German Barquero

@joanrod_ai @QuiverAI @a16z Very happy to see what you and the team have achieved! Exciting journey ahead 💪🏼💪🏼

Introducing @QuiverAI, a new AI lab and product company focused on frontier vector design. We’ve raised an $8.3M seed round led by @a16z, with support from amazing angels and investors. Our first model, Arrow-1.0, generates SVGs from images and text. It’s available now in public beta at app.quiver.ai

UB Hupba Group attending BSC. They lead AI research powered by supercomputing capabilities, aiming to keep us at the frontier of new AI advances worldwide.

Join us! A unique opportunity to do groundbreaking research with huge impact in one of the most beautiful countries in the world!! 🇨🇭⛰️🚠

We have unlocked parallel training of non-linear RNNs! > LSTM entered the chat 🔥

Excellent opportunity to work with an amazing and talented team on a super cool and exciting project!!

Excited to spend the week at #CVPR2025 in Nashville! 🇺🇸 If you’re interested in learning more about MixerMDM, come check out our poster at the following sessions: 📍 💃 [@humogen11384] Wednesday 11, 16:00-18:00, ExHall D #430 📹 [AI4CC] Thursday 12, 17:00, ExHall D #412 👑 Saturday 14, 10:30-12:30, ExHall D #165 Feel free to reach out if you want to chat, connect, or just say hi! ☺️

@dancasas @Meta @AlbertPumarola Gràcies, Dan!!

@Germs96 @Meta @AlbertPumarola Congrats German!

Thrilled to share that I’ve just joined the GenAI team at @Meta in Zurich as an AI Research Scientist! Super excited to work alongside an incredible team led by @AlbertPumarola, pushing the frontiers of generative models for multimedia 🚀

@SergioEscalera_ @Meta @AlbertPumarola Thank you, Sergio!! Extremely grateful for your guidance all these years!

@Germs96 @Meta @AlbertPumarola Congrats german. You'll do great things, no doubt. best of luck in your new adventure!

@prlz77 @Meta @AlbertPumarola Gràcies, Pau!

@Germs96 @Meta @AlbertPumarola Congrats German!

@HamzaKebiri @Meta @AlbertPumarola Thanks Hamza :)

@Germs96 @Meta @AlbertPumarola Congrats German 👏🏼

@PDillis @Meta @AlbertPumarola For sure, Diego!! Thanks :)

@Germs96 @Meta @AlbertPumarola Congrats German! We'll make sure to visit in Zurich :)

@johnnync13 @AlbertPumarola @Meta Thanks Johnny!!

@AlbertPumarola @Germs96 @Meta Proud of you German!! I got my first master's thesis right. Rough diamond💎

@AlbertPumarola @Meta Gràcies, Albert!! Let’s goo!💪🏼💪🏼

@Germs96 @Meta Thrilled to have you on the team—let’s crank it up and make great things happen!

Step into the circle. Get Scanned. Get Paid. Get your 3D model. Join our virtual clothing research project at UB! linktr.ee/vitonify

We have new open positions: Looking for strong candidates with experience on diffusion / flow matching models to work on the next generation image and video models in GenAI. Apply Now: 👇 linkedin.com/posts/meghadas…

Call for participation - Chalearn Looking at People series - 6th Face Anti-Spoofing Workshop and Challenge@ICCV2025 sites.google.com/view/face-anti…

Scanning sessions are ongoing! 3D scan + 30€. Now you can repeat by bringing different outfits! Also, +5€ for every friend you refer. Slots open until June! 👉 linktr.ee/vitonify

Why do we even need to learn how to mix motions? 🤔In our previous work (DualMDM), we found something very interesting 💡: when mixing motions throughout the denoising process, it actually works better to mix them differently at different steps. Our initial hypothesis was that early in the denoising process, interpersonal dynamics (how people relate to each other) matter more 👫. Later on the denoising, interpersonal dynamics (how a person moves individually) become more important 💃. So, we proposed different mixing schedulers that adapt across steps 🧠📈. This approach worked better than others, but it still had limitations. While some schedulers performed better on average, none of them worked perfectly across all motion combinations 🎭. Different motion pairs needed slightly different schedulers to perform well. 💥 That’s what motivated MixerMDM! We introduce the Mixer — a lightweight transformer module that predicts a mixing weight at every step of the denoising process, conditioned on the motions to be mixed and their textual descriptions 🌀🤖. This allows us to dynamically and consistently mix motions while maintaining strong individual controllability 💃🕺✨