paper-with-me

Papers

Efficient Vision-Language Pre-training by Cluster Masking

2024-05-14 · CVPR 2024 1 · Zihao Wei, Zixuan Pan, Andrew Owens

We propose a simple strategy for masking image patches during visual-language contrastive learning that improves the quality of the learned representations and the training speed. During each iteration of training, we randomly mask clusters of visually similar image patches, as measured by their raw pixel intensities. This provides an extra learning signal, beyond the contrastive training itself, since it forces a model to predict words for masked visual structures solely from context. It also speeds up training by reducing the amount of data used in each image. We evaluate the effectiveness of our model by pre-training on a number of benchmarks, finding that it outperforms other masking strategies, such as FLIP, on the quality of the learned representation.

📄 PDF Abstract BibTeX arXiv:2405.08815

Code (1)

zi-hao-wei/efficient-vision-language-pre-training-by-cluster-masking 공식 구현 pytorch

Tasks

Contrastive Learning

Methods 이 논문이 사용한 방법론

Contrastive Learning 설명 없음
FLIP https://developer.nvidia.com/blog/flip-a-difference-evaluator-for-alternating-images/

Similar Papers 제목 키워드 기반

Cluster-Wise Spatio-Temporal Masking for Efficient Video-Language Pretraining

2026-03-24 · Weijun Zhuang, Yuqing Huang, Weikang Meng, Xin Li 외 arxiv

Large-scale video-language pretraining enables strong generalization across multimodal tasks but often incurs prohibitive computational costs. Although recent advances in masked visual modeling help mitigate this issue, …

Video Question AnsweringVideo-Text RetrievalVideo Captioning

BackdoorIDS: Zero-shot Backdoor Detection for Pretrained Vision Encoder

2026-03-12 · Siquan Huang, Yijiang Li, Ningzhi Gao, Xingfu Yan 외 arxiv

Self-supervised and multimodal vision encoders learn strong visual representations that are widely adopted in downstream vision tasks and large vision-language models (LVLMs). However, downstream users often rely on thir…

Uniform Masking Prevails in Vision-Language Pretraining

2022-12-10 · Siddharth Verma, Yuchen Lu, Rui Hou, Hanchao Yu 외

Masked Language Modeling (MLM) has proven to be an essential component of Vision-Language (VL) pretraining. To implement MLM, the researcher must make two design choices: the masking strategy, which determines which toke…

Image-text matchingLanguage ModelingLanguage ModellingMasked Language Modeling+1

Frequency Is What You Need: Word-frequency Masking Benefits Vision-Language Model Pre-training

2024-12-20 · Mingliang Liang, Martha Larson

Vision Language Models (VLMs) can be trained more efficiently if training sets can be reduced in size. Recent work has shown the benefits of masking text during VLM training using a variety of approaches: truncation, ran…

Language ModelingLanguage ModellingPOS

Centered Masking for Language-Image Pre-Training

2024-03-23 · Mingliang Liang, Martha Larson

We introduce Gaussian masking for Language-Image Pre-Training (GLIP) a novel, straightforward, and effective technique for masking image patches during pre-training of a vision-language model. GLIP builds on Fast Languag…

Language ModelingLanguage Modelling