BriefGPT.xyz
Oct, 2023
理解CLIP中的可迁移表征学习和零射击迁移
Understanding Transferable Representation Learning and Zero-shot Transfer in CLIP
HTML
PDF
Zixiang Chen, Yihe Deng, Yuanzhi Li, Quanquan Gu
TL;DR
通过对CLIP的理论研究，我们证明了多模态学习的可转移表示学习，并分析了其在零样本学习和下游任务中的性能。在此基础上，我们提出了一种新的CLIP类型方法，在基准数据集上实现了比CLIP和其他最先进方法更好的性能。
Abstract
multi-modal learning
has become increasingly popular due to its ability to leverage information from different data sources (e.g., text and images) to improve the model performance. Recently,
clip
has emerged as
→