🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

There is nothing like tennis at the highest level humanly possible.

Check out the blog post on our work on scaling up embedding dimension in Transformer models efficiently! (NeurIPS 2023 Spotlight paper) (Joint work with Cenk Baykal, Dylan Cutler, Nikhil Ghosh, Rina Panigrahy and Xin Wang)

Oh:

Check out our new multi-image reasoning benchmark where a model needs to reason using information spread across text and multiple images! (An interesting insight we discover: Even the mightiest models struggle to tell time!)

✨ I made a new site called 🧳 💨 LuggageLosers.com It's a live ranking of airlines by how much luggage they are losing right now So you can avoid flying with them (and hopefully they can improve) Airlines losing most luggage rn: 🇮🇳 Air India 🇮🇪 Aer Lingus 🇬🇧 British

Funny thing is that I am convinced intelligence is distilling system 1 into system 2.

Playing around with giving o1-preview a slight modification of a question from AIME 2024. Almost gets it right. One marked improvement over previous models is the reasoning chain stays consistent throughout the solution. chatgpt.com/share/66e4cf93…

Happy to share Michelangelo (arxiv.org/abs/2409.12640), a long-context reasoning benchmark which measures performance beyond needle tasks up to arbitrary context lengths and remains challenging for frontier models. Stay tuned for more Michelangelo evals to come!

Check out our new work showing that causal language modeling alone is sufficient for a Transformer model to learn to solve Sudokus and other constraint satisfaction problems like Zebra puzzles! Lead by Kulin Shah, to appear at NeurIPS Conference 2024!

Beautiful <3

[1] ReMI: A Dataset for Reasoning with Multiple Images Work done with Nishanth Dikkala, Ankit Anand, Fangyu Liu, Bahare Fatemi and several other great colleagues. Fri 13 Dec, 11 am - 2 pm PST, West Ballroom A-D #5104 Details in sub-tweet: x.com/kazemi_sm/stat…

After apparently the entirety of the Indian population, I've come to also learn that: *Y'all are wild with names *Y'all have Lenin's *Y'all have a dude called Napoleon Einstein *At some point I'll literally come to India just to investigate whatever the hell is going on

Found out students are using this website to have an AI generated Lebron James summarise their study material. You cannot make this stuff up. 3blue1bron.com

As LLMs get stronger, we need harder benchmarks to continue tracking progress. Check out our new work in this direction.

💡🚨Check out our ICLR 2025 paper on the inductive bias looping offers for reasoning tasks!

Checkout our latest Gemma models, you'll love them :) Also checkout the results on our BIG-Bench Extra Hard benchmark:

Presenting this work @ ICLR tomorrow! Come talk to us about looped transformers and their inductive bias for reasoning tasks. Poster #272: Hall 3 + 2B. iclr.cc/virtual/2025/p…

Check out the new Gemma 3n model we shipped yesterday which packs a lot of powwow for its size!

We’re fully releasing Gemma 3n, which brings powerful multimodal AI capabilities to edge devices. 🛠️ Here’s a snapshot of its innovations 🧵

Check out alternating updates, our research technique which is part of the Gemma 3n model!