Neural QCFG is a grammar-based sequence-tosequence (seq2seq) model with strong inductive biases on hierarchical structures. It excels in interpretability and generalization but suffers from expensive inference. In this paper, we study two low-rank variants of Neural QCFG for faster inference with different trade-offs between efficiency and expressiveness. Furthermore, utilizing the symbolic interface provided by the grammar, we introduce two soft constraints over tree hierarchy and source coverage. We experiment with various datasets and find that our models outperform vanilla Neural QCFG in most settings.

翻译：神经QCFG是一种基于语法的序列到序列（seq2seq）模型，具有对层级结构的强归纳偏置。该模型在可解释性和泛化能力方面表现优异，但存在推理成本高昂的问题。本文研究了两种低秩变体神经QCFG，它们在效率与表达能力之间呈现不同的权衡，适用于更快速的推理。此外，借助语法提供的符号接口，我们引入了关于树层级结构和源端覆盖率的两种软约束。我们在多个数据集上进行实验，发现所提模型在大多数设置中均优于标准神经QCFG。