Education Background
- Ph.D., Cyberspace Security, Institute of Information
Engineering, Chinese Academy of Sciences, 2019.09~2024.06
- B.Eng., Communication Engineering, Beijing University of Posts and Telecommunications, 2015.09~2019.06
Email: shikunli.ml@gmail.com, lishikun19@mails.ucas.ac.cn
Research Areas
My research interests lie in
trustworthy machine learning and
data-centric AI. Recently, I focus on the following topics:
- Data Selection for Large Language Model SFT.
- Safety of Large Language Model.
- Learning with noisy labels and learning from multiple label sources.
- Dataset distillation and coreset selection.
Selected Publications
*= equal contribution
- Multi-Label Noise Transition Matrix with Label Correlations: Theory and Algorithm
arXiv Preprint
Shikun Li, Xiaobo Xia, Hansong Zhang, Shiming Ge, Tongliang Liu
- DANCE: Dual-View Distribution Alignment for Dataset Condensation
International Joint Conference on Artificial Intelligence (IJCAI), 2024
Hansong Zhang, Shikun Li, Fanzhao Lin, Weiping Wang, Zhenxing Qian, Shiming Ge.
- Transferring Annotator- and Instance-dependent Transition Matrix for Learning from Crowds
IEEE Transactions on Pattern Analysis and Machine Intelligence (IEEE T-PAMI), 2024
Shikun Li, Xiaobo Xia, Jiankang Deng, Shiming Ge, Tongliang Liu
- Multimodal Composition Example Mining for Composed Query Image Retrieval
IEEE Transactions on Image Processing (IEEE T-IP), 2024
Gangjian Zhang*, Shikun Li*, Shikui Wei, Shiming Ge, Na Cai, Yao Zhao
- M3D: Dataset Condensation by Minimizing Maximum Mean Discrepancy
AAAI Conference on Artificial Intelligence (AAAI), 2024
Hansong Zhang*, Shikun Li*, Pengju Wang, Dan Zeng, Shiming Ge.
- Coupled Confusion Correction: Learning from Crowds with Sparse Annotations
AAAI Conference on Artificial Intelligence (AAAI), 2024
Hansong Zhang, Shikun Li, Dan Zeng, Chenggang Yan, Shiming Ge.
- Federated Learning with Label-Masking Distillation
ACM International Conference on Multimedia (ACM MM), 2023, Oral
Jianghu Lu, Shikun Li, Kexin Bao, Pengju Wang, Zhenxing Qian, Shiming Ge.
- Estimating Noise Transition Matrix with Label Correlations for Noisy Multi-Label Learning
Annual Conference on Neural Information Processing Systems (NeurIPS), 2022, Spotlight
Shikun Li, Xiaobo Xia, Hansong Zhang, Yibing Zhan, Shiming Ge, Tongliang Liu
- Selective-Supervised Contrastive Learning with Noisy Labels
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022
Shikun Li, Xiaobo Xia, Shiming Ge, Tongliang Liu
- Trustable Co-Label Learning from Multiple Noisy Annotators
IEEE Transactions on Multimedia (IEEE T-MM), 2021
Shikun Li, Tongliang Liu, Dan Zeng, Shimng Ge
- Coupled-View Deep Classifier Learning from Multiple Noisy Annotators
AAAI Conference on Artificial Intelligence (AAAI), 2020
Shikun Li, Shiming Ge, Yingying Hua, Chunhui Zhang, Hao Wen, Tengfei Liu, Weiqiang Wang
Patents and Software
Awards and Honors
- CAS President's Award (Special Prize), 2024
- Beijing Excellent Graduate, 2024
- IEEE MMTC Best Jounal Paper Award, 2023
- Chinese National Scholarship for Doctoral Student, 2022
- ZhuLiYuehua Scholarship for Excellent Doctoral Student, 2022
- TML Research Excellence Award (Gold Prize), 2022
- IIE President's Scholarship, 2021 (Outstanding Prize), 2023 (Special Prize)
- Merit Student of UCAS, 2022,2023
- First Prize of VisDrone-SOT2019 Challenge, 2019
Academic Service
- Conference Reviewer: NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, AISTATS, AAAI, IJCAI, UAI, KDD, SDM, ICPR, ACCV, etc.
- Journal Reviewer: IEEE T-PAMI, IJCV, IEEE T-NNLS, IEEE T-MM, IEEE T-CSVT, ACM T-KDD, ACM Computing Surveys, etc.
京公网安备 11011602000736号