LightVLP: A Lightweight Vision-Language Pre-training via Gated Interactive Masked AutoEncoders.
Xingwu SunZhen YangRuobing XieFengzong LianZhanhui KangChengzhong XuPublished in: LREC/COLING (2024)
Keyphrases
- lightweight
- denoising
- natural language
- virtual reality
- vision system
- feedforward neural networks
- programming language
- computer vision
- image processing
- training set
- dos attacks
- supervised learning
- wireless sensor networks
- real time
- smart camera
- restricted boltzmann machine
- development environments
- language learning
- machine translation
- neural network