发表机构
Hong Kong Baptist University; Tokyo University of Science; University of Illinois Urbana-Champaign; The University of Tokyo（香港浸会大学; 东京理科大学; 伊利诺伊大学厄巴纳-香槟分校; 东京大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
EviGraph通过区分关键决策需求与可未解决信息，利用语言智能体和确定性检查器在时序知识图谱中验证证据，减少公共服务推荐中的不必要弃权，强调明确决策要求而非增加验证。
AI 中文摘要
公共服务推荐需要与所请求的服务、范围和日期相匹配的证据。然而，将每个缺失的细节都视为决定性因素可能会扣留有用的推荐。我们引入了EviGraph，它区分了关键的决策需求与可以保持未解决的信息。一个语言智能体将这些需求与时序知识图谱中的证据联系起来，而一个确定性检查器则确定推荐是否得到支持。在一个带有可执行政策参考的双语香港公共服务基准上的评估表明，这种区分减少了不必要的弃权（不执行）。然而，额外的验证可能会撤回已支持的推荐，而不会提高决策质量。这些发现表明，可靠的基于证据的导航依赖于明确说明决策必须确立什么，而不是简单地增加更多验证。
英文摘要
Public-service recommendations require evidence that matches the requested service, scope, and date. Yet treating every missing detail as decisive can withhold useful recommendations. We introduce EviGraph, which distinguishes critical decision requirements from information that can remain unresolved. A language agent links these requirements to evidence in a temporal knowledge graph, while a deterministic checker establishes whether a recommendation is supported. Evaluation on a bilingual Hong Kong public-service benchmark with executable policy references shows that this distinction reduces unnecessary abstention. Additional verification, however, can withdraw supported recommendations without improving decision quality. These findings suggest that reliable evidence-based navigation depends on specifying what must be established for a decision, rather than simply adding more verification.
Comments19 pages, including figures and tables
