T2T-ViT

GPTKB entity

Statements (28)
Predicate Object
gptkbp:instanceOf Vision Transformer model
gptkbp:architecture gptkb:transformation
gptkbp:arXivID 2101.11986
gptkbp:author gptkb:Tao_Wang
gptkb:Thomas_Huang
gptkb:Li_Yuan
Jiehao Wang
Qibin Hou
Yujun Shi
Yunpeng Chen
Zhangyang Wang
gptkbp:citation arXiv preprint
Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
gptkbp:contribution better data efficiency
improved performance on ImageNet
progressive tokenization for vision transformers
reduced model size
gptkbp:designedFor image classification
gptkbp:developedBy gptkb:Google_Research
gptkbp:fullName Tokens-to-Token Vision Transformer
https://www.w3.org/2000/01/rdf-schema#label T2T-ViT
gptkbp:improves gptkb:ViT
gptkbp:introducedIn 2021
gptkbp:publishedIn gptkb:arXiv
gptkbp:repository https://github.com/yuanli2333/T2T-ViT
gptkbp:uses Tokens-to-Token module
gptkbp:bfsParent gptkb:Vision_Transformer
gptkbp:bfsLayer 6