Iterative Utterance Segmentation for Neural Semantic Parsing
This work addresses the problem of improving compositional generalization for neural semantic parsers, which is a common challenge in natural language understanding.
Neural semantic parsers struggle with long, complex utterances. This paper introduces an iterative utterance segmentation framework that segments an utterance and maps spans to partial meaning representations, which are then composed. This method significantly improved accuracy on compositional generalization tasks: Geo from 63.1 to 81.2, Formulas from 59.7 to 72.7, and ComplexWebQuestions from 27.1 to 56.3.
Neural semantic parsers usually fail to parse long and complex utterances into correct meaning representations, due to the lack of exploiting the principle of compositionality. To address this issue, we present a novel framework for boosting neural semantic parsers via iterative utterance segmentation. Given an input utterance, our framework iterates between two neural modules: a segmenter for segmenting a span from the utterance, and a parser for mapping the span into a partial meaning representation. Then, these intermediate parsing results are composed into the final meaning representation. One key advantage is that this framework does not require any handcraft templates or additional labeled data for utterance segmentation: we achieve this through proposing a novel training method, in which the parser provides pseudo supervision for the segmenter. Experiments on Geo, ComplexWebQuestions, and Formulas show that our framework can consistently improve performances of neural semantic parsers in different domains. On data splits that require compositional generalization, our framework brings significant accuracy gains: Geo 63.1 to 81.2, Formulas 59.7 to 72.7, ComplexWebQuestions 27.1 to 56.3.