TRANX: A Transition-based Neural Abstract Syntax Parser for Semantic Parsing and Code Generation

Pengcheng Yin, Graham Neubig
10/5/2018
cs.CLcs.PL

Official code: pcyin/tranX (464 stars, Apache-2.0, last commit 2022-11-12).

Abstract

We present TRANX, a transition-based neural semantic parser that maps natural language (NL) utterances into formal meaning representations (MRs). TRANX uses a transition system based on the abstract syntax description language for the target MR, which gives it two major advantages: (1) it is highly accurate, using information from the syntax of the target MR to constrain the output space and model the information flow, and (2) it is highly generalizable, and can easily be applied to new types of MR by just writing a new abstract syntax description corresponding to the allowable structures in the MR. Experiments on four different semantic parsing and code generation tasks show that our system is generalizable, extensible, and effective, registering strong results compared to existing neural semantic parsers.

View on arXivView PDF

Code Implementations(1)

pcyin/tranXOfficial100%
Apache-2.0

A general-purpose neural semantic parser for mapping natural language queries into machine executable code

464110CSS, HTMLAug 25, 20183 years agoApache-2.0

Similar Papers

Coarse-to-Fine Decoding for Neural Semantic Parsing

88% match1 repo1682018

SLING: A framework for frame semantic parsing

84% match1 repo1.9k2017

Discourse Representation Structure Parsing for Chinese

84% match2023

An Incremental Parser for Abstract Meaning Representation

84% match1 repo292016

Language to Logical Form with Neural Attention

83% match1 repo1552016

Laziness Is a Virtue When It Comes to Compositionality in Neural Semantic Parsing

83% match2023

Cite this paper

@article{yin2018tranx,
  title  = {TRANX: A Transition-based Neural Abstract Syntax Parser for Semantic Parsing and Code Generation},
  author = {Pengcheng Yin and Graham Neubig},
  year   = {2018},
  eprint = {1810.02720},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1810.02720}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1810.02720.svg)](https://codeofpaper.com/paper/1810.02720)

Discussion