Kai Ruan

@kai-ruan · 2 works

Recent posts by Kai Ruan

• SPO++: Stream-Aligned Policy Optimization for Asynchronous Agentic RL - Group-relative reinforcement learning waits for sibling rollouts of the same prompt, which is costly for long and variable tool-use trajectories. Single-stream Policy Optimization (SPO) removes this… · Paper · arxiv.org
• When Does Muon Help Agentic Reinforcement Learning? - Muon is competitive with AdamW in large-scale pre-training, but its value for reinforcement-learning (RL) post-training remains unclear. We study vanilla Muon in sparse-reward agentic RL through… · Paper · arxiv.org

Topics Kai Ruan posts about

• Qwen on Instagram - 1 post · Best Qwen creators