MarkTechPost(RSS)📅 Jul 4论文

NVIDIA 联合多所大学提出 ASPIRE:自我改进机器人框架,零样本成功率最高提升 77 分

NVIDIA 联合密歇根大学、UIUC、UC Berkeley 等提出 ASPIRE,一个持续学习机器人框架。它通过协调器-执行器架构、闭环执行引擎、技能库和进化搜索,编写并优化机器人控制程序。编程智能体使用 Claude Code(Claude Opus 4.6,1M token 上下文窗口)。在 LIBERO-Pro 上最高比最强基线提升 77 分;Robosuite 双手交接成功率从 20% 提升至 92%;BEHAVIOR-1K 收音机拾取任务从 56% 提升至 88%。利用 LIBERO-90 积累的技能,ASPIRE 在零样本条件下对 LIBERO-Pro Long 任务达到约 31% 成功率,此前方法饱和在 4% 附近。

💡 Recommended AI Tools

The large video model released by Shengshu Technology and Tsinghua University supports highly consistent simulation of characters and physical laws.
The large video generation model launched by Kuaishou supports the generation of 1080p frame rate videos up to 2 minutes long.
ByteDance's AI painting and short video generation platform supports one-click text generation and shadow generation.
Tencent's official AI assistant deeply integrates WeChat public account ecology, search and file parsing capabilities.