Compositional Semantic Parsing Across Graphbanks
This addresses the need for more generalizable semantic parsing tools in natural language processing, though it is incremental as it builds on existing neural and BERT-based methods.
The paper tackled the problem of semantic parsers being limited to specific graphbanks by introducing a compositional neural semantic parser that achieved competitive accuracies across diverse graphbanks, setting new state-of-the-art results on DM, PAS, PSD, AMR 2015, and EDS.
Most semantic parsers that map sentences to graph-based meaning representations are hand-designed for specific graphbanks. We present a compositional neural semantic parser which achieves, for the first time, competitive accuracies across a diverse range of graphbanks. Incorporating BERT embeddings and multi-task learning improves the accuracy further, setting new states of the art on DM, PAS, PSD, AMR 2015 and EDS.