让模型关注事件对间的隐含时间线索，提升时序关系抽取的可解释性。
WISTERIA: Weak Implicit Signal-based Temporal Relation Extraction with Attention
- 用配对条件化的注意力池化，提取每对事件的关键上下文词元。
- 在多个数据集上达到领先性能，且推理过程与语言学规律一致。
- 适合需要透明决策过程的医疗、法律等时间敏感场景应用。
时序关系抽取（TRE）旨在识别两个事件或时序表达之间的时序关联。现有基于注意力的模型常聚焦全局显著词元，忽视决定时序关系的配对特异性线索。本文提出WISTERIA框架，通过分析每个事件对条件下的前K个注意力成分是否蕴含可解释的时序分类证据，来改进模型。不同于依赖显式标记（如before、after）的方法，WISTERIA将任何隐含表达时序顺序的词汇、句法或形态元素视为信号。结合多头注意力与配对条件化的前K池化，模型能为每对事件隔离最相关上下文词元。在TimeBank-Dense、MATRES、TDDMan和TDDAuto上进行大量实验，并对前K词元进行语言学分析。结果表明，WISTERIA在准确率上表现竞争力，且揭示的配对级推理依据与时序语言学线索高度一致，提供局部化、可解释的时序推理视角。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。