# Compositional Semantic Parsing Across Graphbanks

## TL;DR

This paper introduces a neural semantic parser capable of accurately mapping sentences to various graph-based meaning representations, leveraging BERT embeddings and multi-task learning to achieve state-of-the-art results across multiple graphbanks.

## Contribution

It presents a novel compositional neural semantic parser that generalizes across different graphbanks, a significant advancement over hand-designed parsers for specific datasets.

## Key findings

- Achieves competitive accuracy across diverse graphbanks
- Incorporates BERT embeddings and multi-task learning for improved performance
- Sets new state-of-the-art results on DM, PAS, PSD, AMR 2015, and EDS

## Abstract

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.