🔥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

Michael Noukhovitch This is a simple gist demonstrating how async RL for LLM works: gist.github.com/vwxyzjn/473b8b…

🚨 New paper alert! [NeurIPS 2024 spotlight] 🚨 Trajectory Flow Matching with Applications to Clinical Time Series Modeling ⏳📈 With: Yuan Pu , Yuki Kawamura 川村祐貴 , Andrew Loza, Yoshua Bengio, Dennis Shung, Alex Tong 💻: github.com/nZhangx/Trajec… 📄: arxiv.org/abs/2410.21154 🧵👇

Big news for the little guy (me and my 48Gb gpus)

🚀 Happy to share Tülu 3! We trained the model with actual RL: the model only receives rewards if its generations are verified to be correct (e.g., correct math solution). ❤️ Check out our beautiful RL curves. Code is also available: ~single file PPO that scales to 70B.

Models like DeepSeek-R1 🐋 mark a fundamental shift in how LLMs approach complex problems. In our preprint on R1 Thoughtology, we study R1’s reasoning chains across a variety of tasks; investigating its capabilities, limitations, and behaviour. 🔗: mcgill-nlp.github.io/thoughtology/

The code and model weights for this paper are finally open! Despite being a little late for releasing them, I hope you will find them useful! Code: github.com/facebookresear… Models: - (ViT-G): huggingface.co/lavoies/llip-v… - (ViT-B): huggingface.co/lavoies/llip-v…

Open sourcing code and models from industry is the invisible labour that actually creates progress in the field. It's a massive effort and requires authors to overcomes so many logistical and administrative hurdles. This stuff deserves more recognition!

🧵 Everyone is chasing new diffusion models—but what about the representations they model from? We introduce Discrete Latent Codes (DLCs): - Discrete representation for diffusion models - Uncond. gen. SOTA FID (1.59 on ImageNet) - Compositional generation - Integrates with LLM 🧱

LLMs can speak in DLC! We fine-tune a language model to sample DLC tokens from text, giving us a pipeline: Text → DLC → Image This also enables generation beyond ImageNet.

We're releasing a cool paper! DLCs are image tokens that enable better diffusion modelling. For now, we show this is the right representation. But in the future, this can allow LLMs to "speak in images"🤯to enable visual reasoning and more powerful text-image generalization. ⬇️