Universal Semantic Parsing

Siva Reddy; Oscar Täckström; Slav  Petrov; Mark Steedman; Mirella Lapata

doi:10.18653/v1/D17-1009

Universal Semantic Parsing

Siva Reddy, Oscar Täckström, Slav Petrov, Mark Steedman, Mirella Lapata

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Universal Dependencies (UD) offer a uniform cross-lingual syntactic representation, with the aim of advancing multilingual applications. Recent work shows that semantic parsing can be accomplished by transforming syntactic dependencies to logical forms. However, this work is limited to English, and cannot process dependency graphs, which allow handling complex phenomena such as control. In this work, we introduce UDEPLAMBDA, a semantic interface for UD, which maps natural language to logical forms in an almost language-independent fashion and can process dependency graphs. We perform experiments on question answering against Freebase and provide German and Spanish translations of the WebQuestions and GraphQuestions datasets to facilitate multilingual evaluation. Results show that UDEPLAMBDA outperforms strong base-lines across languages and datasets. For English, it achieves a 4.9F1point improvement over the state-of-the-art on GraphQuestions.

Original language	English
Title of host publication	Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
Place of Publication	Copenhagen, Denmark
Publisher	Association for Computational Linguistics
Pages	89–101
Number of pages	13
ISBN (Print)	978-1-945626-83-8
DOIs	https://doi.org/10.18653/v1/D17-1009
Publication status	Published - 11 Sept 2017
Event	EMNLP 2017: Conference on Empirical Methods in Natural Language Processing - Copenhagen, Denmark Duration: 7 Sept 2017 → 11 Sept 2017 http://emnlp2017.net/index.html http://emnlp2017.net/

Conference

Conference	EMNLP 2017: Conference on Empirical Methods in Natural Language Processing
Abbreviated title	EMNLP 2017
Country/Territory	Denmark
City	Copenhagen
Period	7/09/17 → 11/09/17
Internet address	http://emnlp2017.net/index.html http://emnlp2017.net/

Access to Document

10.18653/v1/D17-1009Licence: Creative Commons: Attribution (CC-BY)

1702.03196-1Accepted author manuscript, 763 KB
D17-1009Final published version, 464 KBLicence: Creative Commons: Attribution (CC-BY)

http://aclweb.org/anthology/D17-1009Licence: Creative Commons: Attribution (CC-BY)

TransModal: Translating from Multiple Modalities into Text
Lapata, M.
EU government bodies
1/09/16 → 31/08/22
Project: Research

Mirella Lapata
- mlapinf.ed.acuk
- School of Informatics - Personal Chair in Natural Language Processing
- Institute of Language, Cognition and Computation
- Language, Interaction, and Robotics
Person: Academic: Research Active

Cite this

@inproceedings{09d7a4a4486a4fdd870c4075f9bd6acf,

title = "Universal Semantic Parsing",

abstract = "Universal Dependencies (UD) offer a uniform cross-lingual syntactic representation, with the aim of advancing multilingual applications. Recent work shows that semantic parsing can be accomplished by transforming syntactic dependencies to logical forms. However, this work is limited to English, and cannot process dependency graphs, which allow handling complex phenomena such as control. In this work, we introduce UDEPLAMBDA, a semantic interface for UD, which maps natural language to logical forms in an almost language-independent fashion and can process dependency graphs. We perform experiments on question answering against Freebase and provide German and Spanish translations of the WebQuestions and GraphQuestions datasets to facilitate multilingual evaluation. Results show that UDEPLAMBDA outperforms strong base-lines across languages and datasets. For English, it achieves a 4.9F1point improvement over the state-of-the-art on GraphQuestions.",

author = "Siva Reddy and Oscar T{\"a}ckstr{\"o}m and Slav Petrov and Mark Steedman and Mirella Lapata",

year = "2017",

month = sep,

day = "11",

doi = "10.18653/v1/D17-1009",

language = "English",

isbn = "978-1-945626-83-8",

pages = "89–101",

booktitle = "Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing",

publisher = "Association for Computational Linguistics",

note = "EMNLP 2017: Conference on Empirical Methods in Natural Language Processing, EMNLP 2017 ; Conference date: 07-09-2017 Through 11-09-2017",

url = "http://emnlp2017.net/index.html, http://emnlp2017.net/",

}

Reddy, S, Täckström, O, Petrov, S, Steedman, M & Lapata, M 2017, Universal Semantic Parsing. in Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Association for Computational Linguistics, Copenhagen, Denmark , pp. 89–101, EMNLP 2017: Conference on Empirical Methods in Natural Language Processing, Copenhagen, Denmark, 7/09/17. https://doi.org/10.18653/v1/D17-1009

TY - GEN

T1 - Universal Semantic Parsing

AU - Reddy, Siva

AU - Täckström, Oscar

AU - Petrov, Slav

AU - Steedman, Mark

AU - Lapata, Mirella

PY - 2017/9/11

Y1 - 2017/9/11

N2 - Universal Dependencies (UD) offer a uniform cross-lingual syntactic representation, with the aim of advancing multilingual applications. Recent work shows that semantic parsing can be accomplished by transforming syntactic dependencies to logical forms. However, this work is limited to English, and cannot process dependency graphs, which allow handling complex phenomena such as control. In this work, we introduce UDEPLAMBDA, a semantic interface for UD, which maps natural language to logical forms in an almost language-independent fashion and can process dependency graphs. We perform experiments on question answering against Freebase and provide German and Spanish translations of the WebQuestions and GraphQuestions datasets to facilitate multilingual evaluation. Results show that UDEPLAMBDA outperforms strong base-lines across languages and datasets. For English, it achieves a 4.9F1point improvement over the state-of-the-art on GraphQuestions.

AB - Universal Dependencies (UD) offer a uniform cross-lingual syntactic representation, with the aim of advancing multilingual applications. Recent work shows that semantic parsing can be accomplished by transforming syntactic dependencies to logical forms. However, this work is limited to English, and cannot process dependency graphs, which allow handling complex phenomena such as control. In this work, we introduce UDEPLAMBDA, a semantic interface for UD, which maps natural language to logical forms in an almost language-independent fashion and can process dependency graphs. We perform experiments on question answering against Freebase and provide German and Spanish translations of the WebQuestions and GraphQuestions datasets to facilitate multilingual evaluation. Results show that UDEPLAMBDA outperforms strong base-lines across languages and datasets. For English, it achieves a 4.9F1point improvement over the state-of-the-art on GraphQuestions.

U2 - 10.18653/v1/D17-1009

DO - 10.18653/v1/D17-1009

M3 - Conference contribution

SN - 978-1-945626-83-8

SP - 89

EP - 101

BT - Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

PB - Association for Computational Linguistics

CY - Copenhagen, Denmark

T2 - EMNLP 2017: Conference on Empirical Methods in Natural Language Processing

Y2 - 7 September 2017 through 11 September 2017

ER -

Universal Semantic Parsing

Abstract

Conference

Access to Document

Fingerprint

Projects

TransModal: Translating from Multiple Modalities into Text

Profiles

Mirella Lapata

Cite this