DINOv2 has never seen a caption, and Qwen3 has never seen an image. We still aligned their embedding spaces without a single image-caption pair. It even works when the images and the captions come from different datasets. Project page: dominik-schnaus.github.io/unpaired-rosetta ⬇️

Show Post

buddy, if this bothers you: you definitely don't want to know how little we actually understand about the mechanism of action of many common medicines.

Show Post

New online! Rethinking ocean alkalinity enhancement

Show Post

Great to see this paper finally out. Patiently Led by Prof Zanna Chase as part of c-side PAGES working group on past changes and the role of sea ice in the earth system. 🌊🧊 cp.copernicus.org/articles/22/...

Show Post

The problem is the anti AI side don't use AI and aren't experienced enough to actually have a conversation about it. It's all what they heard in the media or what they remember from last year when they barely tried AI to do a google search. That's why it's hard to take them seriously

Show Post