TEMPS：用于时间信息检索的时间句子嵌入

中文摘要 AI 辅助

现代信息检索（IR）系统很少表示时间，然而许多信息需求依赖于时间：在临床、新闻和法律检索中，事件发生的时间可以决定文档是否相关。密集检索器和检索增强生成（RAG）流水线在主题上能很好地将查询与文档匹配，但在时间上匹配不佳，因此它们会返回主题相关但时间错误的内容。我们引入了时间文本相似度（TTS）任务，该任务衡量两个锚定文本在时间上对齐的程度，与它们的主题相似性无关。然后，我们提出了TEMPS（精确搜索的时间嵌入模型），这是一个模块化的时间分支，附加到冻结的语义检索器上，并基于该信号进行训练。它将锚定的时间表达式解析为区间，并将每个表达式与高斯分布进行矩匹配；由此产生的排序监督一个锚定日期条件编码器，其得分在推理时与语义得分融合。接地提供了监督，因此训练不使用手工标注的时间数据。时间得分本身是来自分布嵌入的高斯KL包含度量；TEMPS增加的是接地和矩匹配监督。在三个时间基准上，TEMPS提高了每个测试的语义骨干的MRR，并且在TS-Retriever上，将R@1从19.92提升到25.39，超过了先前的时间最先进水平。

英文摘要

Modern information retrieval (IR) systems rarely represent time, yet many information needs depend on it: in clinical, journalistic, and legal search, when an event occurred can decide whether a document is relevant. Dense retrievers and Retrieval-Augmented Generation (RAG) pipelines match queries to documents well on topic but poorly on time, so they surface content that is on-topic yet temporally wrong. We introduce Temporal Textual Similarity (TTS), a task that measures how well two anchored texts align in time, independent of their topical similarity. We then present TEMPS (Temporal Embedding Model for Precise Search), a modular temporal branch that attaches to a frozen semantic retriever and trains on that signal. It resolves anchored temporal expressions to intervals and moment-matches each one to a Gaussian; the resulting ordering supervises an anchor-date-conditioned encoder, whose score we fuse with the semantic score at inference. Grounding supplies the supervision, so training uses no hand-labeled temporal data. The temporal score itself is the Gaussian-KL inclusion measure from distributional embeddings; what TEMPS adds is the grounding and the moment-matched supervision. On three temporal benchmarks, TEMPS improves MRR for every semantic backbone tested and, on TS- Retriever, lifts R@1 from 19.92 to 25.39 over the prior temporal state of the art.

发表机构

• SAMOVAR, Télécom SudParis, Institut Polytechnique de Paris（SAMOVAR，巴黎电信 SudParis，巴黎综合理工学院）
• Aldebaran Care

机构由 AI 辅助整理，请以论文原文为准。