Compositional Generalization via Semantic Tagging
This addresses a key limitation in semantic parsing for applications requiring systematic generalization, though it is an incremental improvement over existing methods.
The paper tackles the problem of compositional generalization in neural sequence-to-sequence models for semantic parsing by proposing a decoding framework that uses semantic tagging to decompose the process, resulting in consistent improvements across datasets, domains, and architectures.
Although neural sequence-to-sequence models have been successfully applied to semantic parsing, they fail at compositional generalization, i.e., they are unable to systematically generalize to unseen compositions of seen components. Motivated by traditional semantic parsing where compositionality is explicitly accounted for by symbolic grammars, we propose a new decoding framework that preserves the expressivity and generality of sequence-to-sequence models while featuring lexicon-style alignments and disentangled information processing. Specifically, we decompose decoding into two phases where an input utterance is first tagged with semantic symbols representing the meaning of individual words, and then a sequence-to-sequence model is used to predict the final meaning representation conditioning on the utterance and the predicted tag sequence. Experimental results on three semantic parsing datasets show that the proposed approach consistently improves compositional generalization across model architectures, domains, and semantic formalisms.