Training Reasoning Models in Simulation to Natively Use Real Robots
暂无中文版,显示英文原文。
摘自论文摘要
Robot-use agents (RUAs) have emerged as a promising approach to general-purpose robot control. However, strong performance has so far been achieved only by frontier proprietary models such as GPT-6 Astra and Claude Fable 5.2, whereas open-weight models remain substantially less capable.
摘自论文摘要,摘要生成中。
受许可协议限制,本站不转载该论文全文,仅提供摘要和原文链接。