Tavis Siebert (@tavis_siebert) profile picture

MS CS @ ETH Zurich

Joined June 2024
Tavis Siebert retweeted
Scared for the future of open source? Black Friday discounts got you disappointed? NO MORE! With stable-envs: - save 99% of disk - rendering speedups of up to 20x faster than baseline - train world models up to 30% faster end-to-end - on 80+ of the most popular envs - focus on training, modeling, and NOT data pipelines! - open source! (see thread)🧡 We'd love to see you use it! If your environment is not supported yet, please open a PR!
3
6
2
36
2,421
The Platonic Representation Hypothesis is now a Theorem ◼️ Train JEPAs separately, on any modalities: they provably learn the same representation up to one rotation. The result? A 109M JEPA-Text model that aligns with vision models better than 72B LLMs. w/ @ylecun 🧡
41
188
30
1,119
86,252
Tavis Siebert retweeted
Your world model may know less about the world than you think. I've always wondered what world models really learn. Humans don't memorize every trajectory. We learn rules we can reuse. Can world models do the same? We built AnyGoal to find out. hazel-heejeong-nam.github.io…
13
27
4
186
11,104
Tavis Siebert retweeted
Preparing for AI Research Scientist interviews? πŸ”₯ We just open-sourced PaperLeet πŸˆβ€β¬› LeetCode for AI researchers: implement the core of ~100 key papers (llm, rl, diffusion, flow matching, JEPA, world models), graded by tests that check the math. paperleet.com
25
116
12
1,027
108,321
Tavis Siebert retweeted
JEPAs can 𝐧𝐨𝐰 jointly imagine what happens and what to do next! Meet π‹πžπ–π€πŒ: A JEPA for the real world with 32.3Γ— faster planning, 19 ms per plan Improves success 28.6% β†’ 89.7% on contact-rich manipulations But also ... 🫢 a WAM for everyone with 17M Β· 1 GPU Β· 1 hyperparameter Works on a real robot, not just a simulation !! πŸ‘‰ le-wam.github.io/
27
84
12
731
69,128