持续学习与提示学习

在现实世界的应用中，数据环境是动态变化的，模型需要能够持续学习新知识，同时保留已有知识而不会遗忘。我们的研究涉及基于任务的不平衡持续学习、增量学习中的概念漂移处理和领域泛化，旨在提高模型在动态和不平衡数据环境中的稳定性和适应性。包括任务不平衡的持续学习、概念漂移的持续学习和提示学习等子方向。

相关成果：

• 任务不平衡的持续学习：

• [1] Dynamically Anchored Prompting for Task-Imbalanced Continual Learning, Chenxing Hong, Yan Jin, Zhiqi Kang, Yizhou Chen, Mengke Li, Yang Lu*, and Hanzi Wang, International Joint Conference on Artificial Intelligence (IJCAI), Jeju, Korea, August 3-9, 2024. (CCF-A)

• 概念漂移的持续学习:

  [1] Adaptive Middle Modality Alignment Learning for Visible-Infrared Person Re-identification, Yukang Zhang, Yan Yan, Yang Lu, and Hanzi Wang, International Journal of Computer Vision (IJCV), 2024. (JCR 1区 / CCF-A)

  [2] Dynamic Weighted Majority for Incremental Learning of Imbalanced Data Streams with Concept Drift, Yang Lu, Yiu-ming Cheung, and Yuan Yan Tang, International Joint Conference on Artificial Intelligence (IJCAI), pp. 2393-2399, Melbourne, Australia, August 19-25, 2017. (CCF-A)

• 提示学习:

  [1] PRO-VPT: Distribution-Adaptive Visual Prompt Tuning via Prompt Relocation, Chikai Shang, Mengke Li, Yiqun Zhang, Zhen Chen, Jinlin Wu, Fangqing Gu, Yang Lu*, and Yiu-ming Cheung, IEEE/CVF International Conference on Computer Vision (ICCV), Honolulu, Hawaii, October 19 - 23, 2025. (CCF-A)

  [2] FATE: A Prompt-Tuning-Based Semi-Supervised Learning Framework for Extremely Limited Labeled Data, Hezhao Liu, Mengke Li, Yiqun Zhang, Shreyank N Gowda, Yang Lu*, Chen Gong, and Hanzi Wang, ACM Multimedia (MM), Dublin, Ireland, October 27 - 31, 2025. (CCF-A)

  [3] Transitive Vision-Language Prompt Learning for Domain Generalization, Liyuan Wang, Yan Jin, Zhen Chen, Jinlin Wu, Mengke Li, Yang Lu*, and Hanzi Wang, IEEE Transactions on Emerging Topics in Computational Intelligence (TETCI), 2025. (JCR 1区)