Jacob Berg

He’s back @arhanjain.bsky.social

Andre Ye – Paul & Daisy Soros Fellowships for New Americans

Inspired by his father’s study of neural networks, #UWAllen CS and Philosophy major @andreiskiii.bsky.social taught himself the subject on his way to enrolling in UW at age 15. He just won a @pdsoros.bsky.social Fellowship to pursue a Ph.D. at MIT in #AI. pdsoros.org/fellows/andr... #HuskyExperience

College students get a taste of startup life inside dynamic UW entrepreneurship class Team "Stimma" pitches at Pioneer Square Labs on Wednesday in front of Seattle venture capitalists. (GeekWire Photo / Taylor Soper) Best class ever? That

#UWAllen & PSL - www.geekwire.com/2025/college...

Interested in retrieving (sub-)trajectories from enormous datasets like DROID? 🦣

Check out our retrieval code using DTW and vision foundation model features!

Code: github.com/WEIRDLabUW/S...
Website: weirdlabuw.github.io/strap

GitHub - deepseek-ai/DeepSeek-R1 Contribute to deepseek-ai/DeepSeek-R1 development by creating an account on GitHub.

The deepseek r1 paper and download links: github.com/deepseek-ai/...

Pilluuuuuuuuuu

Have some offline data lying around? Use it to robustify few-shot imitation learning! 🤖

STRAP 🎒 is a retrieval-based method that leverages semantic sub-trajectories in offline datasets to augment the training data.

🧵 1/6

Overview of PixMo and its relation to Molmo's ability. PixMo's captions data enables Molmo's fine-grained understanding; PixMo's AskModelAnything enables Molmo's user interaction; PixMo's pointing data enables Molmo's pointing and counting; PixMo's synthetic data enables Molmo's visual skills.

Remember Molmo? The full recipe is finally out!

Training code, data, and everything you need to reproduce our models. Oh, and we have updated our tech report too!

Links in thread 👇

the science of LMs should be fully open✨

today @akshitab.bsky.social @natolambert.bsky.social and I are giving our #neurips2024 tutorial on language model development.

everything from data, training, adaptation. published or not, no secrets 🫡

tues, 12/10, 9:30am PT ☕️

neurips.cc/virtual/2024...

Can’t forget about Vienna

@arhanjain.bsky.social plz come home and play with me

@arhanjain.bsky.social’s dog at it again

We’ve been investigating how sim, while wrong, can be useful for real-world robotic RL! In our #NeurIPS2024 work, we theoretically showed how naive sim2real transfer can be inefficient, but if you *learn to explore* in sim, this transfers to the real world! We show this works on real robots! 🧵(1/6)

I'm excited about scaling up robot learning! We’ve been scaling up data gen with RL in realistic sims generated from crowdsourced videos. Enables data collection far more cheaply than real world teleop. Importantly, data becomes *cheaper* with more environments and transfers to real robots! 🧵 (1/N)

Just wanted to say @arhanjain.bsky.social has a very cute dog