Andrej Risteski @ ICML ✈️
@risteski_a
Machine learning researcher. Associate Professor, ML department at CMU (@mldcmu).
Giving a talk tomorrow at the SFU at ICML Workshop, at 9:00am local time! (link below) Title: "Why AI is Harder in the Physical World 🤖 than the discrete world 📑 .... and what to maybe do about it"
I'll be giving the first H-Net talk this afternoon at 4:30-5 PT at the ES-FoMo workshop! come support the fight against Big Token 🙏
Looking forward to seeing everyone for ES-FoMo part three tomorrow! We'll be in East Exhibition Hall A (the big one), and we've got an exciting schedule of invited talks, orals, and posters planned for you tomorrow. Let's meet some of our great speakers! 1/
If you are at #icml25 and are interested in RL algorithms, scaling laws for RL, and test-time scaling (& related stuff), come talk to us at various poster sessions (details ⬇️). We are also presenting some things at workshops later in the week, more on that later.
Will be at #icml2025 next week, and looking forward to catching up with folks. Reach out if you'd like to meet up!
Tokenization is just a special case of "chunking" - building low-level data into high-level abstractions - which is in turn fundamental to intelligence. Our new architecture, which enables hierarchical *dynamic chunking*, is not only tokenizer-free, but simply scales better.
Tokenization has been the final barrier to truly end-to-end language models. We developed the H-Net: a hierarchical network that replaces tokenization with a dynamic chunking process directly inside the model, automatically discovering and operating over meaningful units of data
🧵generative models are sweet, but navigating existing repositories can be overwhelming, particularly when starting a new research project so i built jax-interpolants, a clean & flexible implementation of the stochastic interpolant framework in jax github.com/nmboffi/jax-in…
11/10 BWLer was just presented at the Theory of AI for Scientific Computing (TASC) workshop at COLT 2025, where it received Best Paper 🏆 Huge thanks to the organizers (@nmboffi, @khodakmoments, Jianfeng Lu, @__tm__157, @risteski_a) for a fantastic event!
If you're coming to COLT --- reminder, this is happening first thing Monday morning (June 30) :-) Schedule for the workshop here: tasc-workshop.github.io/#schedule
@khodakmoments, @__tm__157, along with myself, @nmboffi and Jianfeng Lu are organizing a COLT 2025 workshop on the Theory of AI for Scientific Computing, to be held on the first day of the conference (June 30).