Hi! I am a fourth-year PhD student at PAttern Recognition and NEural Computing (PARNEC) group, NUAA where I am fortunate enough to be advised by Prof. Songcan Chen who is also the leader of PARNEC group. Currently, I work as a visiting student in Spain closely with Bartłomiej Twardowski and Alexandra Gomez-Villa in LAMP team, led by Dr. Joost van de Weijer in Computer Vision Center, Autonomous University of Barcelona, Spain.

My research interests lie broadly in Machine Learning, especially Continual Learning, Weakly Supervised Learning, Backdoor Safety and Multi-Modal LLMs – check my Google Scholar for more information. I am currently seeking new opportunities as a postdoc or researcher in the industry, approximately starting in October, 2026. I am open to discussing potential collaborations or openings that align with my research expertise. Feel free to reach out if you’d like to chat!

Recent News

• May, 2026 Online Continual Learning with Dynamic Label Hierarchies. accepted to ICML 2026.

• May, 2026 Shortcut-Resistant CAM Distillation for Long-Tailed Recognition. accepted to ICML 2026.

• March, 2026 Filter, Obstruct and Dilute: Defending Against Backdoor Attacks on Semi-Supervised Learning. accepted to IEEE Transactions on Information Forensics & Security (TIFS).

• Sep, 2025 Global Pre-fixing, Local Adjusting: A Simple yet Effective Contrastive Strategy for Continual Learning. accepted to Frontiers of Computer Science (FCS).

• June, 2025 You Never Walk Alone: A Generalizable and Non-Parametric Structure Learning Framework. accepted to IEEE Transactions on Neural Networks and Learning Systems (TNNLS).

• May, 2025 Cut out and Replay: A Simple yet Versatile Strategy for Multi-Label Online Continual Learning. accepted to ICML 2025.

• April, 2025 LoD: Loss-difference OOD Detection by Intentionally Label-Noisifying Unlabeled Wild Data. accepted to IJCAI 2025.

• April, 2025 Filter, Obstruct and Dilute: Defending Against Backdoor Attacks on Semi-Supervised Learning. is online.

• Feb, 2025 Enhanced Adaptive Gradient Algorithms for Nonconvex-PL Minimax Optimization. accepted to AISTATS 2025.

• Sep, 2024 Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online Continual Learning. accepted to NeurIPS 2024.

• Jan, 2024 Pushing One Pair of Labels Apart Each Time in Multi-Label Learning: From Single Positive to Full Labels. accepted to SCIENCE CHINA Information Sciences.

• Jan, 2024 Adaptive federated minimax optimization with lower complexities. accepted to AISTATS 2024.

• Dec, 2023 Unlocking the Power of Open Set: A New Perspective for Open-set Noisy Label Learning. accepted to AAAI 2024.

• Sep, 2023 Beyond Myopia: Learning from Positive and Unlabeled Data through Holistic Predictive Trends. accepted to NeurIPS 2023.

Academic Services

• Conference Reviewing: NeurIPS 2024/2025/2026, ICLR 2025/2026/2026, ICML 2025/2026, AISTATS 2025/2026, ACML 2025, AAAI 2026, CVPR 2025/2026.
• Journal Reviewing: TMLR, TKDD.