Excited to share our latest work on Video Compression using Transformers (VCT), check it out! PS: Fabian, Sergi and I will be attending CVPR, looking forward to catch up with everyone 😊

It’s surprising that in 2022, there remains little movement away from LaTeX toward a new language. It has some of the most unintuitive designs and syntax you’d expect in a language today.

This is an elegant approach to use a diffusion model as a generative image compressor!

Dear #CVPR2026 -- for next year can we please increase the space allocated for name+affiliation? Now it is only 6% of the badge!

Apparently Twitter isn't confident that a tweet by the Chairman @ Twitter about the world's wealthiest person is about "Business personalities"? 😅

Looking forward to present our paper on Neural Video Compression with GANs at #ECCV2022 😊

This Icelandic company just reveived a 100M+€ EU grant to build a massive CO2 storage hub in my home town, Hafnarfjordur, Iceland. Super cool tech that basically turns CO2 into stone by pumping carbonated water into lava!

Will be presented at #NeurIPS! 🎉🎉🎉

A brief life update: 🇨🇭➡️🇮🇸 After 10 wonderful years in Switzerland, I have finally returned back to my home country, Iceland. Goodbye Google Zurich, and hello Google Iceland 🥶🌋

I'll present this work on Thursday in New Orleans at #NeurIPS2022 !! See u there ☀️👾

Have a look at our work on using diffusion models for SoTA generative image compression!

1/ Excited to share that our work on masked transformers for compression has been accepted at #ICCV2023! We show two things: a) how a simple transformer based architecture can lead to state-of-the-art image compression results: arxiv.org/abs/2304.07313

Finite Scalar Quantization: VQ-VAE Made Simple paper page: huggingface.co/papers/2309.15… propose to replace vector quantization (VQ) in the latent representation of VQ-VAEs with a simple scheme termed finite scalar quantization (FSQ), where we project the VAE representation down to a

Check out our latest work, showing that Finite Scalar Quantization greatly simplifies VQ based representation learning!

Just the usual Monday evening in Iceland, aside from a volcanic eruption seen from the backyard...

We build neural codecs from a *single* image or video, achieving compression performance close to SOTA models trained on large datasets, while requiring ~100x fewer FLOPs for decoding ⚡ #CVPR2024 c3-neural-compression.github.io

What a way to celebrate one year of incredible Gemini progress -- #1🥇across the board on overall ranking, as well as on hard prompts, coding, math, instruction following, and more, including with style control on. Thanks to the hard work of everyone in the Gemini team and

Check out our latest Gemini models!