Analyzing the Hierarchical Reasoning Model by Guan Wang We verified scores on hidden tasks, ran ablations, and found that performance comes from an unexpected source ARC-AGI Semi Private Scores: * ARC-AGI-1: 32% * ARC-AGI-2: 2% Our 4 findings:

basically this - hunyuan-gamecraft.github.io

"𝗪𝗵𝘆 𝗰𝗮𝗻'𝘁 𝘄𝗲 𝗷𝘂𝘀𝘁 𝘁𝗮𝗸𝗲 𝗚𝗣𝗧-𝟰, 𝗽𝗹𝘂𝗴 𝗶𝘁 𝗶𝗻𝘁𝗼 𝗮 𝗿𝗼𝗯𝗼𝘁, 𝗮𝗻𝗱 𝗰𝗮𝗹𝗹 𝗶𝘁 𝗮 𝗱𝗮𝘆?" Because foundation models for robotics face fundamentally different challenges than language models. Let me explain the technical reality:

Our paper 'Self‑Labelling via Simultaneous Clustering and Representation Learning' just got its 1000th citation. On that occasion, I want to give my perspective on this question: Who or what is Sinkhorn–Knopp? Short answer: It’s the little ~1960s matrix‑normalization workhorse

Respect.

✈️ On my way to Hawaii for #ICCV2025! Join us at our workshop 🌍 Reliable and Interactive World Models (#RIWM2025) 🌍 📍 Room 316A 🗓️ Mon, Oct 20 | 11 AM – 8 PM PDT riwm-2025.github.io/RIWM-2025/ #WorldModel #EmbodiedAI #InteractiveAI

heading to #ICCV2025, anyone up for a ☕️? also, my team at FAIR has an internship opening on world modeling, planning, and their robotics applications. DM me if you’re interested.

wtf! Yuandong is one of the GOATs of not just FAIR, but Meta's entire AI push. Absolute madness.