Zhijie Deng (邓志杰) Assistant Professor, Ph.D. Advisor Qing Yuan Research Institute School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai, China Office: SEIEE-3-301/剑川路930号A栋318C Email: zhijied@sjtu[dot]edu[dot]cn, dengzhijiethu@gmail[dot]com [GitHub] [Google Scholar]

Biography

I am an Assistant Professor at Qing Yuan Research Institute at Shanghai Jiao Tong University. I obtained the Ph.D. degree from Department of Computer Science and Technology, Tsinghua University in June 2022, under the supervision of Prof. Bo Zhang and Prof. Jun Zhu. Before that, I received the B.S. degree from Department of Computer Science and Technology, Tsinghua University in 2017. I was a visiting scholar from July 2016 to September 2016 in Machine Learning Department at Carnegie Mellon University, advised by Prof. Eric P. Xing.

My principal research interests lie in the development of machine learning and statistical methodology, spanning generative models, spectral methods, Bayesian inference, etc. The developed methods have been applied to scenarios such as image/language generation, semi-/self-/un-supervised learning, PDE solving, uncertainty quantification, and trustworthy ML.

Openings

A PhD position (starting from September 2025) in the area of machine learning/deep learning.

I am looking for self-motivated students who have a strong background in mathematics/physics/computer science and are eager to get involved in cutting-edge, fundamental AI research (especially in fields of LLMs, diffusion models, and AI4PDE). If you are interested, feel free to drop me an email.

Selected Publications [More]

* equal contribution. † corresponding author.

Large Language Models

• CLLMs: Consistency Large Language Models
  Siqi Kou, Lanxiang Hu, Zhezhi He, Zhijie Deng†, and Hao Zhang
• Online Speculative Decoding
  Xiaoxuan Liu, Lanxiang Hu, Peter Bailis, Ion Stoica, Zhijie Deng†, Alvin Cheung, and Hao Zhang†
• Efficient Detection of LLM-generated Texts with a Bayesian Surrogate Model
  Zhijie Deng†, Hongcheng Gao, Yibo Miao, and Hao Zhang†

Generative Models

• LOVECon: Text-driven Training-Free Long Video Editing with ControlNet
  Zhenyi Liao and Zhijie Deng†
• BayesDiff: Estimating Pixel-wise Uncertainty in Diffusion via Bayesian Inference
  Siqi Kou, Lei Gan, Dequan Wang, Chongxuan Li†, and Zhijie Deng†
  International Conference on Learning Representations (ICLR), Vienna, Austria, 2024
  
• On Calibrating Diffusion Probabilistic Models
  Tianyu Pang†, Cheng Lu, Chao Du, Min Lin, Shuicheng Yan, and Zhijie Deng†
  Advances in Neural Information Processing Systems (NeurIPS), New Orleans, USA, 2023
  
• Structured Generative Adversarial Networks
  Zhijie Deng*, Hao Zhang*, Xiaodan Liang, Luona Yang, Shizhen Xu, Jun Zhu, and Eric P Xing
  Advances in Neural Information Processing Systems (NeurIPS), Long Beach, USA, 2017 (NVAIL Pioneering Research Award, 2017.12)
  [ArXiv] [code]

Deep Spectral Methods

• Improved Operator Learning by Orthogonal Attention
  Zipeng Xiao, Zhongkai Hao, Bokai Lin, Zhijie Deng†, and Hang Su†
• Neural Eigenfunctions Are Structured Representation Learners
  Zhijie Deng*, Jiaxin Shi*, Hao Zhang, Peng Cui, Cewu Lu, and Jun Zhu
  
• Learning Neural Eigenfunctions for Unsupervised Semantic Segmentation
  Zhijie Deng and Yucen Luo
  International Conference on Computer Vision (ICCV), Paris, France, 2023
  [ArXiv] [code]
• NeuralEF: Deconstructing Kernels by Deep Neural Networks
  Zhijie Deng, Jiaxin Shi, and Jun Zhu
  International Conference on Machine Learning (ICML), Baltimore, USA, 2022
  [ArXiv] [code] [slides]

Bayesian Deep Learning and Uncertainty Estimation

• Bayesian Exploration of Pre-trained Models for Low-shot Image Classification
  Yibo Miao, Yu Lei, Feng Zhou†, and Zhijie Deng†
  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, USA, 2024
• Towards Accelerated Model Training via Bayesian Data Selection
  Zhijie Deng*, Peng Cui*, and Jun Zhu
  Advances in Neural Information Processing Systems (NeurIPS), New Orleans, USA, 2023
  
• Accelerated Linearized Laplace Approximation for Bayesian Deep Learning
  Zhijie Deng, Feng Zhou, and Jun Zhu
  Advances in Neural Information Processing Systems (NeurIPS), New Orleans, USA, 2022 (Spotlight)
  [ArXiv] [code] [slides]
• LiBRe: A Practical Bayesian Approach to Adversarial Detection
  Zhijie Deng, Xiao Yang, Shizhen Xu, Hang Su, and Jun Zhu
  IEEE Conference on Computer Vision and Pattern Recognition (CVPR), VIRTUAL, 2021 (Valse 2021 Spotlight, 2021.10)
  [code] [slides]

Talks

• 扩散模型：方法与应用, CSIG2023青年科学家会议扩散模型与内容生成论坛, 2023.12
  
• Online Speculative Decoding，安徽省安全人工智能研究院, 2023.12
  
• Deep Spectral Methods: Another Way to Unsupervised Learning, Bosch Center for AI (Shanghai), Shanghai Artificial Intelligence Laboratory, Kuaishou, SJTU, HFUT, and NUDT
  
• Bayesian Deep Learning: Methods and Applications, PRCV 2021, 2021.12
  

Selected Honors & Awards

• Outstanding Graduates, Department of Computer Science and Technology, Tsinghua University, 2022.06
  
• Huawei-Tsinghua Outstanding Service Award, 2021.06
• '84' Future Innovation Scholarship, 2020.12
• Tsinghua Good Graduates, Tsinghua University, 2017.06
  
• Outstanding Graduates, Department of Computer Science and Technology, Tsinghua University, 2017.06
  

Services

Conference reviewer: NeurIPS 2019-2023; ICML 2021-2024; ICLR 2022-2024; CVPR 2021-2024; ICCV 2021 2023; AISTATS 2024; ECCV 2022 2024; AAAI 2021-2022; AutoML 2022
Journal reviewer: TPAMI, TIP, TNNLS, Artificial Intelligence , TITS , Neural Networks , TMLR

Teaching

2024 spring, [CS7352] Advanced Neural Network Theory and Application, in cooperation with Prof. Weiran Huang and Prof. Yonglu Li, Shanghai Jiao Tong University
2024 spring, [CS3507] Science and Technology Innovation (Part 4-J), Shanghai Jiao Tong University
2023 fall, [AI1601] Artificial Intelligence Thinking and Ethics, in cooperation with Prof. Leilei Gu and Prof. Ge Zheng, Shanghai Jiao Tong University
2023 spring, [CS7310H] Algorithm Design and Analysis, in cooperation with Prof. Zhilei Xu, Shanghai Jiao Tong University
2022 fall, assistant of Deep Learning and the Application by Prof. Junchi Yan, Shanghai Jiao Tong University
2019 spring, TA in Statistical Machine Learning, instructed by Prof. Jun Zhu, Tsinghua University
2018 fall, TA in Machine Learning, instructed by Prof. Jie Tang and Prof. Jun Zhu, Tsinghua University

© 2024.01 Zhijie Deng