Hui-Shuai Zhang

Most-cited papers

On Layer Normalization in the Transformer Architecture
Ruibin Xiong, Yunchang Yang, Di He, Kai Zheng +6 more
2020International Conference on Machine Learning1,685 citations
Semantic Scholar
Differentially Private Fine-tuning of Language Models
Da Yu, Saurabh Naik, A. Backurs, Sivakanth Gopi +8 more
2021International Conference on Learning Representations558 citations Open access
Semantic Scholar
xRAG: Extreme Context Compression for Retrieval-augmented Generation with One Token
Xin Cheng, Xun Wang, Xingxing Zhang, Tao Ge +4 more
2024Neural Information Processing Systems170 citations
Semantic Scholar
Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning
Da Yu, Hui-Shuai Zhang, Wei Chen, Tie-Yan Liu
2021International Conference on Learning Representations147 citations
Semantic Scholar
Large Scale Private Learning via Low-rank Reparametrization
Da Yu, Hui-Shuai Zhang, Wei Chen, Jian Yin +1 more
2021International Conference on Machine Learning129 citations
Semantic Scholar
Differentially Private Synthetic Data via Foundation Model APIs 2: Text
Chulin Xie, Zinan Lin, A. Backurs, Sivakanth Gopi +8 more
2024International Conference on Machine Learning103 citations
Semantic Scholar
Convergence of AdaGrad for Non-convex Objectives: Simple Proofs and Relaxed Assumptions
Bo-Han Wang, Hui-Shuai Zhang, Zhirui Ma, Wei Chen
2023Annual Conference Computational Learning Theory91 citations Open access
Semantic Scholar
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
Xin Cheng, Rui Tian, Wang-Ding Zeng, Da-Mai Dai +17 more
2026Annual Meeting of the Association for Computational Linguistics90 citations
Semantic Scholar
SGD Converges to Global Minimum in Deep Learning via Star-convex Path
Yi Zhou, Junjie Yang, Hui-Shuai Zhang, Ying-Bin Liang +1 more
2019International Conference on Learning Representations90 citations
Semantic Scholar
On the Generalization Properties of Diffusion Models
Puheng Li, Zhong Li, Hui-Shuai Zhang, Jiang Bian
2023Neural Information Processing Systems89 citations
Semantic Scholar
Availability Attacks Create Shortcuts
Da Yu, Hui-Shuai Zhang, Wei Chen, Jian Yin +1 more
2021Knowledge Discovery and Data Mining84 citations Open access
Semantic Scholar
Exploring the Limits of Differentially Private Deep Learning with Group-wise Clipping
Jiyan He, Xuechen Li, Da Yu, Hui-Shuai Zhang +5 more
2022International Conference on Learning Representations75 citations Open access
Semantic Scholar
Adaptive Inertia: Disentangling the Effects of Adaptive Learning Rate and Momentum
Zeke Xie, Xin-Rui Wang, Hui-Shuai Zhang, Issei Sato +1 more
2020International Conference on Machine Learning73 citations
Semantic Scholar
How Does Data Augmentation Affect Privacy in Machine Learning?
Da Yu, Hui-Shuai Zhang, Wei Chen, Jian Yin +1 more
2020AAAI Conference on Artificial Intelligence70 citations Open access
Semantic Scholar
DSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation
Xin Cheng, Xing-Kai Yu, Chen-Ze Shao, Jia-Shi Li +29 more
2026arXiv.org59 citations
Semantic Scholar
VideoLLM Knows When to Speak: Enhancing Time-Sensitive Video Comprehension with Video-Text Duet Interaction Format
Yue-Qian Wang, Xiao-Jun Meng, Yuxuan Wang, Jianxin Liang +3 more
2024Conference on Empirical Methods in Natural Language Processing58 citations
Semantic Scholar
ResiDual: Transformer with Dual Residual Connections
Shufang Xie, Hui-Shuai Zhang, Junliang Guo, Xu Tan +5 more
2023Artificial Intelligence50 citations Open access
Semantic Scholar
G-SGD: Optimizing ReLU Neural Networks in its Positively Scale-Invariant Space
Qi Meng, Shuxin Zheng, Hui-Shuai Zhang, Wei Chen +4 more
2018International Conference on Learning Representations48 citations
Semantic Scholar
Normalized/Clipped SGD with Perturbation for Differentially Private Non-Convex Optimization
Xiaodong Yang, Hui-Shuai Zhang, Wei Chen, Tie-Yan Liu
2022arXiv.org47 citations Open access
Semantic Scholar
Closing the Gap Between the Upper Bound and the Lower Bound of Adam's Iteration Complexity
Bo-Han Wang, Jingwen Fu, Hui-Shuai Zhang, Nan-Ning Zheng +1 more
2023Neural Information Processing Systems44 citations
Semantic Scholar
Provable Adaptivity of Adam under Non-uniform Smoothness
Bo-Han Wang, Yu-Shun Zhang, Hui-Shuai Zhang, Qi Meng +5 more
2022Knowledge Discovery and Data Mining42 citations Open access
Semantic Scholar
Solving Formal Math Problems by Decomposition and Iterative Reflection
Yichi Zhou, Jianqiu Zhao, Yong-Xin Zhang, Bo-Han Wang +13 more
2025arXiv.org41 citations
Semantic Scholar
Understanding generalization error of SGD in nonconvex optimization
Yi Zhou, Ying-Bin Liang, Hui-Shuai Zhang
2019Machine-mediated learning41 citations Open access
Semantic Scholar
ProactiveVideoQA: A Comprehensive Benchmark Evaluating Proactive Interactions in Video Large Language Models
Yue-Qian Wang, Xiao-Jun Meng, Yi-Fan Wang, Hui-Shuai Zhang +1 more
2025arXiv.org34 citations
Semantic Scholar
Individual Privacy Accounting for Differentially Private Stochastic Gradient Descent
Da Yu, Gautam Kamath, Janardhan Kulkarni, Tie-Yan Liu +2 more
2022Trans. Mach. Learn. Res.34 citations
Semantic Scholar
Denoising Masked Autoencoders Help Robust Classification
Quanlin Wu, Hang Ye, Yun-Tian Gu, Hui-Shuai Zhang +2 more
2022International Conference on Learning Representations32 citations
Semantic Scholar
Does Momentum Change the Implicit Regularization on Separable Data?
Bo-Han Wang, Qi Meng, Hui-Shuai Zhang, Ruoyu Sun +3 more
2021Neural Information Processing Systems31 citations Open access
Semantic Scholar
MMDuet2: Enhancing Proactive Interaction of Video MLLMs with Multi-Turn Reinforcement Learning
Yue-Qian Wang, Songxiang Liu, Disong Wang, Nuo Xu +3 more
2025arXiv.org30 citations
Semantic Scholar
Selective Pre-training for Private Fine-tuning
Da Yu, Sivakanth Gopi, Janardhan Kulkarni, Zinan Lin +4 more
2023Trans. Mach. Learn. Res.30 citations Open access
Semantic Scholar
Convergence of Distributed Stochastic Variance Reduced Methods Without Sampling Extra Data
Shicong Cen, Hui-Shuai Zhang, Yuejie Chi, Wei Chen +1 more
2019IEEE Transactions on Signal Processing29 citations Open access
Semantic Scholar
Automatic Jailbreaking of the Text-to-Image Generative AI Systems
Minseon Kim, Hyomin Lee, Boqing Gong, Hui-Shuai Zhang +1 more
2024arXiv.org28 citations
Semantic Scholar
Efficient Continual Pre-training by Mitigating the Stability Gap
Yiduo Guo, Jie Fu, Hui-Shuai Zhang, Dongyan Zhao +1 more
2024arXiv.org28 citations
Semantic Scholar
Two Coupled Rejection Metrics Can Tell Adversarial Examples Apart
Tianyu Pang, Hui-Shuai Zhang, Di He, Yin-Peng Dong +4 more
2021Computer Vision and Pattern Recognition28 citations Open access
Semantic Scholar
When and Why Momentum Accelerates SGD: An Empirical Study
Jingwen Fu, Bo-Han Wang, Hui-Shuai Zhang, Zhi-Zheng Zhang +2 more
2023arXiv.org27 citations Open access
Semantic Scholar
DiffKendall: A Novel Approach for Few-Shot Learning with Differentiable Kendall's Rank Correlation
Kai Zheng, Hui-Shuai Zhang, Weiran Huang
2023Neural Information Processing Systems27 citations Open access
Semantic Scholar
FD-Align: Feature Discrimination Alignment for Fine-tuning Pre-Trained Models in Few-Shot Learning
Kun Song, Huimin Ma, Bo-Chao Zou, Hui-Shuai Zhang +1 more
2023Neural Information Processing Systems23 citations
Semantic Scholar
Indiscriminate Poisoning Attacks Are Shortcuts
Da Yu, Hui-Shuai Zhang, Wei Chen, Jian Yin +1 more
2021arXiv.org20 citations
Semantic Scholar
Shorten After You're Right: Lazy Length Penalties for Reasoning RL
Danlong Yuan, Tian Xie, Shao-Han Huang, Zhuo-Cheng Gong +4 more
202519 citations
Semantic Scholar
On the Convergence of Adam under Non-uniform Smoothness: Separability from SGDM and Beyond
Bo-Han Wang, Hui-Shuai Zhang, Qi Meng, Ruoyu Sun +2 more
2024arXiv.org19 citations
Semantic Scholar
ReMamba: Equip Mamba with Effective Long-Sequence Modeling
Danlong Yuan, Jia-Hao Liu, Bei Li, Hui-Shuai Zhang +3 more
2024Conference on Empirical Methods in Natural Language Processing18 citations
Semantic Scholar
Efficient Domain Continual pretraining by Mitigating the Stability Gap
Yiduo Guo, Jie Fu, Hui-Shuai Zhang, Dong-Yan Zhao
2025Annual Meeting of the Association for Computational Linguistics17 citations Open access
Semantic Scholar
Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation
Zekai Zhang, Jia-Hao Li, Jie Zhang, Kai-Yuan Gao +17 more
2026arXiv.org15 citations
Semantic Scholar
Do LLMs "Feel"? Emotion Circuits Discovery and Control
Chen-Xi Wang, Yixuan Zhang, Rui Yu, Yufei Zheng +7 more
2025arXiv.org15 citations
Semantic Scholar
K-frames: Scene-Driven Any-k Keyframe Selection for long video understanding
Yi-Feng Yao, Yi-Ke Yun, Jing Wang, Hui-Shuai Zhang +5 more
2025arXiv.org15 citations
Semantic Scholar
Similarity Distribution based Membership Inference Attack on Person Re-identification
Jun-Yao Gao, Xinyang Jiang, Hui-Shuai Zhang, Yifan Yang +5 more
2022AAAI Conference on Artificial Intelligence15 citations Open access
Semantic Scholar
Stabilize deep ResNet with a sharp scaling factor τ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\tau$$\end{document
Hui-Shuai Zhang, Da Yu, Ming-Yang Yi, Wei Chen +1 more
2019Machine-mediated learning15 citations Open access
Semantic Scholar
Training Over-parameterized Deep ResNet Is almost as Easy as Training a Two-layer Network
Hui-Shuai Zhang, Da Yu, Wei Chen, Tie-Yan Liu
2019arXiv.org15 citations
Semantic Scholar
Convergence Theory of Learning Over-parameterized ResNet: A Full Characterization.
Hui-Shuai Zhang, Da Yu, Ming-Yang Yi, Wei Chen +1 more
201914 citations
Semantic Scholar
Synthesize Privacy-Preserving High-Resolution Images via Private Textual Intermediaries
Hao-Xiang Wang, Zinan Lin, Da Yu, Hui-Shuai Zhang
2025Neural Information Processing Systems12 citations
Semantic Scholar
Mixture-of-Modules: Reinventing Transformers as Dynamic Assemblies of Modules
Zhuo-Cheng Gong, Ang Lv, Jian Guan, Wei Wu +4 more
2024Conference on Empirical Methods in Natural Language Processing12 citations
Semantic Scholar