iiinova — where your digital self keeps evolving

Olivier Hénaff

@olivierhenaff.bsky.social

Working on something new, combining active, multimodal, and memory-augmented learning. Formerly Senior Staff Scientist @GoogleDeepMind, PhD @NYU, @Polytechnique

Posts Replies Media

Olivier Hénaff @olivierhenaff.bsky.social · 14/02/2025

After an amazing 6 years at Google DeepMind, I'm thrilled to announce that I'll be starting a new project at the intersection of multimodal foundation modeling, data curation, and human behavior. If this is of interest to you please reach out!

110

Olivier Hénaff @olivierhenaff.bsky.social · 02/12/2024

Active data curation keeps on giving. This time we enabled the distillation of large multimodal models into much smaller ones, simply by choosing the data they learn from. Sets a new state of the art in small multimodal models that are very efficient for inference!

050

Reposted by Olivier Hénaff

Karsten Roth @confusezius.bsky.social · 28/11/2024

This was an insightful project I worked on at Google DeepMind alongside the amazing @zeynepakata.bsky.social , @dimadamen.bsky.social , @ibalazevic.bsky.social and @olivierhenaff.bsky.social: 👉Language-image pretraining with CLIP or SigLIP is widely used due to strong zero-shot transfer, but ....

1121

Reposted by Olivier Hénaff

Ivana Balazevic @ibalazevic.bsky.social · 28/11/2024

We maintain strong zero-shot transfer of CLIP / SigLIP across model size and data scale, while achieving up to 4x few-shot sample efficiency and up to +16% performance gains! Fun project with @confusezius.bsky.social, @zeynepakata.bsky.social, @dimadamen.bsky.social and @olivierhenaff.bsky.social.

0203

Olivier Hénaff @olivierhenaff.bsky.social · 28/11/2024

More than zero-shot generalization, few-shot *adaptation* is critical for many applications. We find simple changes to multimodal pretraining are sufficient to yield outsized gains on a wide range of few-shot tasks. Congratulations @confusezius.bsky.social on a very successful internship!

071