~400 people have joined us for the Research Mentorship Session by @sarahookr at @Cohere_Labs Machine Learning Summer School. So much great advice for early career and aspiring researchers 🔥🔥❤️

Minimal example (ViT, inet10) now available with comments and details at https://t.co/s4yNTPg42B - ~150 lines *total* - 91% top1 (simple linear probe) in a couple hours (1gpu) - no teacher-student/stop-gradient/SWA/... - losses that are as smooth as you can hope for:

Hey Nano Banana Pro, please annotate the original Transformer architecture diagram. Just look at how precisely it added little insights to the main operations. 🤯 Great for infographics and for improving technical visual communication.

Excited to announce our MIT Press book “Neuroevolution: Harnessing Creativity in AI Agent Design” by Sebastian Risi (@risi1979), Yujin Tang (@yujin_tang), Risto Miikkulainen, and myself. We explore decades of work on evolving intelligent agents and shows how neuroevolution can

Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve

Introducing Gemini 3 ✨ It’s the best model in the world for multimodal understanding, and our most powerful agentic + vibe coding model yet. Gemini 3 can bring any idea to life, quickly grasping context and intent so you can get what you need with less prompting.  Find Gemini

Claude finally has some self-respect, previous sonnet version would always rate CGPT's answer higher even when Claude’s answer was clearly better.

i've reviewed my own PR and found nothing wrong with it

Sharing an interesting recent conversation on AI's impact on the economy. AI has been compared to various historical precedents: electricity, industrial revolution, etc., I think the strongest analogy is that of AI as a new computing paradigm (Software 2.0) because both are

https://t.co/cUOIHEbfvf

Python 3.7 feels like yesterday.

Nvidia presents TiDAR Think in Diffusion, Talk in Autoregression

LeJEPA: a novel pretraining paradigm free of the (many) heuristics we relied on (stop-grad, teacher, ...) - 60+ arch., up to 2B params - 10+ datasets - in-domain training (>DINOv3) - corr(train loss, test perf)=95% Paper: https://t.co/NpfB9G1pOP Code: https://t.co/BsK5wmNEHc

Regular reminder to learn einsum & einops if you haven't already!

starting to learn diffusion models, trying a nano implementation, will keep posting what i learn along the way

really enjoying reading the trio tutorial https://t.co/sclrte6eaI

Prototyping is easy; productizing is difficult

Here's your weekend challenge: Implement speculative decoding. Step 1: Read the following paper and/or blog: https://t.co/yJ7Rkb7yv9 https://t.co/8A4LWmruxM (cc @jaygala223) Step 2: Choose a family of models which come in various sizes. My choice would be the Gemma3 or Qwen

Introducing Cambrian-S it’s a position, a dataset, a benchmark, and a model but above all, it represents our first steps toward exploring spatial supersensing in video. 🧶

Junior students who have just started doing research? Check out the (75 and counting) awesome tips! https://t.co/5CTTuJm3Jg

people in drought celebrate a drop of water