RASMUS's picture
Training in progress, step 1000
878dbce
|
raw
history blame
928 Bytes

Fiber: Coarse-to-Fine Vision-Language Pre-Training with Fusion in the Backbone

Session by johko

Recording πŸ“Ί

YouTube

Session Slides πŸ–₯️

Google Drive

Original Paper πŸ“„

Hugging Face / arxiv

GitHub Repo πŸ§‘πŸ½β€πŸ’»

https://github.com/microsoft/fiber

Additional Resources πŸ“š