arXiv · cs.AI · Genliang Zhu, Xiaobo Chen, Ziyuan Wang, Chu Wang · 11 小时前模型评分41
证据引导的 人工智能（AI，Artificial Intelligence） 智能体恢复方法：应对工具结果不明与部分执行
Evidence-Guided Recovery for AI Agents with Ambiguous Tool Outcomes and Partial Execution
导读
针对工具超时导致 人工智能（AI，Artificial Intelligence） 智能体面临结果不明与部分执行的问题，研究者提出证据引导的任务恢复方法，结合标准观测 contingent 规划与独立实现的效果历史裁决器。
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
- “We formulate evidence-guided task recovery over approved obligations, provider capabilities and states compatible with public evidence.”
- “In a finite study of 37 assignments, planning and three active controls each certify 28 completions; four saved synthetic fee units incur additional computation.”
- “Across 58 paired units in 19 previously developed task families, advice certifies 30 completions versus 9, a 36.2-percentage-point difference (exact family sign-flip p=0.0078), using 272 proposals versus 399.”
- “Unknown and handoff outcomes remain explicit, with guarantees tied to the declared provider capabilities rather than arbitrary third-party exactly-once execution.”
来源：arXiv · cs.AI · arxiv.org