[2106.03348] ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2106.03348] ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Computer Vision and Pattern Recognition arXiv:2106.03348 (cs) [Submitted on 7 Jun 2021 ( v1 ), last revised 24 Dec 2021 (this version, v4)] Title: ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Bias Authors: Yufei Xu , Qiming Zhang , Jing Zhang , Dacheng Tao View a PDF of the paper titled ViTAE: Vision Transformer Advanced by Exp

related reading

• [2302.05442] Scaling Vision Transformers to 22 Billion Parametersarxiv.org
• A Mathematical Framework for Transformer Circuitstransformer-circuits.pub
• In-context Learning and Induction Headstransformer-circuits.pub
• [2605.05331] ViTok-v2: Scaling Native Resolution Auto-Encoders to 5 Billion Parametersarxiv.org
• Stand-Alone Self-Attention in Vision Models - NeurIPS-2019-stand-alone-self-attention-in-vision-models-Paper.pdfpapers.nips.cc
• Aman's AI Journal • Primers • Ilya Sutskever's Top 30aman.ai
• [2201.03545] A ConvNet for the 2020sarxiv.org
• [2201.06618] VAQF: Fully Automatic Software-Hardware Co-Design Framework for Low-Bit Vision Transformerarxiv.org
• The Transformer Family | Hacker Newsnews.ycombinator.com
• How I built my own Tesla-style self-driving AI | Transformer Lablab.cloud
• Towards Multimodal Interpretability: Learning Sparse Interpretable Features in Vision Transformers — LessWronglesswrong.com
• On the speed of ViTs and CNNslucasb.eyer.be