CVPR 2024 accepted papers on Continual learning
48 of the 2,716 papers accepted at CVPR 2024 (1.8%) are on Continual learning, compared with 1.9% at CVPR 2023. They fall into 4 topics.
See CVPR 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Class-incremental · 17 papers
- Expandable Subspace Ensemble for Pre-Trained Model-Based Class-Incremental Learning (117 citations)
- Gradient Reweighting: Towards Imbalanced Class-Incremental Learning (57 citations)
- OrCo: Towards Better Generalization via Orthogonality and Contrast for Few-Shot Class-Incremental Learning (45 citations)
- SDDGR: Stable Diffusion-based Deep Generative Replay for Class Incremental Object Detection (42 citations)
- Class Incremental Learning with Multi-Teacher Distillation (31 citations)
- FCS: Feature Calibration and Separation for Non-Exemplar Class Incremental Learning (30 citations)
- Resurrecting Old Classes with New Data for Exemplar-Free Continual Learning (27 citations)
- Generative Multi-modal Models are Good Class Incremental Learners (26 citations)
- Solving the Catastrophic Forgetting Problem in Generalized Category Discovery (20 citations)
- Continual Segmentation with Disentangled Objectness Learning and Class Recognition (15 citations)
- Text-Enhanced Data-free Approach for Federated Class-Incremental Learning (13 citations)
- Dual-Consistency Model Inversion for Non-Exemplar Class Incremental Learning (12 citations)
and 5 more in the interactive map.
Online and replay-based · 12 papers
- Towards Efficient Replay in Federated Incremental Learning (34 citations)
- Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning (27 citations)
- Incremental Residual Concept Bottleneck Models (21 citations)
- Orchestrate Latent Expertise: Advancing Online Continual Learning with Multi-Level Supervision and Reverse Self-Distillation (18 citations)
- Learning Equi-angular Representations for Online Continual Learning (12 citations)
- Online Task-Free Continual Generative and Discriminative Learning via Dynamic Cluster Memory (11 citations)
- Continual Learning for Motion Prediction Model via Meta-Representation Learning and Optimal Memory Buffer Retention Strategy (9 citations)
- BrainWash: A Poisoning Attack to Forget in Continual Learning (8 citations)
- Improving Plasticity in Online Continual Learning via Collaborative Learning (8 citations)
- Defense without Forgetting: Continual Adversarial Defense with Anisotropic & Isotropic Pseudo Replay (7 citations)
- RCL: Reliable Continual Learning for Unified Failure Detection (6 citations)
- Dual-Enhanced Coreset Selection with Class-wise Collaboration for Online Blurry Class Incremental Learning (3 citations)
Continual fine-tuning of LLMs · 12 papers
- Boosting Continual Learning of Vision-Language Models via Mixture-of-Experts Adapters (124 citations)
- InfLoRA: Interference-Free Low-Rank Adaptation for Continual Learning (69 citations)
- Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners (37 citations)
- Continual Self-supervised Learning: Towards Universal Multi-modal Medical Data Representation Learning (36 citations)
- Convolutional Prompting meets Language Models for Continual Learning (20 citations)
- Interactive Continual Learning: Fast and Slow Thinking (20 citations)
- Semantically-Shifted Incremental Adapter-Tuning is A Continual ViTransformer (20 citations)
- Continual Forgetting for Pre-trained Vision Models (15 citations)
- Enhancing Visual Continual Learning with Language-Guided Supervision (13 citations)
- Brain Decodes Deep Nets (10 citations)
- Overcoming Generic Knowledge Loss with Selective Parameter Update (10 citations)
- Towards Backward-Compatible Continual Learning of Image Compression (6 citations)
Domain and prompt-based · 7 papers
- Each Test Image Deserves A Specific Prompt: Continual Test-Time Adaptation for 2D Medical Image Segmentation (44 citations)
- Consistent Prompting for Rehearsal-Free Continual Learning (38 citations)
- Distribution-aware Knowledge Prototyping for Non-exemplar Lifelong Person Re-identification (22 citations)
- Continual-MAE: Adaptive Distribution Masked Autoencoders for Continual Test-Time Adaptation (21 citations)
- ECLIPSE: Efficient Continual Learning in Panoptic Segmentation with Visual Prompt Tuning (17 citations)
- Learning Continual Compatible Representation for Re-indexing Free Lifelong Person Re-identification (15 citations)
- A Versatile Framework for Continual Test-Time Domain Adaptation: Balancing Discriminability and Generalizability (6 citations)