Abstract
Temporal Relation Extraction (TRE) requires identifying how two events or temporal expressions are related in time. Existing attention-based models often highlight globally salient tokens but overlook the pair-specific cues that actually determine the temporal relation. We propose WISTERIA (Weak Implicit Signal-based Temporal Relation Extraction with Attention), a framework that examines whether the top-K attention components conditioned on each event pair truly encode interpretable evidence for temporal classification. Unlike prior works assuming explicit markers such as before, after, or when, WISTERIA considers signals as any lexical, syntactic, or morphological element implicitly expressing temporal order. By combining multi-head attention with pair-conditioned top-K pooling, the model isolates the most informative contextual tokens for each pair. We conduct extensive experiments on TimeBank-Dense, MATRES, TDDMan, and TDDAuto, including linguistic analyses of top-K tokens. Results show that WISTERIA achieves competitive accuracy and reveals pair-level rationales aligned with temporal linguistic cues, offering a localized and interpretable view of temporal reasoning.
Abstract (translated)
时间关系提取（Temporal Relation Extraction, TRE）需要识别两个事件或时间表达式在时间上的关联方式。现有的基于注意力的模型常聚焦于全局显著标记，却忽略了实际决定时间关系的事件对特定线索。我们提出WISTERIA（基于弱隐式信号与注意力机制的时间关系提取）框架，该框架检验了在 conditioning on 每个事件对条件下的top-K注意力成分是否真正编码了可解释的时间分类证据。与先前假设"之前"、"之后"或"当...时"等显性标记的研究不同，WISTERIA将信号定义为任何隐式表达时间顺序的词汇、句法或形态学元素。通过将多头注意力与事件对条件化的top-K池化相结合，模型能为每个事件对分离出最具信息量的上下文标记。我们在TimeBank-Dense、MATRES、TDDMan和TDDAuto数据集上进行了广泛实验，包括对top-K标记的语言学分析。结果表明，WISTERIA实现了具有竞争力的准确率，并揭示了与时间语言学线索对齐的事件级推理依据，为时间推理提供了局部化且可解释的视角。
URL
https://arxiv.org/abs/2603.23319
