paper-with-me

홈 › Papers

Beyond CLIP Generalization: Against Forward&Backward Forgetting Adapter for Continual Learning of Vision-Language Models

2025-05-12 · Songlin Dong, Chenhao Ding, Jiangyang Li, Jizhou Han, Qiang Wang, Yuhang He, Yihong Gong

This study aims to address the problem of multi-domain task incremental learning~(MTIL), which requires that vision-language models~(VLMs) continuously acquire new knowledge while maintaining their inherent zero-shot recognition capability. Existing paradigms delegate the testing of unseen-domain samples to the original CLIP, which only prevents the degradation of the model's zero-shot capability but fails to enhance the generalization of the VLM further. To this end, we propose a novel MTIL framework, named AFA, which comprises two core modules: (1) an against forward-forgetting adapter that learns task-invariant information for each dataset in the incremental tasks to enhance the zero-shot recognition ability of VLMs; (2) an against backward-forgetting adapter that strengthens the few-shot learning capability of VLMs while supporting incremental learning. Extensive experiments demonstrate that the AFA method significantly outperforms existing state-of-the-art approaches, especially in few-shot MTIL tasks, and surpasses the inherent zero-shot performance of CLIP in terms of transferability. The code is provided in the Supplementary Material.

📄 PDF Abstract BibTeX arXiv:2505.07690

Code (0)

등록된 구현이 없습니다.

Tasks

Continual LearningFew-Shot LearningIncremental LearningZero-Shot Learning

Methods 이 논문이 사용한 방법론

Adapter 설명 없음
CLIP Contrastive Language-Image Pre-training (CLIP), consisting of a simplified version of ConVIRT trained from scratch, is an efficient method of image representation learning…

Similar Papers 제목 키워드 기반

Beyond Human-Like Processing: Large Language Models Perform Equivalently on Forward and Backward Scientific Text

2024-11-17 · Xiaoliang Luo, Michael Ramscar, Bradley C. Love

The impressive performance of large language models (LLMs) has led to their consideration as models of human language processing. Instead, we suggest that the success of LLMs arises from the flexibility of the transforme…

Protein Design

Seeing the Arrow of Time

2014-06-01 · CVPR 2014 6 · Lyndsey C. Pickup, Zheng Pan, Donglai Wei, YiChang Shih 외

We explore whether we can observe Time's Arrow in a temporal sequence--is it possible to tell whether a video is running forwards or backwards? We investigate this somewhat philosophical question using computer vision an…

General ClassificationVideo Compression

Combining detection and tracking for human pose estimation in videos

2020-03-30 · CVPR 2020 6 · Manchen Wang, Joseph Tighe, Davide Modolo

We propose a novel top-down approach that tackles the problem of multi-person human pose estimation and tracking in videos. In contrast to existing top-down approaches, our method is not limited by the performance of its…

Pose EstimationPose Tracking

Reverse Thinking Makes LLMs Stronger Reasoners

2024-11-29 · Justin Chih-Yao Chen, Zifeng Wang, Hamid Palangi, Rujun Han 외

Reverse thinking plays a crucial role in human reasoning. Humans can reason not only from a problem to a solution but also in reverse, i.e., start from the solution and reason towards the problem. This often enhances ove…

Data AugmentationKnowledge DistillationLogical ReasoningMath+1

Backpropagation Clipping for Deep Learning with Differential Privacy

2022-02-10 · Timothy Stevens, Ivoline C. Ngong, David Darais, Calvin Hirsch 외

We present backpropagation clipping, a novel variant of differentially private stochastic gradient descent (DP-SGD) for privacy-preserving deep learning. Our approach clips each trainable layer's inputs (during the forwa…

Deep LearningPrivacy PreservingPrivacy Preserving Deep LearningSensitivity