Beijing Institute of Mathematical Sciences and Applications Beijing Institute of Mathematical Sciences and Applications

  • About
    • President
    • Governance
    • Partner Institutions
    • Visit
  • People
    • Management
    • Faculty
    • Postdocs
    • Visiting Scholars
    • Administration
    • Academic Support
  • Research
    • Research Groups
    • Courses
    • Seminars
  • Join Us
    • Faculty
    • Postdocs
    • Students
  • Events
    • Conferences
    • Workshops
    • Forum
  • Life @ BIMSA
    • Accommodation
    • Transportation
    • Facilities
    • Tour
  • News
    • News
    • Announcement
    • Downloads
About
President
Governance
Partner Institutions
Visit
People
Management
Faculty
Postdocs
Visiting Scholars
Administration
Academic Support
Research
Research Groups
Courses
Seminars
Join Us
Faculty
Postdocs
Students
Events
Conferences
Workshops
Forum
Life @ BIMSA
Accommodation
Transportation
Facilities
Tour
News
News
Announcement
Downloads
Qiuzhen College, Tsinghua University
Yau Mathematical Sciences Center, Tsinghua University (YMSC)
Tsinghua Sanya International  Mathematics Forum (TSIMF)
Shanghai Institute for Mathematics and  Interdisciplinary Sciences (SIMIS)
BIMSA > Yunfeng Cai

Yunfeng Cai

     Professor    
Professor Yunfeng Cai

Group: Artificial Intelligence and Machine Learning

Office: A3-4-307

Email: caiyunfeng@bimsa.cn

Research Field: Machine Learning

Biography


Yunfeng Cai studied in Mathematics at the University of Science and Technology of China from 2000 to 2004. He then pursued his PhD in Computing Mathematics at Peking University, which he obtained in Jan. 2009. From Jan. 2009 to June 2012, he worked as a postdoctoral researcher at the Academy of Mathematics and Systems Science / University of California, Davis. From Sep. 2012 to Sep. 2018, he served as a researcher at Peking University. In Sep. 2018, he joined Baidu Research as a Research Scientist and left on May 2024. In June 2024, he began his current role as a Professor at the Beijing Institute of Mathematical Sciences and Applications (BIMSA).

Research Interest


  • LLM and Reasoning
  • AI for science
  • Computer Vision
  • RL and Control

Education Experience


  • 2004 - 2009      Peking University      Computing Mathematics      Ph.D
  • 2000 - 2004      University of Science and Technology of China      Computing Mathematics      Bachelor

Work Experience


  • 2024 -      BIMSA      Professor
  • 2018 - 2024      Baidu Research      Researcher
  • 2012 - 2018      Peking University      Researcher
  • 2010 - 2012      University of California, Davis      Postdoc
  • 2009 - 2009      Academy of Mathematics and Systems Science, CAS      Postdoc

Publication


  • [1] Haoyu Liang, Youran Sun, Yunfeng Cai, Jun Zhu, Bo Zhang, Jailbreaking LLMs' Safeguard with Universal Magic Words for Text Embedding Models (2025)
  • [2] Xiaochen Zhang, Yunfeng Cai, Haoyi Xiong, Knoop: Practical Enhancement of Knockoff with Over-Parameterization for Variable Selection, Machine Learning, 114(1) (2025)
  • [3] Yang Liu, Huang Fang, Yunfeng Cai, Mingming Sun , MQuinE: a Cure for “Z-paradox” in Knowledge Graph Embedding, , 9837-9850 (2024)
  • [4] Z Yu, H Wang, J Yang, H Wang, Z Xie, Y Cai, J Cao, Z Ji, M Sun, Sgd: Street view synthesis with gaussian splatting and diffusion prior, arXiv preprint arXiv:2403.20079 (2024)
  • [5] Xindi Yang, Zeke Xie, Xiong Zhou, Boyu Liu, Buhua Liu, Yi Liu, Haoran Wang, Yunfeng Cai, Mingming Sun, Neural Field Classifiers via Target Encoding and Classification Loss, arXiv preprint arXiv:2403.01058 (2024)
  • [6] Y Liu, H Fang, Y Cai, M Sun, MQuinE: a cure for" Z-paradox" in knowledge graph embedding models, arXiv preprint arXiv:2402.03583 (2024)
  • [7] Shengyuan Chen, Yunfeng Cai, Huang Fang, and Mingming Sun, Differentiable neuro-symbolic reasoning on large-scale knowledge graphs, Advances in Neural Information Processing Systems, 36, 28139-28154 (2023)
  • [8] Huang Fang, Yang Liu, Yunfeng Cai, and Mingming Sun, Mln4kb: an efficient markov logic network engine for large-scale knowledge bases and structured logic rules, , 2423-2432 (2023)
  • [9] Ming Zhou, Zhaojun Bai, Yunfeng Cai, Klaus Neymeyr, Convergence analysis of a block preconditioned steepest descent eigensolver with implicit deflation, NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, 35(2023)
  • [10] Zeke Xie, Xindi Yang, Yujie Yang, Qi Sun, Yixiang Jiang, Haoran Wang, Yunfeng Cai, and Mingming Sun, S3im: Stochastic structural similarity and its unreasonable effectiveness for neural fields, , 18024-18034 (2023)
  • [11] Tan Yu, Jun Li, Yunfeng Cai, and Ping Li, Constructing orthgonal convolutions in an explicit manner, The Tenth International Conference on Learning Representations (ICLR), Virtual Event, April 25-29, 2022(2022)
  • [12] Jinxing Yu, Yunfeng Cai, Mingming Sun, and Ping Li, Spacee: Knowledge graph embedding by relational linear transformation in the entity space, , 64-72 (2022)
  • [13] Yunfeng Cai, and Ping Li, Identification of Matrix Joint Block Diagonalization, Proceedings of Machine Learning Research, 130(2022), 1495-1503
  • [14] Tan Yu, Yunfeng Cai, and Ping Li, Efficient Compact Bilinear Pooling via Kronecker Product, Proceedings of the AAAI Conference on Artificial Intelligence, 36(2022), 3, 3170-3178
  • [15] Tan Yu, Xu Li, Yunfeng Cai, Mingming Sun, and Ping Li, $S^2$-MLP: Spatial-Shift MLP Architecture for Vision, 2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, HI, USA, 2022(2022), 297-306
  • [16] Z Xie, QY Tang, Y Cai, M Sun, P Li, On the power-law hessian spectrums in deep learning, arXiv preprint arXiv:2201.13011 (2022)
  • [17] Z Xie, QY Tang, Y Cai, M Sun, P Li, On the power-law spectrum in deep learning: A bridge to protein science, arXiv preprint arXiv:2201.13011, 2 (2022)
  • [18] T Yu, X Li, Y Cai, M Sun, P Li, S2-mlp: Spatial-shift mlp architecture for vision, Proceedings of the IEEE/CVF winter conference on applications of computer vision, 297-306 (2022)
  • [19] Yunfeng Cai, Guanhua Fang, and Ping Li, A Note on Sparse Generalized Eigenvalue Problem, Advances in Neural Information Processing Systems, 34(2021), 23036-23048
  • [20] T Yu, X Li, Y Cai, M Sun, P Li, S$^2$-MLPv2: Improved Spatial-Shift MLP Architecture for Vision, arXiv preprint arXiv:2108.01072 (2021)
  • [21] Tan Yu, Xu Li, Yunfeng Cai, Mingming Sun, and Ping Li, Rethinking token-mixing MLP for MLP-based vision backbone, arXiv preprint arXiv:2106.14882 (2021)
  • [22] Jinxing Yu, Yunfeng Cai, Mingming Sun, and Ping Li, Mquade: a unified model for knowledge fact embedding, , 3442-3452 (2021)
  • [23] Yunfeng Cai, and Ping Li, A Blind Block Term Decomposition of High Order Tensors, Proceedings of the AAAI Conference on Artificial Intelligence, 35(2021), 8, 6868-6876
  • [24] Hexuan Liu, Yunfeng Cai, You-Lin Chen, and Ping Li, Ratio Trace Formulation of Wasserstein Discriminant Analysis, Advances in Neural Information Processing Systems, 33(2020), 16821-16832
  • [25] Yunfeng Cai, and Ping Li, Solving the Robust Matrix Completion Problem via a System of Nonlinear Equations, Proceedings of Machine Learning Research, 108(2020), 4162-4172
  • [26] Yunfeng Cai, and Ping Li, An Inverse-free Truncated Rayleigh-Ritz Method for Sparse Generalized Eigenvalue Problem, Proceedings of Machine Learning Research, 108(2020), 3460-3470
  • [27] Da-Wu Xiao, Wen-Hui Hu, Yunfeng Cai, and Nan Zhao, Magnetic Noise Enabled Biocompass, PHYSICAL REVIEW LETTERS, 124(2020)
  • [28] Tan Yu, Yunfeng Cai, and Ping Li, Toward Faster and Simpler Matrix Normalization via Rank-1 Update, Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020(2020), 203-219
  • [29] Yunfeng Cai, Rencang Li, Perturbation Analysis for Matrix Joint Block Diagonalization, LINEAR ALGEBRA AND ITS APPLICATIONS, 581(2019)
  • [30] C Cai, Y Cai, M Sun, Z Xu, Group representation theory for knowledge graph embedding, arXiv preprint arXiv:1909.05100 (2019)
  • [31] Yunfeng Cai, Zhigang Jia, Zhengjian Bai, Perturbation analysis of an eigenvector-dependent nonlinear eigenvalue problem with applications, BIT NUMERICAL MATHEMATICS, 60(2019)
  • [32] Yunfeng Cai, Guanghui Cheng, Decai Shi, Solving the general joint block diagonalization problem via linearly independent eigenvectors of a matrix polynomial, NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, 26(2019)
  • [33] Yunfeng Cai, Zhaojun Bai, John E. Pask and N. Sukumar, Convergence analysis of a locally accelerated preconditioned steepest descent method for Hermitian-definite generalized eigenvalue problems, JOURNAL OF COMPUTATIONAL MATHEMATICS, 36(2018)
  • [34] Jifei Miao, Guanghui Cheng, Yunfeng Cai, Jing Xia, Approximate joint singular value decomposition algorithm based on Givens-like rotation, IEEE Signal Processing Letters, 25(2018)
  • [35] Yunfeng Cai, Lei-Hong Zhang, Zhaojun Bai, Ren-Cang Li, On an Eigenvector-Dependent Nonlinear Eigenvalue Problem, SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 39(2018)
  • [36] Yunfeng Cai, Tiejun Li, Jiushu Shao, Zhiming Wang, Analysis of the Closure Approximation for a Class of Stochastic Differential Equations, Numerical Mathematics-Theory Methods and Applications, 10(2017)
  • [37] Jiang Qian, Yunfeng Cai, Delin Chu, Roger C. E. Tan, Eigenvalue Embedding of Undamped Vibroacoustic Systems with No-spillover, SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 38(2017)
  • [38] Yunfeng Cai, Chengyu Liu, An Algebraic Approach to Nonorthogonal General Joint Block Diagonalization, SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 38(2017)
  • [39] Hao Li, Yunfeng Cai, A Jacobi-Davidson type method for computing real eigenvalues of the quadratic eigenvalue problem, CALCOLO, 53(2015)
  • [40] Decai Shi, Yunfeng Cai, Shufang Xu, Some perturbation results for a normalized Non-Orthogonal Joint Diagonalization problem, LINEAR ALGEBRA AND ITS APPLICATIONS, 484(2015)
  • [41] Zhen-Chen Guo, Jiang Qian, Yun-feng Cai, Shu-fang Xu, Refined Schur Method for Robust Pole Assignment with Repeated Poles, IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 61(2015)
  • [42] Peichang Guo, Yunfeng Cai, Jiang Qian, Shufang Xu, Perturbation analysis of the extinction probability of a Markovian binary tree, LINEAR ALGEBRA AND ITS APPLICATIONS, 475(2015)
  • [43] Yunfeng Cai, Decai Shi, Shufang Xu, A matrix polynomial spectral approach for general joint block diagonalization, SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 36(2015)
  • [44] Hao Li, Yunfeng Cai, Solving the real eigenvalues of Hermitian quadratic eigenvalue problems via bisection, Electronic Journal of Linear Algebra, 30(2015)
  • [45] Zhenchen Guo, Yunfeng Cai, Jiang Qian, Shu-fang Xu, A modified Schur method for robust pole assignment in state feedback control, Automatica, 52(2015)
  • [46] Yunfeng Cai, Zhaojun Bai, John E. Pask, N. Sukumar, Hybrid preconditioning for iterative diagonalization of ill-conditioned generalized eigenvalue problems in electronic structure calculations, JOURNAL OF COMPUTATIONAL PHYSICS, 255(2013)
  • [47] Xin Lu, Shufang Xu, Yunfeng Cai, Partial derivatives of the eigen-triplet of the quadratic eigenvalue problem depending on several parameters, APPLIED MATHEMATICS AND COMPUTATION, 219(2013)
  • [48] Yunfeng Cai, Jiang Qian, Shufang Xu , Robust partial pole assignment problem for high order control systems, Automatica, 48(2012)
  • [49] Yunfeng Cai, Jian Qian, Shufang Xu, The formulation and numerical method for partial quadratic eigenvalue assignment problems, NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS, 18(2011)
  • [50] Yunfeng Cai, Shufang Xu, A new eigenvalue embedding approach for finite element model updating, TAIWANESE JOURNAL OF MATHEMATICS, 14(2010)
  • [51] Yunfeng Cai, Shufang Xu, On a quadratic inverse eigenvalue problem, INVERSE PROBLEMS, 25(2009)
  • [52] Yunfeng Cai, Yuencheng Kuo, Wenwei Lin, Shufang Xu, Solutions to a quadratic inverse eigenvalue problem, LINEAR ALGEBRA AND ITS APPLICATIONS, 430(2009)
  • [53] Y Cai, X Li, M Sun, P Li, Recovering Linear Causal Models with Latent Variables via Cholesky Factorization of Covariance Matrix, arXiv preprint arXiv:2311.00674 (2023)
  • [54] T Yu, J Li, Y Cai, P Li, Constructing orthogonal convolutions in an explicit manner, International Conference on Learning Representations (2022)
  • [55] Y Cai, G Fang, P Li, Sensitivity of Under-Determined Linear System, IEEE International Symposium on Information Theory (ISIT), 2267-2272 (2022)
  • [56] M Zhao, Z Jia, Y Cai, X Chen, D Gong, Advanced variations of two-dimensional principal component analysis for face recognition, Neurocomputing, 452, 653-664 (2021)
  • [57] Y Cai, P Li, Tensor Completion via Tensor Networks with a Tucker Wrapper, arXiv preprint arXiv:2010.15819 (2020)
  • [58] X Chen, Z Jia, Y Cai, M Zhao, Relaxed 2-D Principal Component Analysis by LNorm for Face Recognition, Intelligent Computing Theories and Application (2019)
  • [59] J Miao, G Cheng, Y Cai, J Xia, Approximate joint singular value decomposition algorithm based on givens-like rotation, IEEE Signal Processing Letters, 25(5), 620-624 (2018)
  • [60] Y Cai, J Qian, On some inverse eigenvalue problems of quadratic palindromic systems, arXiv preprint arXiv:1606.03840 (2016)
  • [61] H Li, Y Cai, A Jacobi–Davidson type method for computing real eigenvalues of the quadratic eigenvalue problem, Calcolo, 53(4), 737-749 (2016)
  • [62] D Shi, Y Cai, S Xu, A Polynomial Eigenproblem Approach for General Joint Block Diagonalization (2013)
  • [63] T Yu, Y Cai, P Li, Fast Bilinear Matrix Normalization via Rank-1 Update
  • [64] S Shao, B LiChen, T Ye, Y Cai, K Wu, Z Xie, The Blessing of Smooth Initialization for Video Diffusion Models
  • [65] X Li, Y CAI, M Sun, P Li, Causal Discovery via Cholesky Factorization
  • [66] Z Xie, X Yang, Y Yang, Q Sun, Y Jiang, H Wang, Y Cai, M Sun, Stochastic Structural SIMilarity and Its Unreasonable Effectiveness for Neural Field
  • [67] YF Cai, J Qian, SF Xu, A New Approach for Partial Quadratic Eigenvalue Assignment Problem
  • [68] Y Cai, Z Wang, T Li, J Shao, Analysis of the Closure Approximation for a Class
  • [69] Z Xie, QY Tang, Y CAI, M Sun, On the Power-Law Hessian Spectra in Deep Learning
  • [70] Yufei Gu, Qian-Yuan Tang, Yunfeng Cai, Mingming Sun, Ping Li, zhou Xun, Zeke Xie, Investigating the Overlooked Hessian Structure: From CNNs to LLMs, ICML

We are recruiting postdocs. Send me an email with your CV if you are interested. 

 

Update Time: 2025-08-12 17:00:06


Beijing Institute of Mathematical Sciences and Applications
CONTACT

No. 544, Hefangkou Village Huaibei Town, Huairou District Beijing 101408

北京市怀柔区 河防口村544号
北京雁栖湖应用数学研究院 101408

Tel. 010-60661855 Tel. 010-60661855
Email. administration@bimsa.cn

Copyright © Beijing Institute of Mathematical Sciences and Applications

京ICP备2022029550号-1

京公网安备11011602001060 京公网安备11011602001060