FedV-KGQA 实践落地：设计经验与可交互原型系统
英文标题：FedV-KGQA in Practice: Design Lessons and an Interactive Prototype
针对现实中知识图谱常被不同机构垂直切分、单一系统无法掌握完整推理链的痛点，研究团队提出了面向垂直分区图谱的多跳问答框架 FedV-KGQA。该方案在原始三元组和关系嵌入完全不出本地的前提下，通过联邦融合恢复了中心化训练的大部分准确率，且效果远超单个数据孤岛。实验发现，实体锚定与本地图增强对性能的提升作用明显大于嵌入模型本身的选择，且最具性价比的编码器取决于目标精度而非参数规模。这项工作不仅总结了 4 条系统设计经验，还开源了支持单问题全链路追踪的交互式原型，为跨机构隐私保护与知识推理系统的落地提供了重要参考。
2.1 KB
640 B
612 B
380 B
9.1 KB
32.7 KB
22.8 KB
24.2 KB
34.7 KB
中文
[提交于 2026 年 9 月 12 日]
标题：FedV-KGQA 实践落地：设计经验与可交互原型系统
查看由 Md Saikat Islam Khan Bappy 及另 1 位作者撰写的论文《FedV-KGQA in Practice: Design Lessons and an Interactive Prototype》PDF：
摘要：知识图谱问答（KGQA）通常假设单个系统即可访问完整的图谱。但在实际应用中，事实信息往往分散在不同的机构手中，各方共享实体标识符却拥有互不相交的关系类型，没有任何一方能够掌握完整的推理链。本海报论文展示了 FedV-KGQA 在此类垂直分区图谱上进行多跳问答的实证研究结果。各数据孤岛（silo）丰富其本地图谱，并仅使用自身的三元组训练知识图谱嵌入。随后，服务器拼接来自各个孤岛的特定实体视图，将投影后的问题锚定在主题实体上，并通过相似度对候选实体进行排序。在此过程中，原始三元组与关系嵌入从未离开本地孤岛。对比 FedV-KGQA 的多组实验得出了三项核心结论：第一，联邦融合（federated fusion）能够恢复中心化模型的大部分准确率，而单个孤岛几乎无法有效推理；第二，实体锚定与本地图丰富化比嵌入模型的选择更为关键；第三，最经济高效的编码器取决于预期的目标准确率，而非单纯的参数量大小。本海报论文提供了这一跨实验对比分析、从中提炼出的四项设计经验，以及一个基于已发布检查点运行真实推理并追踪每道问题完整流水线的交互式原型系统。
提交历史
来自：Md Saikat Islam Khan Bappy [
]
[v1] 2026 年 9 月 12 日 星期六 02:39:19 UTC (585 KB)
English / 原文
[Submitted on 12 Sep 2026]
Title:FedV-KGQA in Practice: Design Lessons and an Interactive Prototype
View a PDF of the paper titled FedV-KGQA in Practice: Design Lessons and an Interactive Prototype, by Md Saikat Islam Khan Bappy and 1 other authors
Abstract:Knowledge graph question answering usually assumes that one system can reach the whole graph. In practice, facts are often held by organizations that share entity identifiers but own disjoint relation types, so no single party sees a complete reasoning chain. This poster presents the empirical findings of FedV-KGQA on multi-hop question answering over such vertically partitioned graphs. Each silo enriches its local graph and trains a knowledge graph embedding on its own triples. A server then concatenates the silo-specific entity views, anchors the projected question at the topic entity, and ranks candidates by similarity. Raw triples and relation embeddings never leave a silo. Comparing the FedV-KGQA experiments with one another yields three results. First, federated fusion recovers most of the centralized accuracy, while a single silo recovers little. Second, anchoring and enrichment matter more than the choice of embedding model. Third, the cheapest encoder depends on the target accuracy rather than on parameter count. This poster paper contributes that cross-experiment comparison, four design lessons drawn from it, and an interactive prototype that runs real inference and traces the full pipeline, per question, on released checkpoints.
Submission history
From: Md Saikat Islam Khan Bappy [
]
[v1]
Sat, 12 Sep 2026 02:39:19 UTC (585 KB)