Profile avatar
kolesnikov.ch
Member of technical staff at OpenAI (Zurich).
20 posts 713 followers 71 following
Getting Started
Conversation Starter
Posts 7 Comments 15

Looking for a small or medium sized VLM? PaliGemma 2 spans more than 150x of compute! Not sure yet if you want to invest the time 🪄finetuning🪄 on your data? Give it a try with our ready-to-use "mix" checkpoints: 🤗 huggingface.co/blog/paligem... 🎤 developers.googleblog.com/en/introduci...

submitted 8 days ago • 0 comments

With some delay, JetFormer's *prequel* paper is finally out on arXiv: a radically simple ViT-based normalizing flow (NF) model that achieves SOTA results in its class. Jet is one of the key components of JetFormer, deserving a standalone report. Let's unpack: 🧵⬇️

submitted 69 days ago • 2 comments

Paligemma2 is out! Bigger models, better results. For the best experience, do not forget to finetune. Congrats Paligemma2 team!

submitted 84 days ago • 0 comments

Ok, it is yesterdays news already, but good night sleep is important. After 7 amazing years at Google Brain/DM, I am joining OpenAI. Together with @xzhai.bsky.social and @giffmana.ai, we will establish OpenAI Zurich office. Proud of our past work and looking forward to the future.

submitted 85 days ago • 9 comments

In arxiv.org/abs/2303.00848, @dpkingma.bsky.social and @ruiqigao.bsky.social had suggested that noise augmentation could be used to make other likelihood-based models optimise perceptually weighted losses, like diffusion models do. So cool to see this working well in practice!

submitted 87 days ago • 0 comments

The answer has just dropped: bsky.app/profile/kole...

submitted 87 days ago • 2 comments

I always dreamed of a model that simultaneously 1. optimizes NLL of raw pixel data, 2. generates competitive high-res. natural images, 3. is practical. But it seemed too good to be true. Until today! Our new JetFormer model (arxiv.org/abs/2411.19722) ticks on all of these. 🧵

submitted 87 days ago • 3 comments