🛠️ Steven Gong

Search

SearchSearch

Jun 17, 2025, 1 min read

AdaLN

Used in the pi0 paper.

AdaLN, short for Adaptive Layer Normalization, is a Vision Transformer (ViT) architecture designed for multidomain learning, particularly in the context of extracting building information from satellite and street view images

https://web.eecs.umich.edu/~stellayu/publication/doc/2022AdaLN.pdf

Graph View

Backlinks

  • No backlinks found

Created with Quartz, © 2025

  • Blog
  • LinkedIn
  • Twitter
  • GitHub