Browsing: Reinforcement

Most fine-tuning teaches a model to produce a better answer to a single prompt. Agentic tasks are different. An agent that files an expense report or triages a support ticket takes a sequence of actions, and each step depends on what came before. There’s rarely one correct sequence to imitate, and the only reliable feedback…

Having machines learning by themselves is the ultimate Science Fiction dream. Today that gap is closing, with programs like AlphaGo and the different LLMs that are out there, but would you believe it if I told you researchers started formulating these kinds of problems and developing algorithms to tackle this in the late 1950s?

When you post-train a Mixture-of-Experts (MoE) model with Reinforcement Learning from Human Feedback (RLHF) or Group Relative Policy Optimization (GRPO) at scale, three simultaneous challenges emerge. The first requires coordinating heterogeneous compute for rollout generation and policy training. Second, sustaining high-throughput communication across hundreds of accelerators. And third, dynamically orchestrating every subsystem to keep them…

Reinforcement learning (RL) is the central training paradigm for advancing large foundation models towards self-improvement. This report introduces the MiMo-V2.6 series, an omni-modal family that pushes the frontier of model intelligence by scaling RL compute. Prior to RL, we conduct mid-training on a broad multimodal corpus to provide ample exploration space, and build a solid…