Pinned post
Can a language model be based purely on self-conditioning? 🤔
Yes! Meet LAW 👋 (a.k.a. Loopholing All the Way!)
🧵1/8
Yes! Meet LAW 👋 (a.k.a. Loopholing All the Way!)
🧵1/8
0:36
@nadjushenka_
Pinned post
(and thanks @nadjushenka_ for the super nice visuals 😻)View quoted post on X
🔥 New paper: Language Modeling with Hyperspherical FlowsView quoted post on X
Recent flow language models (FLMs) all use Gaussian noise. Makes sense for images, but not necessarily for text 🫠 We propose to add noise by rotating embeddings on 𝕊^{d−1} instead 🌐
w/ @caglarml
(1/9)
Still at #ICLR2026 for the workshops and interested in discrete diffusion? Check-out the RealmGen workshop ☀️ I’ll be giving a spotlight talk at ~11:18. See you there 😄View quoted post on X