In continual learning, the learner faces a stream of data whose distribution changes over time. Modern neural networks are known to suffer under this setting, as they quickly forget previously acquired knowledge. To address such catastrophic forgetting, many continual learning methods implement different types of experience replay, re-learning on past data stored in a small buffer known as episodic memory. In this work, we complement experience replay with a new objective that we call anchoring, where the learner uses bilevel optimization to update its knowledge on the current task, while keeping intact the predictions on some anchor points of past tasks. These anchor points are learned using gradient-based optimization to maximize forgetting, which is approximated by fine-tuning the currently trained model on the episodic memory of past tasks. Experiments on several supervised learning benchmarks for continual learning demonstrate that our approach improves the standard experience replay in terms of both accuracy and forgetting metrics and for various sizes of episodic memories.
Dynamically Anchored Prompting for Task-Imbalanced Continual Learning
Xiamen University（厦门大学）；Univ. Grenoble Alpes（格勒诺布尔大学）；Inria（法国国家信息与自动化研究所）；CNRS（法国国家科学研究中心）；Grenoble INP（格勒诺布尔理工学院）；LJK（让·昆廷诺实验室）；Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ)（广东省人工智能与数字经济实验室（深圳））
Canonicalized Stable-List Replay for Private Federated Continual Learning over Language-Model Embeddings
Department of Computer Science, Iowa State University（爱荷华州立大学计算机科学系）；Department of Computer Science & Engineering, BRAC University（BRAC大学计算机科学与工程系）；Department of Civil, Construction & Environmental Engineering, Iowa State University（爱荷华州立大学土木、建设与环境工程系）
Handling Spatial-Temporal Data Heterogeneity for Federated Continual Learning via Tail Anchor
Southwestern University of Finance and Economics（西南财经大学）；University of Electronic Science and Technology of China（电子科技大学）；WeBank（微众银行）；Southwest Jiaotong University（西南交通大学）；Hong Kong University of Science and Technology（香港科技大学）
Department of Computer Science and Technology, Tsinghua University, China（计算机科学与技术系，清华大学，中国）；College of Computer Science, Zhejiang University, China（浙江大学计算机科学学院，中国）；College of Computer Science, Beijing University of Posts and Telecommunications, China（北京邮电大学计算机科学学院，中国）
本文提出连续GUI代理任务，通过引入GUI-Anchoring in Flux框架，解决GUI分布变化时持续学习稳定性问题，实验显示其优于现有方法。
Adapting model parameters to incoming streams of data is a crucial factor to deep learning scalability. Interestingly, prior continual learning strategies in online settings inadvertently anchor their updated parameters to a local parameter subspace to remember old tasks, else drift away from the subspace and forget. From this observation, we formulate a trade-off between constructing multiple parameter modes and allocating tasks per mode. Mode-Optimized Task Allocation (MOTA), our contributed adaptation strategy, trains multiple modes in parallel, then optimizes task allocation per mode. We e…
Reasoning Portability: Guiding Continual Learning for MLLMs in the RLVR Era
Shenzhen Graduate School of Peking University（北京大学深圳研究生院）；Centre for Artificial Intelligence and Robotics, HKISI, CAS（香港科学院人工智能与机器人研究中心）；Peng Cheng Laboratory（鹏城实验室）
Overcoming Domain Drift in Online Continual Learning
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；Suzhou University of Science and Technology（苏州科技大学）；Youtu Lab, Tencent（腾讯优图实验室）；College of Intelligence and Computing, Tianjin University（天津大学智能与计算学部）
SAE-FD: Sparse Autoencoder Feature Distillation for Continual Learning of Large Language Models
The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Nanjing University of Aeronautics and Astronautics（南京航空航天大学）；The 63rd Research Institute, National University of Defense Technology, Nanjing（国防科技大学第六十三研究所，南京）
Advanced Research Lab, NavInfo Europe（四维图新欧洲高级研究实验室）
Continual learning (CL) in the brain is facilitated by a complex set of mechanisms. This includes the interplay of multiple memory systems for consolidating information as posited by the complementary learning systems (CLS) theory and synaptic consolidation for protecting the acquired knowledge from erasure. Thus, we propose a general CL method that creates a synergy between SYNaptic consolidation and dual memory Experience Replay (SYNERgy). Our method maintains a semantic memory that accumulates and consolidates information across the tasks and interacts with the episodic memory for effective…
Dynamic Dual Buffer with Divide-and-Conquer Strategy for Online Continual Learning
Department of Computing, Imperial College London（帝国理工学院计算机系）；Department of Bioengineering, Imperial College London（帝国理工学院生物工程系）；School of Information and Software Engineering, University of Electronic Science and Technology of China（电子科技大学信息与软件工程学院）
GAP-Prompt: Gated Adaptive Prompting for Efficient Continual Learning
Université d’Orléans（奥尔良大学）；INSA CVL（中央卢瓦尔河谷国立应用科学学院）；CY Cergy Paris University（塞尔吉-蓬图瓦兹大学）；ENSEA（法国高等电子与数字学院）；CNRS（法国国家科学研究中心）；Université de Technologie de Troyes（特鲁瓦技术大学）
HPCR: Holistic Proxy-based Contrastive Replay for Online Continual Learning
Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳））；Centre for Frontier AI Research, Institute of High Performance Computing, A*STAR（新加坡高性能计算研究院前沿人工智能研究中心（新加坡科技研究局））
Towards Real-World Test-Time Adaptation: Tri-Net Self-Training with Balanced Normalization
South China University of Technology（华南理工大学）；Institute for Infocomm Research, A*STAR（新加坡科技研究局信息通信研究所）；School of Data Science, The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳）数据科学学院）
Sequential Learning and Catastrophic Forgetting in Differentiable Resistor Networks
Mathematics Applications Consortium for Science and Industry (MACSI), Department of Mathematics and Statistics, University of Limerick, Limerick, Ireland（数学应用科学与工业联盟（MACSI）、数学与统计学系、利默里克大学、利默里克、爱尔兰）
CONTRAST: Continual Multi-source Adaptation to Dynamic Distributions
Brookhaven National Laboratory（布鲁克海文国家实验室）；University of California, Riverside（加州大学河滨分校）；AWS AI Labs（亚马逊云科技人工智能实验室）；University of Michigan, Ann Arbor（密歇根大学安娜堡分校）
Learning Stateful Predictive Knowledge From Experience
University College London（伦敦大学学院）；National University of Singapore（新加坡国立大学）；Chinese Academy of Sciences（中国科学院）；Brown University（布朗大学）；University of Bristol（布里斯托尔大学）；University of Edinburgh（爱丁堡大学）；Zhongguancun Academy（中关村科学院）；The Yangtze River Delta（长三角）
University of Texas at Austin（德克萨斯大学奥斯汀分校）；Michigan State University（密歇根州立大学）；MIT-IBM Watson AI Lab, IBM Research（IBM研究MIT-IBM沃森人工智能实验室）；University of California, Santa Barbara（加州大学圣巴巴拉分校）
Class-incremental learning (CIL) suffers from the notorious dilemma between learning newly added classes and preserving previously learned class knowledge. That catastrophic forgetting issue could be mitigated by storing historical data for replay, which yet would cause memory overheads as well as imbalanced prediction updates. To address this dilemma, we propose to leverage "free" external unlabeled data querying in continual learning. We first present a CIL with Queried Unlabeled Data (CIL-QUD) scheme, where we only store a handful of past training samples as anchors and use them to query re…
