TPAMI 2025 papers on Continual learning
22 of the 932 papers published in TPAMI 2025 (2.4%) are on Continual learning, compared with 1.7% at TPAMI 2024. They fall into 4 topics.
See TPAMI 2025 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Online and replay-based · 8 papers
- Simple Lifelong Learning Machines (26 citations)
- Revisiting Flatness-Aware Optimization in Continual Learning With Orthogonal Gradient Projection (9 citations)
- Graph Memory Learning: Imitating Lifelong Remembering and Forgetting of Brain Networks (4 citations)
- Developing Evolving Adaptability in Biological Intelligence: A Novel Biologically-Inspired Continual Learning Model for Video Saliency Prediction (2 citations)
- Continual Unsupervised Generative Modeling (1 citation)
- Schedule-Robust Continual Learning (1 citation)
- Release the Potential of Memory Buffer in Continual Learning: A Dynamic System Perspective
- Rehearsal-Free and Efficient Continual Learning for Cross-Domain Face Anti-Spoofing
Continual fine-tuning of LLMs · 7 papers
- Learning Without Forgetting for Vision-Language Models (115 citations)
- HiDe-PET: Continual Learning via Hierarchical Decomposition of Parameter-Efficient Tuning (18 citations)
- Gradient Projection for Continual Parameter-Efficient Tuning (15 citations)
- Towards Natural Machine Unlearning (13 citations)
- MoE-Adapters++: Toward More Efficient Continual Learning of Vision-Language Models Via Dynamic Mixture-of-Experts Adapters (6 citations)
- VPT-NSP 2 ++: Importance-Aware Visual Prompt Tuning in Null Space for Continual Learning (3 citations)
- Task-Distributionally Robust Data-Free Meta-Learning (1 citation)
Class-incremental · 4 papers
- Few-Shot Class-Incremental Learning for Classification and Object Detection: A Survey (109 citations)
- Replay Without Saving: Prototype Derivation and Distribution Rebalance for Class-Incremental Semantic Segmentation (15 citations)
- Replay Master: Automatic Sample Selection and Effective Memory Utilization for Continual Semantic Segmentation (6 citations)
- PASS++: A Dual Bias Reduction Framework for Non-Exemplar Class-Incremental Learning
Domain and prompt-based · 3 papers
- Distribution-Aware Knowledge Aligning and Prototyping for Non-Exemplar Lifelong Person Re-Identification (6 citations)
- Long Short-Term Knowledge Decomposition and Consolidation for Lifelong Person Re-Identification (6 citations)
- Bi-C 2 R: Bidirectional Continual Compatible Representation for Re-Indexing Free Lifelong Person Re-Identification
