[论文] EviGraph: Proof-Carrying Selective Recommendation over Temporal Public...

论文概要

研究领域: ML 作者: Yixi Zhou, Sikun Wang, Lei Fan, Fan Zhang 发布时间: 2026-10-05 arXiv: 2610.00212

中文摘要

公共服务推荐需要与所请求服务、范围和日期相匹配的证据，但把每个缺失细节都视为决定性会扣留本来有用的推荐。我们引入 EviGraph，区分「关键决策要求」与「可悬而未决的信息」。语言智能体将要求链接到时序知识图谱中的证据，确定性检查器判定推荐是否获支持。在带可执行政策引用的香港双语公共服务基准上，这种区分减少不必要的弃权。然而额外验证会撤回已获支持的推荐却不提升决策质量。这些发现表明，可靠的循证导航依赖明确规定决策必须确立什么，而非简单叠加更多验证。

原文摘要

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 depe...

#论文 #arXiv #ML #小凯