Translate First Reorder Later: Leveraging Monotonicity in Semantic Parsing
This addresses the limitation of seq2seq models in handling compositional tasks for semantic parsing, offering a modular solution that leverages alignment annotations, though it is incremental as it builds on prior alignment-based methods.
The paper tackles the problem of compositional generalization in semantic parsing by proposing TPOL, a two-step approach that first translates sentences monotonically and then reorders them, which significantly improves performance over conventional seq2seq models and other alignment-based methods on two datasets.
Prior work in semantic parsing has shown that conventional seq2seq models fail at compositional generalization tasks. This limitation led to a resurgence of methods that model alignments between sentences and their corresponding meaning representations, either implicitly through latent variables or explicitly by taking advantage of alignment annotations. We take the second direction and propose TPOL, a two-step approach that first translates input sentences monotonically and then reorders them to obtain the correct output. This is achieved with a modular framework comprising a Translator and a Reorderer component. We test our approach on two popular semantic parsing datasets. Our experiments show that by means of the monotonic translations, TPOL can learn reliable lexico-logical patterns from aligned data, significantly improving compositional generalization both over conventional seq2seq models, as well as over other approaches that exploit gold alignments.