ViT
Technology & Computing abbreviation · Artificial Intelligence
| Stands for | Vision Transformer |
|---|---|
| Categories | Technology & Computing |
ViT stands for Vision Transformer, an artificial intelligence abbreviation.
An image recognition model that splits a picture into small patches and processes them as a sequence with a transformer, the architecture originally built for language, instead of convolutional layers.
Example: The ViT backbone outperformed our ResNet once we pretrained it on the larger dataset.
Usage notes
Introduced by Google researchers in the 2020 paper "An Image Is Worth 16x16 Words". ViTs need large training datasets to beat convolutional networks but scale very well. They serve as the image encoders in many vision-language models.
Related abbreviations
More artificial intelligence abbreviations
- Adam – Adaptive moment estimation
- AGI – Artificial general intelligence
- AI – Artificial intelligence
- AIGC – AI-generated content
- ANI – Artificial narrow intelligence
- ANN – Approximate nearest neighbor (search)
- ASI – Artificial superintelligence
- ASL – AI Safety Level
- ASR – Automatic speech recognition
- AUC – Area under the curve
- AutoML – Automated machine learning
- BERT – Bidirectional Encoder Representations from Transformers