New research from Databricks: LLMs Can Learn to Reason... via Off-Policy RL Optimal Advantage-based Policy Optimization with Lagged Inference policy (OAPL) shows you don’t need strict on-policy training to improve reasoning. It matches or beats Group Relative Policy Optimization (GRPO), stays stable with large policy lag, and uses ~3× fewer training generations. For Databricks customers, it’s a simpler, practical, and equally powerful approach to RL that Databricks is pioneering internally — and bringing directly to Databricks customers, so enterprises can improve agents using the same methods we use for our in-house agents, without complex infrastructure changes.show more

Announcing native availability of Anthropic Claude 3.7 Sonnet in... Databricks across AWS, Azure and GCP! Customers can now securely access Claude’s advanced reasoning, planning and agentic capabilities directly within Databricks, no setup required. This launch marks the beginning of a strategic partnership between Anthropic and Databricks. We’re excited to see what our customers build and how they leverage domain-specific AI agents!show more

Does off-policy value-based RL scale? In LLMs, larger scale... predictably improves performance. Value-based RL learns from arbitrary data and is sample-efficient, but folk wisdom says it doesn't scale 🧵⬇️We show predictability for scaling value-based RL!show more

🤔 How to fine-tune an Imitation Learning policy (e.g.,... Diffusion Policy, ACT) with RL? As an RL practitioner, I’ve been struggling with this problem for a while. Here’s why it’s tough: 1️⃣ Special designs (usually for multimodal action distributions) in modern IL models make them non-trivial to fine-tune by RL. 2️⃣ Large policy models + RL's poor sample efficiency = a nightmare But finally, we figured out a simple solution that works for any model architecture! 🌟 Check out our #ICLR2025 paper: “Policy Decorator: Model-Agnostic Online Refinement for Large Policy Models”, led by my amazing mentee Xiu Yuan. 🔗 🧵 Read more below!show more

Meet #DBRX: a general-purpose LLM that sets a new... standard for efficient open source models. Use the DBRX model in your RAG apps or use the DBRX design to build your own custom LLMs and improve the quality of your GenAI applications.show more

Haven't been to a conference in a while, really... excited to be at #NeurIPS2024! I'll be helping present 4 of our group's recent papers: 1. Overcoming the Sim-to-Real Gap: Leveraging Simulation to Learn to Explore for Real-World RL 2. Distributional Successor Features Enable Zero-Shot Policy Optimization 3. Learning to Cooperate with Humans using Generative Agents 4. Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning Find more details on each paper and where to find us in this thread (1/6)show more

How does high-fidelity tactile simulation help robots nail the... last millimeter? We’re releasing VT-Refine, accepted to CoRL: a real-to-sim-to-real visuo-tactile policy using a GPU-parallel tactile sim for our piezoresistive skin FlexiTac. Then fine-tuning a diffusion policy with large-scale RL in simulation. Website: #CoRL2025 #RobotLearning #Sim2Realshow more

Zombie robot RL policy

New project! Flow Policy Gradients for Robot Control tldr;... a simple online RL recipe for training and fine-tuning flow policies for robots co-led w/ Hongsuk Benjamin Choi:show more

Netanyahu: My opposition to a Palestinian State is not... simply my policy, it is the policy of Israel and its people.show more

🚨Current scalable RL algos train a policy w/o value... func, which is limiting with learning in open-ended, non-stationary, dynamic environments. But, how to scale value-based RL with more data/compute is unclear... Not anymore: presenting scaling laws for value-based RL 🧵⬇️show more

Join our mission to advance human-centered AI! Stanford HAI... is seeking a new Director of Policy and Society to lead our team in producing critical scholarship and convening global discussions on the intersection of technology, policy, and society.show more

Here is a tutorial on training LLaSA (LLaMA-based TTS)... using GRPO to improve prosody, rhythm, and expressiveness in synthesized speech with TRL!show more

ICML 2026: Latent Reasoning in TRMs is Secretly a... Policy Improvement Operator Why does recursive reasoning, especially latent reasoning, actually work? The theory is still young, and even mechanistic explanations are limited. We close part of this gap by showing that latent reasoning is secretly doing policy improvement. Each recursion pushes the model steadily toward the target. Based on this view, we propose an algorithm that boosts learning and inference efficiency by up to 18x.show more

Generalist robot policies need a benchmark that works across... any robot and any policy. 🦾 Introducing RoboLab, a high‑fidelity simulation benchmark built on NVIDIA Isaac and Omniverse to evaluate generalist robot policies in diverse, photoreal, physics‑based environments. Coming soon to the NVIDIA Isaac Lab‑Arena roadmap for large‑scale, robotic policy evaluation. 📖 #NationalRoboticsWeekshow more

You asked. We listened. Introducing StockX Returns. With our... new policy, you can now return eligible orders up to 14 days after delivery for StockX Credit.show more

The Pope says you can’t vote for Trump because... he has a strict immigration policy …do you know who else has a strict immigration policy?show more

“But I wish you could explain to me what... the hell’s going on with the mind of the public. Because we have the right policy. [Democrats] don’t. They have horrible policy.” —Trump at the House GOP retreatshow more