Hy-AI

⚡️ As LLM reinforcement learning scales to larger GPU clusters and more training data, training effi...

Hunyuan(@TXhunyuan) · 8.5 分

腾讯在LLM强化学习中通过调整批处理大小和学习率，显著提升训练效率。PPO吞吐量提升2.29倍，GRPO验证时间减少29%。