吴泳涛 讲师
返回列表邮箱：yongtaowu@ruc.edu.cn
菠菜平台高瓴人工智能学院讲师。2026年获瑞士洛桑联邦理工学院博士学位，曾任加州大学洛杉矶分校会见学者。主要研究偏向为机械学习理论与优化，重点从优化理论视角研究神经网络的训练动力学；同时也将优化要领应用于可信大模型研究。在 NeurIPS、ICML、ICLR、IEEE TPAMI 等人工智能领域重要聚会会议和期刊发表论文10余篇。曾组织瑞士洛桑联邦理工学院数字信任中心大模型微调陷阱研讨会，并协助组织 NeurIPS 2024 FITML 研讨会。曾获瑞典皇家理工学院年度奖学金。恒久担任 NeurIPS、ICLR、ICML、AISTATS、IJCV、TMLR 等国际聚会会议和期刊审稿人。
教育经历
2022-2026年 瑞士洛桑联邦理工学院 博士
2020-2022年 瑞典皇家理工学院 硕士
2016-2020年 中山大学 本科
教授课程
研究偏向
机械学习理论（优化、泛化、训练动力学）
可信大模型（鲁棒性、宁静性、隐私）
学术论文
From Spectral to Frobenius: When Muon Loses Its Spectral Implicit Bias
Yongtao Wu, Mahdi Mansouri, Zhenyu Zhu
arXiv preprint , 2026
DiffCAP: Diffusion-based Cumulative Adversarial Purification for Vision Language Models
Jia Fu, Yongtao Wu, Yihang Chen, Kunyu Peng, Xiao Zhang, Volkan Cevher, Sepideh Pashami, Anders Holst
Transactions on Machine Learning Research, 2026
Hadamard Product in Deep Learning: Introduction, Advances and Challenges
Grigorios Chrysos, Yongtao Wu, Razvan Pascanu, Philip Torr, Volkan Cevher
Transactions on Pattern Analysis and Machine Intelligence, 2025
Machine learning-aided hysteretic response prediction of double skin composite wall under earthquake loads
Shiye Wang, Wei Wang, Yongtao Wu, Zhiyang Xie, Yuqing Gao
Journal of Building Engineering, 2025
Robustness in Both Domains: CLIP Needs a Robust Text Encoder
Elias Abad Rocamora, Christian Schlarmann, Naman Deep Singh, Yongtao Wu, Matthias Hein, Volkan Cevher
Conference on Neural Information Processing Systems, 2025
Multi-Step Alignment as Markov Games: An Optimistic Online Gradient Descent Approach with Convergence Guarantees
Yongtao Wu (配合一作), Luca Viano, Yihang Chen, Zhenyu Zhu, Kimon Antonakopoulos, Quanquan Gu, Volkan Cevher
Transactions on Machine Learning Research, 2025
Quantum-PEFT: Ultra Parameter-efficient Fine-tuning
Toshiaki Koike-Akino, Francesco Tonin, Yongtao Wu, Zhenqing Wu, Leyla Naz Candogan, Volkan Cevher
International Conference on Learning Representations, 2025
Linear Attention for Efficient Bidirectional Sequence Modeling
Arshia Afzal, Elias Abad Rocamora, Leyla Naz Candogan, Pol Puigdemont, Francesco Tonin, Yongtao Wu, Mahsa Shoaran, Volkan Cevher
Conference on Neural Information Processing Systems, 2025
Robust NAS under Adversarial Training: Benchmark, Theory, and Beyond
Yongtao Wu, Fanghui Liu, Carl-Johann Simon-Gabriel, Grigorios Chrysos, Volkan Cevher
International Conference on Learning Representations, 2024
Revisiting Character-level Adversarial Attacks for Language Models
Elias Abad Rocamora, Yongtao Wu (通讯作者), Fanghui Liu, Grigorios G. Chrysos, Volkan Cevher
International Conference on Machine Learning, 2024
Membership Inference Attacks against Large Vision-Language Models
Zhan Li, Yongtao Wu (配合一作), Yihang Chen, Francesco Tonin, Elias Abad Rocamora, Volkan Cevher
Conference on Neural Information Processing Systems, 2024
Universal Gradient Methods for Stochastic Convex Optimization
Anton Rodomanov, Ali Kavis, Yongtao Wu, Kimon Antonakopoulos, Volkan Cevher
International Conference on Machine Learning, 2024
Single-pass Detection of Jailbreaking Input in Large Language Models
Leyla Naz Candogan, Yongtao Wu, Elias Abad Rocamora, Grigorios G. Chrysos, Volkan Cevher
Transactions on Machine Learning Research, 2024
On the Convergence of Encoder-only Shallow Transformers
Yongtao Wu, Fanghui Liu, Grigorios G Chrysos, Volkan Cevher
Conference on Neural Information Processing Systems, 2023
Extrapolation and Spectral Bias of Neural Nets with Hadamard Product: a Polynomial Net Study
Yongtao Wu, Zhenyu Zhu, Fanghui Liu, Grigorios G Chrysos, Volkan Cevher
Conference on Neural Information Processing Systems, 2022
社会兼职
聚会会议审稿人： Conference on Neural Information Processing Systems (NeurIPS), International Conference on Learning Representations (ICLR), International Conference on Machine Learning (ICML), International Conference on Artificial Intelligence and Statistics (AISTATS)
期刊审稿人： International Journal of Computer Vision (IJCV), Transactions on Machine Learning Research (TMLR), Annals of the Institute of Statistical Mathematics (AISM)