发表机构
University of Kansas; Capital One（堪萨斯大学; 第一资本）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出双尺度图变换器HERMES，联合建模局部关系与全局生态上下文，在1.3亿高风险会话上实现客户摩擦降低44.44%、欺诈召回提升24.66%。
AI 中文摘要
账户接管（ATO）欺诈对数字银行构成日益严重的威胁，需要在有效检测的同时尽量减少对合法客户的干扰。生产系统主要依赖表格模型，这些模型孤立地对会话进行评分，丢弃了底层交互网络的关系结构。尽管基于图的模型利用了会话与网络实体之间的关系，但它们主要基于局部邻域进行推理，因此仅捕获了问题的一部分：欺诈风险同时取决于会话周围的局部关系结构和欺诈生态系统演变的全局状态。我们提出了HERMES（用于高风险会话的异构关系微-宏观图变换器编码器），这是一种双尺度架构，联合建模这些互补的信息尺度。Micro-GT通过结构化的、关系感知的注意力机制，在时间安全的会话邻域上捕获局部异构图结构。作为局部表示的补充，Macro-GT使用非预期性的气候令牌来建模生态系统级上下文，这些令牌总结了欺诈动态、平台转移和基础设施重用，同时使用自适应类原型来跟踪随时间变化的代表性欺诈和良性会话模式。在来自美国一家领先金融机构的超过1.3亿个高风险交易会话上进行的评估中，HERMES始终优于生产系统和强大的基于图的基线，与生产系统相比，客户摩擦相对减少44.44%，欺诈召回率相对提高24.66%。消融和时间稳定性分析进一步证明了局部关系建模和全局生态系统上下文在不断变化的欺诈体制中的互补增益。
英文摘要
Account takeover (ATO) fraud is a growing threat to digital banking, requiring effective detection while minimizing friction for legitimate customers. Production systems predominantly rely on tabular models that score sessions in isolation, discarding the relational structure of the underlying interaction network. Although graph-based models exploit relationships among sessions and network entities, they primarily reason over local neighborhoods and therefore capture only part of the problem: fraud risk depends jointly on the local relational structure surrounding a session and the evolving global state of the fraud ecosystem. We present HERMES (HEterogeneous Relational Micro--macro graph transformer Encoder for high-risk Sessions), a dual-scale architecture that jointly models these complementary scales of information. Micro-GT captures local heterogeneous graph structure through structured, relation-aware attention over a temporally safe session neighborhood. Complementing this local representation, Macro-GT models ecosystem-level context using non-anticipative climate tokens that summarize fraud dynamics, platform shifts, and infrastructure reuse, together with adaptive class prototypes that track representative fraud and benign session patterns over time. Evaluated on more than 130 million high-risk transaction sessions from a leading U.S. financial institution, HERMES consistently outperforms production and strong graph-based baselines, achieving a 44.44% relative reduction in customer friction and a 24.66% relative improvement in fraud recall over the production system. Ablation and temporal-stability analyses further demonstrate complementary gains from local relational modeling and global ecosystem context across changing fraud regimes.
CommentsThis paper has been accepted at the Geometric Distributional Deep Learning (GDDL) workshop at NeurIPS 2026
