🔥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

Once again I am recommending that you NOT optimize your intelligent agents around usage metrics. Please. Pretty please. Just stop. I know that's how you optimize other products like recommendation engines but these things are NOT THE SAME.

Neel Nanda I think we should scale up TMLR so that it can be the central publication venue for ML. 4/

🚨 Preprint Alert 🚀 📄 seq-JEPA: Autoregressive Predictive Learning of Invariant-Equivariant World Models arxiv.org/abs/2505.03176 Can we simultaneously learn both transformation-invariant and transformation-equivariant representations with self-supervised learning (SSL)?

Traj level objectives (RLOO, GRPO) seem better than value function based methods for LLM training (Q-learning, PPO). Quite unfortunate. Traj level RL can't do skill stitching, this is extremely sample inefficient for learning compositional reasoning (such as multi turn tool use).

Is there a universal strategy to turn any generative model—GANs, VAEs, diffusion models, or flows—into a conditional sampler, or finetuned to optimize a reward function? Yes! Outsourced Diffusion Sampling (ODS) accepted to ICML Conference , does exactly that!

Is AdamW the best inner optimizer for DiLoCo? Does the inner optimizer affect the compressibility of the DiLoCo delta? Excited to introduce MuLoCo: Muon is a practical inner optimizer for DiLoCo! 🧵arxiv.org/abs/2505.23725 1/N

Super stoked to share my first first-author paper that introduces a hybrid architecture approach for real-time neural decoding. It's been a lot of work, but happy to showcase some very cool results!

(1/n)🚨You can train a model solving DFT for any geometry almost without training data!🚨 Introducing Self-Refining Training for Amortized Density Functional Theory — a variational framework for learning a DFT solver that predicts the ground-state solutions for different

🚀 Our method, Poutine, was the best-performing entry in the 2025 Waymo Vision-based End-to-End Driving Challenge at #CVPR2025! Our 3 B-parameter VLM Poutine scored 7.99 RFS on the official test set—comfortably ahead of every other entry (see figure).