@inproceedings{tian-etal-2023-useclean,
title = "{U}se{C}lean: learning from complex noisy labels in named entity recognition",
author = "Tian, Jinjin and
Zhou, Kun and
Wang, Meiguo and
Zhang, Yu and
Yao, Benjamin and
Liu, Xiaohu and
Guo, Chenlei",
editor = "Breitholtz, Ellen and
Lappin, Shalom and
Loaiciga, Sharid and
Ilinykh, Nikolai and
Dobnik, Simon",
booktitle = "Proceedings of the 2023 CLASP Conference on Learning with Small Data (LSD)",
month = sep,
year = "2023",
address = "Gothenburg, Sweden",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.clasp-1.14",
pages = "120--130",
abstract = "We investigate and refine denoising methods for NER task on data that potentially contains extremely noisy labels from multi-sources. In this paper, we first summarized all possible noise types and noise generation schemes, based on which we built a thorough evaluation system. We then pinpoint the bottleneck of current state-of-art denoising methods using our evaluation system. Correspondingly, we propose several refinements, including using a two-stage framework to avoid error accumulation; a novel confidence score utilizing minimal clean supervision to increase predictive power; an automatic cutoff fitting to save extensive hyper-parameter tuning; a warm started weighted partial CRF to better learn on the noisy tokens. Additionally, we propose to use adaptive sampling to further boost the performance in long-tailed entity settings. Our method improves F1 score by on average at least 5 10{\%} over current state-of-art across extensive experiments.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tian-etal-2023-useclean">
<titleInfo>
<title>UseClean: learning from complex noisy labels in named entity recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jinjin</namePart>
<namePart type="family">Tian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kun</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meiguo</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Yao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaohu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenlei</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2023 CLASP Conference on Learning with Small Data (LSD)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ellen</namePart>
<namePart type="family">Breitholtz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shalom</namePart>
<namePart type="family">Lappin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sharid</namePart>
<namePart type="family">Loaiciga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolai</namePart>
<namePart type="family">Ilinykh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Dobnik</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Gothenburg, Sweden</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We investigate and refine denoising methods for NER task on data that potentially contains extremely noisy labels from multi-sources. In this paper, we first summarized all possible noise types and noise generation schemes, based on which we built a thorough evaluation system. We then pinpoint the bottleneck of current state-of-art denoising methods using our evaluation system. Correspondingly, we propose several refinements, including using a two-stage framework to avoid error accumulation; a novel confidence score utilizing minimal clean supervision to increase predictive power; an automatic cutoff fitting to save extensive hyper-parameter tuning; a warm started weighted partial CRF to better learn on the noisy tokens. Additionally, we propose to use adaptive sampling to further boost the performance in long-tailed entity settings. Our method improves F1 score by on average at least 5 10% over current state-of-art across extensive experiments.</abstract>
<identifier type="citekey">tian-etal-2023-useclean</identifier>
<location>
<url>https://aclanthology.org/2023.clasp-1.14</url>
</location>
<part>
<date>2023-09</date>
<extent unit="page">
<start>120</start>
<end>130</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T UseClean: learning from complex noisy labels in named entity recognition
%A Tian, Jinjin
%A Zhou, Kun
%A Wang, Meiguo
%A Zhang, Yu
%A Yao, Benjamin
%A Liu, Xiaohu
%A Guo, Chenlei
%Y Breitholtz, Ellen
%Y Lappin, Shalom
%Y Loaiciga, Sharid
%Y Ilinykh, Nikolai
%Y Dobnik, Simon
%S Proceedings of the 2023 CLASP Conference on Learning with Small Data (LSD)
%D 2023
%8 September
%I Association for Computational Linguistics
%C Gothenburg, Sweden
%F tian-etal-2023-useclean
%X We investigate and refine denoising methods for NER task on data that potentially contains extremely noisy labels from multi-sources. In this paper, we first summarized all possible noise types and noise generation schemes, based on which we built a thorough evaluation system. We then pinpoint the bottleneck of current state-of-art denoising methods using our evaluation system. Correspondingly, we propose several refinements, including using a two-stage framework to avoid error accumulation; a novel confidence score utilizing minimal clean supervision to increase predictive power; an automatic cutoff fitting to save extensive hyper-parameter tuning; a warm started weighted partial CRF to better learn on the noisy tokens. Additionally, we propose to use adaptive sampling to further boost the performance in long-tailed entity settings. Our method improves F1 score by on average at least 5 10% over current state-of-art across extensive experiments.
%U https://aclanthology.org/2023.clasp-1.14
%P 120-130
Markdown (Informal)
[UseClean: learning from complex noisy labels in named entity recognition](https://aclanthology.org/2023.clasp-1.14) (Tian et al., CLASP 2023)
ACL