Lin Li
PhD student @ King's College London, Associate Member @ Sea AI Lab
Department of Informatics
King's College London
London, WC2B 4BG, UK
I am a PhD student in machine learning under the supervision of Dr Michael Spratling and Dr Dimitrios Letsios at the Department of Informatics, King’s College London. My PhD study is funded by King’s-China Scholarship Council PhD Scholarship. Prior to coming to King’s, I received a MSc degree in computing with my thesis advised by Professor Wayne Luk from Imperial College London. I also received a B.B.M. in finance with my thesis advised by Professor Zheng Qiao from Xiamen University.
My research interests include
- Trustworthy ML: robustness, safety and interpretability
- Data-centric ML: generative models for data augmentation
- AI+ Applications: healthcare, finance
Collaboration: my collaborators and I are looking for new mates to join the team to develop new methods for robust large multimodal models and using generative models for data augmentation. If interested, you are more than welcome to contact me to discuss.
news
Mar 27, 2024 | I was invited by AI Time to give a talk of our recent CVPR2024 publication: One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models. |
---|---|
Mar 4, 2024 | Our work, OODRobustBench: benchmarking and analyzing adversarial robustness under distribution shift, is accepted by ICLR 2024 Workshop Data-centric Machine Learning Research (DMLR) |
Feb 27, 2024 | One work, One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models, is accepted by CVPR2024! |
Dec 23, 2023 | I am invited to serve as reviewer for Internation Conference on Machine Learning (ICML) 2024. |
Dec 22, 2023 | I join the program committee of Workshop on Wearable Intelligence for Healthcare Robotics (WIHR): from Brain Activity to Body Movements at 2024 IEEE International Conference on Robotics and Automation (ICRA) in PACIFICO Yokohama, Japan. |
selected publications
- One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language ModelsIn IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024
- Data Augmentation Alone Can Improve Adversarial TrainingIn International Conference on Learning Representations, 2023
- Understanding and combating robust overfitting via input loss landscape analysis and regularizationPattern Recognition, 2023
- Large AI Models in Health Informatics: Applications, Challenges, and the FutureIEEE Journal of Biomedical and Health Informatics (JBHI), 2023
- OODRobustBench: benchmarking and analyzing adversarial robustness under distribution shiftICLR 2024 Workshop Data-centric Machine Learning Research (DMLR), 2024