arXiv · cs.AI · Kangning Zhang, Wenxiang Jiao, Bowen Fang, Shuai Shao, Jiachen Zhu, Rong Shan, Shijian Wang, Hao Wang, Yuan Lu, Weiwen Liu, Weinan Zhang · 4 小时前模型评分45
Harness-S：用同级 Harness 互相补位，提升智能体训练泛化能力
Harness-S: Learning from Sibling Harnesses for Generalizable Agent Training
导读
Harness-S 通过“同级 Harness 池”在基础执行系统全部尝试失败时调用辅助 Harness 重新生成轨迹，替换失败组用于强化学习，从而把能力迁移进单一策略。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
事件类型：研究发布
证据摘录
- “Rather than seeking a universally superior harness, we introduce Harness-S to selectively rescue failures of a base execution system through sibling harnesses and transfer the resulting capabilities into a single policy that remains effective under the original harness.”
来源：arXiv · cs.AI · arxiv.org