Gendered Ambiguous Pronoun (GAP) Shared Task at the Gender Bias in NLP Workshop 2019

Kellie Webster; Marta R. Costa-jussà; Christian Hardmeier; Will Radford

doi:10.18653/v1/W19-3801

Gendered Ambiguous Pronoun (GAP) Shared Task at the Gender Bias in NLP Workshop 2019

Kellie Webster, Marta R. Costa-jussà, Christian Hardmeier, Will Radford

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

Abstract

The 1st ACL workshop on Gender Bias in Natural Language Processing included a shared task on gendered ambiguous pronoun (GAP) resolution. This task was based on the coreference challenge defined in Webster et al. (2018), designed to benchmark the ability of systems to resolve pronouns in real-world contexts in a gender-fair way. 263 teams competed via a Kaggle competition, with the winning system achieving logloss of 0.13667 and near gender parity. We review the approaches of eleven systems with accepted description papers, noting their effective use of BERT (Devlin et al., 2018), both via fine-tuning and for feature extraction, as well as ensembling.

Originalsprog	Engelsk
Titel	Proceedings of the First Workshop on Gender Bias in Natural Language Processing
Antal sider	7
Publikationsdato	2 aug. 2019
ISBN (Trykt)	978-1-950737-40-6
DOI	https://doi.org/10.18653/v1/W19-3801
Status	Udgivet - 2 aug. 2019
Udgivet eksternt	Ja

Adgang til dokumentet

10.18653/v1/W19-3801

Citationsformater

@inproceedings{30d0e713d5b8401e8aabc80941dcb5fc,

title = "Gendered Ambiguous Pronoun (GAP) Shared Task at the Gender Bias in NLP Workshop 2019",

abstract = "The 1st ACL workshop on Gender Bias in Natural Language Processing included a shared task on gendered ambiguous pronoun (GAP) resolution. This task was based on the coreference challenge defined in Webster et al. (2018), designed to benchmark the ability of systems to resolve pronouns in real-world contexts in a gender-fair way. 263 teams competed via a Kaggle competition, with the winning system achieving logloss of 0.13667 and near gender parity. We review the approaches of eleven systems with accepted description papers, noting their effective use of BERT (Devlin et al., 2018), both via fine-tuning and for feature extraction, as well as ensembling.",

author = "Kellie Webster and Costa-juss{\`a}, {Marta R.} and Christian Hardmeier and Will Radford",

year = "2019",

month = aug,

day = "2",

doi = "10.18653/v1/W19-3801",

language = "English",

isbn = "978-1-950737-40-6",

booktitle = "Proceedings of the First Workshop on Gender Bias in Natural Language Processing",

}

Gendered Ambiguous Pronoun (GAP) Shared Task at the Gender Bias in NLP Workshop 2019. / Webster, Kellie; Costa-jussà, Marta R.; Hardmeier, Christian et al.
Proceedings of the First Workshop on Gender Bias in Natural Language Processing. 2019.

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Gendered Ambiguous Pronoun (GAP) Shared Task at the Gender Bias in NLP Workshop 2019

AU - Webster, Kellie

AU - Costa-jussà, Marta R.

AU - Hardmeier, Christian

AU - Radford, Will

PY - 2019/8/2

Y1 - 2019/8/2

N2 - The 1st ACL workshop on Gender Bias in Natural Language Processing included a shared task on gendered ambiguous pronoun (GAP) resolution. This task was based on the coreference challenge defined in Webster et al. (2018), designed to benchmark the ability of systems to resolve pronouns in real-world contexts in a gender-fair way. 263 teams competed via a Kaggle competition, with the winning system achieving logloss of 0.13667 and near gender parity. We review the approaches of eleven systems with accepted description papers, noting their effective use of BERT (Devlin et al., 2018), both via fine-tuning and for feature extraction, as well as ensembling.

AB - The 1st ACL workshop on Gender Bias in Natural Language Processing included a shared task on gendered ambiguous pronoun (GAP) resolution. This task was based on the coreference challenge defined in Webster et al. (2018), designed to benchmark the ability of systems to resolve pronouns in real-world contexts in a gender-fair way. 263 teams competed via a Kaggle competition, with the winning system achieving logloss of 0.13667 and near gender parity. We review the approaches of eleven systems with accepted description papers, noting their effective use of BERT (Devlin et al., 2018), both via fine-tuning and for feature extraction, as well as ensembling.

U2 - 10.18653/v1/W19-3801

DO - 10.18653/v1/W19-3801

M3 - Article in proceedings

SN - 978-1-950737-40-6

BT - Proceedings of the First Workshop on Gender Bias in Natural Language Processing

ER -

Gendered Ambiguous Pronoun (GAP) Shared Task at the Gender Bias in NLP Workshop 2019

Abstract

Adgang til dokumentet

Fingeraftryk

Citationsformater