研究人员开发了InterEvolve，一种用于人形机器人测试时奖励程序演化的新颖系统。该方法允许机器人在不重新训练的情况下解决新任务，方法是重新利用现有技能并从自己的尝试中学习。InterEvolve利用了面向对象的基金模型和大型语言模型代理来动态修改和优化奖励程序，使机器人能够适应并提高其在复杂运动操纵任务上的性能。 AI
影响 使机器人能够在不重新训练的情况下适应新任务，有可能加速在动态环境中的部署。
排序理由 该集群包含一篇详细介绍机器人学习新方法的学术论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, without retraining. Our key insight is that a broad controller alr…
arXiv:2610.02196v1 Announce Type: cross Abstract: We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, without retraini…