A pointer network architecture for context-dependent semantic parsing

Xuanli He; Quan Huang Tran; Reza Haffari

A pointer network architecture for context-dependent semantic parsing

Xuanli He, Quan Huang Tran, Reza Haffari

Department of Data Science & AI

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

Abstract

Semantic parsing targets at mapping human
utterances into structured meaning representations,
such as logical forms, programming
snippets, SQL queries etc. In this work, we
focus on logical form generation, which is extracted
from an automated email assistant system.
Since this task is dialogue-oriented, information
across utterances must be well handled.
Furthermore, certain inputs from users
are used as arguments for the logical form,
which requires a parser to distinguish the functional
words and content words. Hence, an
intelligent parser should be able to switch between
generation mode and copy mode. In order
to address the aforementioned issues, we
equip the vanilla seq2seq model with a pointer
network and a context-dependent architecture
to generate more accurate logical forms. Our
model achieves state-of-the-art performance
on the email assistant task.

Original language	English
Title of host publication	Proceedings of the The 17th Annual Workshop of the Australasian Language Technology Association
Editors	Meladel Mistica
Place of Publication	Stroudsburg PA USA
Publisher	Association for Computational Linguistics (ACL)
Pages	94–99
Number of pages	6
Publication status	Published - 2019
Event	Australasian Language Technology Association Workshop 2019 - Sydney, Australia Duration: 4 Nov 2019 → 6 Nov 2019 Conference number: 17th https://aclanthology.org/volumes/U19-1/ (Proceedings)

Conference

Conference	Australasian Language Technology Association Workshop 2019
Abbreviated title	ALTA 2019
Country/Territory	Australia
City	Sydney
Period	4/11/19 → 6/11/19
Internet address	https://aclanthology.org/volumes/U19-1/ (Proceedings)

Access to Document

361476912-oaFinal published version, 342 KB

https://aclanthology.org/U19-1013

Exploiting Context in Multilingual Understanding and Generation
Haffari, R. (Primary Chief Investigator (PCI))
Australian Research Council (ARC)
20/11/20 → 28/02/26
Project: Research

Cite this

@inproceedings{4b5a8bf1347042aab1cdb1050a24426f,

title = "A pointer network architecture for context-dependent semantic parsing",

abstract = "Semantic parsing targets at mapping humanutterances into structured meaning representations,such as logical forms, programmingsnippets, SQL queries etc. In this work, wefocus on logical form generation, which is extractedfrom an automated email assistant system.Since this task is dialogue-oriented, informationacross utterances must be well handled.Furthermore, certain inputs from usersare used as arguments for the logical form,which requires a parser to distinguish the functionalwords and content words. Hence, anintelligent parser should be able to switch betweengeneration mode and copy mode. In orderto address the aforementioned issues, weequip the vanilla seq2seq model with a pointernetwork and a context-dependent architectureto generate more accurate logical forms. Ourmodel achieves state-of-the-art performanceon the email assistant task.",

author = "Xuanli He and {Huang Tran}, Quan and Reza Haffari",

year = "2019",

language = "English",

pages = "94–99",

editor = "Meladel Mistica",

booktitle = "Proceedings of the The 17th Annual Workshop of the Australasian Language Technology Association",

publisher = "Association for Computational Linguistics (ACL)",

note = "Australasian Language Technology Association Workshop 2019, ALTA 2019 ; Conference date: 04-11-2019 Through 06-11-2019",

url = "https://aclanthology.org/volumes/U19-1/",

}

He, X, Huang Tran, Q & Haffari, R 2019, A pointer network architecture for context-dependent semantic parsing. in M Mistica (ed.), Proceedings of the The 17th Annual Workshop of the Australasian Language Technology Association., U19-1013, Association for Computational Linguistics (ACL), Stroudsburg PA USA, pp. 94–99, Australasian Language Technology Association Workshop 2019, Sydney, New South Wales, Australia, 4/11/19. <https://aclanthology.org/U19-1013>

A pointer network architecture for context-dependent semantic parsing. / He, Xuanli; Huang Tran, Quan; Haffari, Reza.
Proceedings of the The 17th Annual Workshop of the Australasian Language Technology Association. ed. / Meladel Mistica. Stroudsburg PA USA: Association for Computational Linguistics (ACL), 2019. p. 94–99 U19-1013.

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

TY - GEN

T1 - A pointer network architecture for context-dependent semantic parsing

AU - He, Xuanli

AU - Huang Tran, Quan

AU - Haffari, Reza

N1 - Conference code: 17th

PY - 2019

Y1 - 2019

N2 - Semantic parsing targets at mapping humanutterances into structured meaning representations,such as logical forms, programmingsnippets, SQL queries etc. In this work, wefocus on logical form generation, which is extractedfrom an automated email assistant system.Since this task is dialogue-oriented, informationacross utterances must be well handled.Furthermore, certain inputs from usersare used as arguments for the logical form,which requires a parser to distinguish the functionalwords and content words. Hence, anintelligent parser should be able to switch betweengeneration mode and copy mode. In orderto address the aforementioned issues, weequip the vanilla seq2seq model with a pointernetwork and a context-dependent architectureto generate more accurate logical forms. Ourmodel achieves state-of-the-art performanceon the email assistant task.

AB - Semantic parsing targets at mapping humanutterances into structured meaning representations,such as logical forms, programmingsnippets, SQL queries etc. In this work, wefocus on logical form generation, which is extractedfrom an automated email assistant system.Since this task is dialogue-oriented, informationacross utterances must be well handled.Furthermore, certain inputs from usersare used as arguments for the logical form,which requires a parser to distinguish the functionalwords and content words. Hence, anintelligent parser should be able to switch betweengeneration mode and copy mode. In orderto address the aforementioned issues, weequip the vanilla seq2seq model with a pointernetwork and a context-dependent architectureto generate more accurate logical forms. Ourmodel achieves state-of-the-art performanceon the email assistant task.

M3 - Conference Paper

SP - 94

EP - 99

BT - Proceedings of the The 17th Annual Workshop of the Australasian Language Technology Association

A2 - Mistica, Meladel

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg PA USA

T2 - Australasian Language Technology Association Workshop 2019

Y2 - 4 November 2019 through 6 November 2019

ER -

A pointer network architecture for context-dependent semantic parsing

Abstract

Conference

Access to Document

Projects

Exploiting Context in Multilingual Understanding and Generation

Cite this