD
Hey, thanks for stopping by! đź‘‹
I’m a PhD student in Computer Science at the Department of Computer Science and Technology, University of Cambridge, supervised by Nic Lane at the Cambridge Machine Learning Systems Lab (CaMLSys). I’m also a member of Cambridge AI Safety Lab, working on AI Alignment and Interpretability. I’m currently visiting the Vector Institute, working with Colin Raffel at the University of Toronto. I was a research intern at Microsoft Research, mentored by Alessandro Sordoni and Lucas Caccia.
My research focuses on:
- Modular, distributed/decentralised training (model merging, Mixture-of-Experts) and decentralised inference;
- Data attribution/selection/curation/balancing/mixing, synthetic data generation and curriculum design for foundation model training;
- Compositional reasoning of large language models (in math and coding domains)
Prior to Cambridge, I was glad to be advised by Yongxin Tong and Xianglong Liu. I also spent wonderful time doing research at AWS AI Lab mentored by Minjie Wang and Zheng Zhang, where I worked on Deep Graph Library and its interpretability toolkits GNNLens2
.
I’m open to collaborations in all forms and would love to explore any opportunities! Feel free to contact me by Email (wz341 [AT] cam.ac.uk or zhaowrenee [AT] gmail.com), Wechat, or Calendly.
news
Sep 26, 2024 | One first-author paper and one co-authored paper accepted by NeurIPS 2024 (with an acceptance rate of 25.8%)! See you in Vancouver 🇨🇦 |
---|---|
Mar 15, 2024 | I presented my work on multilinguality and data quality in federated learning on Flower AI Summit! |
Feb 1, 2024 | One first-author conference paper and two first-author workshop papers are accepted by ICLR 2024! See you in Vienna 🇦🇹 Hit me up if you are also attending and want to talk about decentralised ml / data quality / multilingual / AI safety! |
Mar 30, 2023 | Our team got the winner of the US-UK Privacy-Enhancing Technologies Prize Challenges! We will present our solution at Innovation and Technology’s Centre for Data Ethics and Innovation (CDEI) in London at the end of May. Check out the report on the Cambridge University website! |
Mar 30, 2023 | Our paper “Harms from Increasingly Agentic Algorithmic Systems” is accepted by FAccT 2023! Honored to be cited by GPT-4 Technical Report. I’m also the volunteer helping the conference organisation. |
Mar 8, 2023 | I was invited to give a talk about “Challenges and Prospective Technologies for Privacy-Enhanced Federated Learning” on Women@CL Talklet. Happy International Women’s Day! |
selected publications
- Graph Attention Based Proposal 3D ConvNets for Action DetectionProceedings of the AAAI Conference on Artificial Intelligence Apr 2020
- Protea: Client Profiling within Federated Systems using FlowerIn Proceedings of the 1st ACM Workshop on Data Privacy and Federated Learning Technologies for Mobile Edge Network Oct 2022
- Breaking Physical and Linguistic Borders: Multilingual Federated Prompt Tuning for Low-Resource LanguagesIn The Twelfth International Conference on Learning Representations Oct 2024
- Attacks on Third-Party APIs of Large Language ModelsThe Twelfth International Conference on Secure and Trustworthy Large Language Models Oct 2024
- Enhancing Data Quality in Federated Fine-Tuning of Foundation ModelsThe Twelfth International Conference on Learning Representations Workshop on Navigating and Addressing Data Problems for Foundation Models Oct 2024