讀紙 ReadPaper

由 Claude Code Max 翻譯的 AI 論文繁體中文版 · 101 篇

標籤 願望清單
篩選標籤： Reward Hacking Defense 1 篇 × 清除
hf-XiaomiMiMo-MiMo-V2.6-Flash-RL MIT License MIT

MiMo-V2.6：將強化學習擴展至自我改進

MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement

LLM-Core Xiaomi

Reinforcement LearningSelf-ImprovementMoELong ContextMixed-Task Agentic RLReward Hacking Defense
arXiv 110
Liang-Hsun Huang 許願