SABET-QA：时序知识图谱问答

中文摘要 AI 辅助

时序知识图谱问答（TKGQA）需要对时间敏感的事实进行推理，但现有的基于嵌入的方法因采用单步推理管道，难以处理多步查询。我们提出SABET-QA框架，通过双向实体-时间评分机制和槽感知上下文模块（将问题语义与时序知识图谱嵌入对齐），跨多跳迭代优化推理状态；可微分工作记忆支持逐步假设优化，辅助时间边界在可用时作为粗监督。在CronQuestions、Complex-CronQuestions、MultiTQ和TimeQuestions上的实验表明，该框架优于强基线，尤其在复杂多步时序查询上表现突出。

英文摘要

Question Answering over Temporal Knowledge Graphs (TKGQA) requires reasoning over time-sensitive facts, yet existing embedding-based methods struggle with multi-step queries due to single-pass reasoning pipelines. We propose SABET-QA, a framework that iteratively refines reasoning states across multiple hops via a bidirectional entity-temporal scoring mechanism and a slot-aware contextualization module that aligns question semantics with temporal KG embeddings. A differentiable working memory enables progressive hypothesis refinement, while auxiliary temporal boundaries serve as coarse supervision when available. Experiments on CronQuestions, Complex-CronQuestions, MultiTQ, and TimeQuestions demonstrate consistent improvements over strong baselines, particularly on complex multi-step temporal queries.

发表机构

• ENS Paris-Saclay（巴黎-萨克雷高等师范学校）
• École Polytechnique（巴黎综合理工学院）
• QuickSort Research（快排研究院）

机构由 AI 辅助整理，请以论文原文为准。