Generate-and-Retrieve: use your predictions to improve retrieval for semantic parsing
This addresses the challenge of effective retrieval for semantic parsing, particularly in low-resource scenarios, with incremental improvements over existing methods.
The paper tackles the problem of retrieving informative exemplars for semantic parsing in low-resource settings by proposing GandR, a method that uses preliminary predictions to improve retrieval, resulting in state-of-the-art performance on multiple tasks.
A common recent approach to semantic parsing augments sequence-to-sequence models by retrieving and appending a set of training samples, called exemplars. The effectiveness of this recipe is limited by the ability to retrieve informative exemplars that help produce the correct parse, which is especially challenging in low-resource settings. Existing retrieval is commonly based on similarity of query and exemplar inputs. We propose GandR, a retrieval procedure that retrieves exemplars for which outputs are also similar. GandRfirst generates a preliminary prediction with input-based retrieval. Then, it retrieves exemplars with outputs similar to the preliminary prediction which are used to generate a final prediction. GandR sets the state of the art on multiple low-resource semantic parsing tasks.