paper-with-me

Papers

Memory-Efficient Continual Learning with CLIP Models

2026-05-05 · Ryan King, Gang Li, Bobak Mortazavi, Tianbao Yang arxiv

Contrastive Language-Image Pretraining (CLIP) models excel at understanding image-text relationships but struggle with adapting to new data without forgetting prior knowledge. To address this, models are typically fine-tuned using both new task data and a memory buffer of past tasks. However, CLIP's contrastive loss suffers when the memory buffer is small, leading to performance degradation on previous tasks. We propose a memory-efficient, distributionally robust method that dynamically reweights losses per class during training. Our approach, tested on class incremental settings (CIFAR-100, ImageNet1K) and a domain incremental setting (DomainNet) adapts CLIP models quickly while minimizing catastrophic forgetting, even with minimal memory usage.

📄 PDF Abstract BibTeX arXiv:2605.03866

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Similar Papers 제목 키워드 기반

kNN-CLIP: Retrieval Enables Training-Free Segmentation on Continually Expanding Large Vocabularies

2024-04-15 · Zhongrui Gui, Shuyang Sun, Runjia Li, Jianhao Yuan 외

Continual segmentation has not yet tackled the challenge of improving open-vocabulary segmentation models with training data for accurate segmentation across large, continually expanding vocabularies. We discover that tr…

Panoptic SegmentationRetrievalSegmentationZero Shot Segmentation

CLIP model is an Efficient Continual Learner

2022-10-06 · Vishal Thengane, Salman Khan, Munawar Hayat, Fahad Khan

The continual learning setting aims to learn new tasks over time without forgetting the previous ones. The literature reports several significant efforts to tackle this problem with limited or no access to previous task …

Continual LearningIncremental LearningKnowledge Distillationmodel

LVP-CLIP:Revisiting CLIP for Continual Learning with Label Vector Pool

2024-12-08 · Yue Ma, Huantao Ren, Boyu Wang, Jingang Jin 외

Continual learning aims to update a model so that it can sequentially learn new tasks without forgetting previously acquired knowledge. Recent continual learning approaches often leverage the vision-language model CLIP f…

Continual LearningIncremental LearningLanguage ModelingLanguage Modelling

AttriCLIP: A Non-Incremental Learner for Incremental Knowledge Learning

2023-05-19 · CVPR 2023 1 · Runqi Wang, Xiaoyue Duan, Guoliang Kang, Jianzhuang Liu 외

Continual learning aims to enable a model to incrementally learn knowledge from sequentially arrived data. Previous works adopt the conventional classification architecture, which consists of a feature extractor and a cl…

AttributeContinual LearningLanguage Modelling

LADA: Scalable Label-Specific CLIP Adapter for Continual Learning

2025-05-29 · Mao-Lin Luo, Zi-Hao Zhou, Tong Wei, Min-Ling Zhang

Continual learning with vision-language models like CLIP offers a pathway toward scalable machine learning systems by leveraging its transferable representations. Existing CLIP-based methods adapt the pre-trained image e…

Continual Learning