Cross Attention

Concept

A method to inject conditions where image patch embeddings (queries) attend to text embeddings (keys and values) to determine relevance for image changes.

Mentioned in 1 video

Videos Mentioning Cross Attention

Stanford CME296 Diffusion & Large Vision Models | Spring 2026 | Lecture 5 - Architectures

Stanford CME296 Diffusion & Large Vision Models | Spring 2026 | Lecture 5 - Architectures

Stanford Online

A method to inject conditions where image patch embeddings (queries) attend to text embeddings (keys and values) to determine relevance for image changes.