Long-Kai Huang 

        Email: HLONGKAI at gmail dot com


I am a senior researcher in Machine Learning Center, Tencent AI Lab. Prior to that, I obtained my Ph.D. in SCSE at NTU, advised by Prof. Sinno Jialin Pan.  I received B.Eng. from Sun Yat-Sen University and did my undergraduate thesis with Prof. Wei-Shi Zheng.

I am generally interested in understanding the knowledge and the generalization mechanism in machine learning models and building general continual learning systems and focus on 1) Knowledge Transfer: continual learning, transfer learning, meta-learning; 2) Model Editing in Transformers; 3) Model Interpretability in Transformers and 4) their applications to real-world applications such as drug discovery.


Recent Publications  |  Full Publications (Google Scholar) 

# indicates equal contributions. * indicates corresponding authors. Underlined authors are research interns mentored by me.     


Mitigating Catastrophic Forgetting in Online Continual Learning by Modeling Previous Task Interrelations Via Pareto Optimization

   Yichen Wu#, Hong Wang#, Peilin Zhao, Yefeng Zheng, Ying Wei*, Long-Kai Huang*

    International Conference on Machine Learning (ICML), 2024

Meta Continual Learning Revisited: Implicitly Enhancing Online Hessian Approximation via Variance Reduction [paper]

Yichen Wu, Long-Kai Huang*, Renzhen Wang, Deyu Meng, Ying Wei*

International Conference on Learning Representations (ICLR) 2024 (Oral, Outstanding Paper Award Honorable Mention

Latent Trajectory Learning for Limited Timestamps under Distribution Shift over Time [paper]

Qiuhao Zeng, Changjian Shui, Long-Kai Huang, Peng Liu, Xi Chen, Charles Ling, Boyu Wang

International Conference on Learning Representations (ICLR) 2024  (Oral

  Deep domain adversarial neural network for the deconvolution of cell type mixtures in tissue proteome profiling

Fang Wang#, Fan Yang#, Long-Kai Huang, Wei Li, Jiangning Song, Robin B Gasser, Ruedi Aebersold, Guohua Wang, Jianhua Yao

Nature Machine Intelligence 2023

Retaining Beneficial Information from Detrimental Data for Neural Network Repair [paper]

Long-Kai Huang, Peilin Zhao, Junzhou Huang, Sinno Jialin Pan

Advances in Neural Information Processing Systems (NeurIPS) 2023

Secure Out-of-Distribution Task Generalization with Energy-Based Models [paper]

Shengzhuang Chen, Long-Kai Huang, Jonathan Richard Schwarz, Yilun Du, Ying Wei

Advances in Neural Information Processing Systems (NeurIPS) 2023

Concept-wise Fine-tuning Matters in Preventing Negative Transfer [paper]

Yunqiao Yang, Long-Kai Huang, Ying Wei

International Conference on Computer Vision (ICCV) 2023

Improving Generalizability of Graph Anomaly Detection Models via Data Augmentation

Shuang Zhou, Xiao Huang, Ninghao Liu, Huachi Zhou, Fu-Lai Chung, Long-Kai Huang*

IEEE Transactions on Knowledge and Data Engineering (TKDE) 2023

  DrugOOD: Out-of-Distribution Dataset Curator and Benchmark for AI-Aided Drug Discovery–a Focus on Affinity Prediction Problems with Noise Annotations

Yuanfeng Ji, Lu Zhang [and 15 others, including Long-Kai Huang]

In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI) 2023  (Oral

Improving Task-Specific Generalization in Few-Shot Learning via Adaptive Vicinal Risk Minimization [paper] [code]

             Long-Kai Huang, Ying Wei

             Advances in Neural Information Processing Systems (NeurIPS) 2022

Adversarial Task Up-sampling for Meta-learning [paper]

             Yichen Wu, Long-Kai Huang*, Ying Wei*

             Advances in Neural Information Processing Systems (NeurIPS) 2022

Frustratingly Easy Transferability Estimation [paper] [code]

Long-Kai Huang, Junzhou Huang, Yu Rong, Qiang Yang, Ying Wei

International Conference on Machine Learning (ICML), 2022

    Improving Generalization in Meta-learning via Task Augmentation [paper] [code]

Huaxiu Yao#, Long-Kai Huang#, Linjun Zhang, Ying Wei, Li Tian, James Zou, Junzhou Huang, Zhenhui Li

International Conference on Machine Learning (ICML), 2021

   Communication-Efficient Distributed PCA by Riemannian Optimization [paper]

Long-Kai Huang and Sinno Jialin Pan

International Conference on Machine Learning (ICML), 2020