论文标题
结构化重新排序和生育层的组成概括
Compositional Generalisation with Structured Reordering and Fertility Layers
论文作者
论文摘要
已显示SEQ2SEQ模型在组成概括中遇到困难，即比训练期间所看到的更复杂的新结构。从基于语法的模型中汲取灵感，在构图概括方面表现出色，我们提出了一个柔性的端到端可区分模型，该模型构成了两个结构性操作：我们在这项工作中介绍的生育步骤，以及基于先前工作的重新排序步骤（Wang等，2021年）。为了确保可怜性，我们使用每个步骤的预期值。我们的模型通过在需要概括到更长示例的逼真的语义解析任务的挑战性构图上，优于SEQ2SEQ模型。它也与针对组成概括的其他模型进行了有利的比较。
Seq2seq models have been shown to struggle with compositional generalisation, i.e. generalising to new and potentially more complex structures than seen during training. Taking inspiration from grammar-based models that excel at compositional generalisation, we present a flexible end-to-end differentiable neural model that composes two structural operations: a fertility step, which we introduce in this work, and a reordering step based on previous work (Wang et al., 2021). To ensure differentiability, we use the expected value of each step. Our model outperforms seq2seq models by a wide margin on challenging compositional splits of realistic semantic parsing tasks that require generalisation to longer examples. It also compares favourably to other models targeting compositional generalisation.
