Jointly predicting predicates and arguments in neural semantic role labeling

Luheng He; Kenton Lee; Omer Levy; Luke Zettlemoyer

doi:10.18653/v1/p18-2058

Jointly predicting predicates and arguments in neural semantic role labeling

Luheng He, Kenton Lee, Omer Levy, Luke Zettlemoyer

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

Abstract

Recent BIO-tagging-based neural semantic role labeling models are very high performing, but assume gold predicates as part of the input and cannot incorporate span-level features. We propose an end-to-end approach for jointly predicting all predicates, arguments spans, and the relations between them. The model makes independent decisions about what relationship, if any, holds between every possible word-span pair, and learns contextualized span representations that provide rich, shared input features for each decision. Experiments demonstrate that this approach sets a new state of the art on PropBank SRL without gold predicates.

Original language	English
Title of host publication	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers)
Publisher	Association for Computational Linguistics (ACL)
Pages	364-369
Number of pages	6
ISBN (Electronic)	9781948087346
DOIs	https://doi.org/10.18653/v1/p18-2058
State	Published - 2018
Externally published	Yes
Event	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 - Melbourne, Australia Duration: 15 Jul 2018 → 20 Jul 2018

Publication series

Name	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
Volume	2

Conference

Conference	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018
Country/Territory	Australia
City	Melbourne
Period	15/07/18 → 20/07/18

Access to Document

10.18653/v1/p18-2058

Cite this

He, L., Lee, K., Levy, O., & Zettlemoyer, L. (2018). Jointly predicting predicates and arguments in neural semantic role labeling. In ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers) (pp. 364-369). (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 2). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/p18-2058

He, Luheng ; Lee, Kenton ; Levy, Omer et al. / Jointly predicting predicates and arguments in neural semantic role labeling. ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). Association for Computational Linguistics (ACL), 2018. pp. 364-369 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)).

@inproceedings{17b9d4443eca42e78912c85134ab5224,

title = "Jointly predicting predicates and arguments in neural semantic role labeling",

abstract = "Recent BIO-tagging-based neural semantic role labeling models are very high performing, but assume gold predicates as part of the input and cannot incorporate span-level features. We propose an end-to-end approach for jointly predicting all predicates, arguments spans, and the relations between them. The model makes independent decisions about what relationship, if any, holds between every possible word-span pair, and learns contextualized span representations that provide rich, shared input features for each decision. Experiments demonstrate that this approach sets a new state of the art on PropBank SRL without gold predicates.",

author = "Luheng He and Kenton Lee and Omer Levy and Luke Zettlemoyer",

note = "Publisher Copyright: c 2018 Association for Computational Linguistics; 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 ; Conference date: 15-07-2018 Through 20-07-2018",

year = "2018",

doi = "10.18653/v1/p18-2058",

language = "אנגלית",

series = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)",

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

pages = "364--369",

booktitle = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers)",

address = "ארצות הברית",

}

He, L, Lee, K, Levy, O & Zettlemoyer, L 2018, Jointly predicting predicates and arguments in neural semantic role labeling. in ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), vol. 2, Association for Computational Linguistics (ACL), pp. 364-369, 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018, Melbourne, Australia, 15/07/18. https://doi.org/10.18653/v1/p18-2058

Jointly predicting predicates and arguments in neural semantic role labeling. / He, Luheng; Lee, Kenton; Levy, Omer et al.
ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). Association for Computational Linguistics (ACL), 2018. p. 364-369 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 2).

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

TY - GEN

T1 - Jointly predicting predicates and arguments in neural semantic role labeling

AU - He, Luheng

AU - Lee, Kenton

AU - Levy, Omer

AU - Zettlemoyer, Luke

PY - 2018

Y1 - 2018

N2 - Recent BIO-tagging-based neural semantic role labeling models are very high performing, but assume gold predicates as part of the input and cannot incorporate span-level features. We propose an end-to-end approach for jointly predicting all predicates, arguments spans, and the relations between them. The model makes independent decisions about what relationship, if any, holds between every possible word-span pair, and learns contextualized span representations that provide rich, shared input features for each decision. Experiments demonstrate that this approach sets a new state of the art on PropBank SRL without gold predicates.

AB - Recent BIO-tagging-based neural semantic role labeling models are very high performing, but assume gold predicates as part of the input and cannot incorporate span-level features. We propose an end-to-end approach for jointly predicting all predicates, arguments spans, and the relations between them. The model makes independent decisions about what relationship, if any, holds between every possible word-span pair, and learns contextualized span representations that provide rich, shared input features for each decision. Experiments demonstrate that this approach sets a new state of the art on PropBank SRL without gold predicates.

UR - http://www.scopus.com/inward/record.url?scp=85063106932&partnerID=8YFLogxK

U2 - 10.18653/v1/p18-2058

DO - 10.18653/v1/p18-2058

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85063106932

T3 - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)

SP - 364

EP - 369

BT - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers)

PB - Association for Computational Linguistics (ACL)

T2 - 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018

Y2 - 15 July 2018 through 20 July 2018

ER -

He L, Lee K, Levy O, Zettlemoyer L. Jointly predicting predicates and arguments in neural semantic role labeling. In ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Short Papers). Association for Computational Linguistics (ACL). 2018. p. 364-369. (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)). doi: 10.18653/v1/p18-2058

Jointly predicting predicates and arguments in neural semantic role labeling

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this