Decoupling Structure and Lexicon for Zero-Shot Semantic Parsing
This addresses the challenge of quickly deploying conversational interfaces in new domains, though it is incremental as it builds on existing zero-shot parsing methods.
The paper tackles the problem of building semantic parsers that can generalize to new domains without expensive supervision, achieving an average accuracy of 53.4% on 7 domains in the Overnight dataset, which is competitive with parsers trained on over 30% of target domain data.
Building a semantic parser quickly in a new domain is a fundamental challenge for conversational interfaces, as current semantic parsers require expensive supervision and lack the ability to generalize to new domains. In this paper, we introduce a zero-shot approach to semantic parsing that can parse utterances in unseen domains while only being trained on examples in other source domains. First, we map an utterance to an abstract, domain-independent, logical form that represents the structure of the logical form, but contains slots instead of KB constants. Then, we replace slots with KB constants via lexical alignment scores and global inference. Our model reaches an average accuracy of 53.4% on 7 domains in the Overnight dataset, substantially better than other zero-shot baselines, and performs as good as a parser trained on over 30% of the target domain examples.