Context-Dependent Semantic Parsing for Temporal Relation Extraction
This work addresses temporal relation extraction for applications like temporal reasoning and question answering, representing an incremental improvement through a hybrid neural-symbolic approach.
The paper tackles the problem of extracting temporal relations from text, which suffers from limited expressivity and inconsistent inference in existing methods, by proposing SMARTER, a neural semantic parser that generates executable logical forms using a custom typed lambda calculus, resulting in precise temporal relation graphs.
Extracting temporal relations among events from unstructured text has extensive applications, such as temporal reasoning and question answering. While it is difficult, recent development of Neural-symbolic methods has shown promising results on solving similar tasks. Current temporal relation extraction methods usually suffer from limited expressivity and inconsistent relation inference. For example, in TimeML annotations, the concept of intersection is absent. Additionally, current methods do not guarantee the consistency among the predicted annotations. In this work, we propose SMARTER, a neural semantic parser, to extract temporal information in text effectively. SMARTER parses natural language to an executable logical form representation, based on a custom typed lambda calculus. In the training phase, dynamic programming on denotations (DPD) technique is used to provide weak supervision on logical forms. In the inference phase, SMARTER generates a temporal relation graph by executing the logical form. As a result, our neural semantic parser produces logical forms capturing the temporal information of text precisely. The accurate logical form representations of an event given the context ensure the correctness of the extracted relations.