@inproceedings{stephan-etal-2022-sepll,
title = "{S}ep{LL}: Separating Latent Class Labels from Weak Supervision Noise",
author = "Stephan, Andreas and
Kougia, Vasiliki and
Roth, Benjamin",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.288/",
doi = "10.18653/v1/2022.findings-emnlp.288",
pages = "3918--3929",
abstract = "In the weakly supervised learning paradigm, labeling functions automatically assign heuristic, often noisy, labels to data samples. In this work, we provide a method for learning from weak labels by separating two types of complementary information associated with the labeling functions: information related to the target label and information specific to one labeling function only. Both types of information are reflected to different degrees by all labeled instances. In contrast to previous works that aimed at correcting or removing wrongly labeled instances, we learn a branched deep model that uses all data as-is, but splits the labeling function information in the latent space. Specifically, we propose the end-to-end model SepLL which extends a transformer classifier by introducing a latent space for labeling function specific and task-specific information. The learning signal is only given by the labeling functions matches, no pre-processing or label model is required for our method. Notably, the task prediction is made from the latent layer without any direct task signal. Experiments on Wrench text classification tasks show that our model is competitive with the state-of-the-art, and yields a new best average performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="stephan-etal-2022-sepll">
<titleInfo>
<title>SepLL: Separating Latent Class Labels from Weak Supervision Noise</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Stephan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vasiliki</namePart>
<namePart type="family">Kougia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Roth</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In the weakly supervised learning paradigm, labeling functions automatically assign heuristic, often noisy, labels to data samples. In this work, we provide a method for learning from weak labels by separating two types of complementary information associated with the labeling functions: information related to the target label and information specific to one labeling function only. Both types of information are reflected to different degrees by all labeled instances. In contrast to previous works that aimed at correcting or removing wrongly labeled instances, we learn a branched deep model that uses all data as-is, but splits the labeling function information in the latent space. Specifically, we propose the end-to-end model SepLL which extends a transformer classifier by introducing a latent space for labeling function specific and task-specific information. The learning signal is only given by the labeling functions matches, no pre-processing or label model is required for our method. Notably, the task prediction is made from the latent layer without any direct task signal. Experiments on Wrench text classification tasks show that our model is competitive with the state-of-the-art, and yields a new best average performance.</abstract>
<identifier type="citekey">stephan-etal-2022-sepll</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.288</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.288/</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>3918</start>
<end>3929</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SepLL: Separating Latent Class Labels from Weak Supervision Noise
%A Stephan, Andreas
%A Kougia, Vasiliki
%A Roth, Benjamin
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F stephan-etal-2022-sepll
%X In the weakly supervised learning paradigm, labeling functions automatically assign heuristic, often noisy, labels to data samples. In this work, we provide a method for learning from weak labels by separating two types of complementary information associated with the labeling functions: information related to the target label and information specific to one labeling function only. Both types of information are reflected to different degrees by all labeled instances. In contrast to previous works that aimed at correcting or removing wrongly labeled instances, we learn a branched deep model that uses all data as-is, but splits the labeling function information in the latent space. Specifically, we propose the end-to-end model SepLL which extends a transformer classifier by introducing a latent space for labeling function specific and task-specific information. The learning signal is only given by the labeling functions matches, no pre-processing or label model is required for our method. Notably, the task prediction is made from the latent layer without any direct task signal. Experiments on Wrench text classification tasks show that our model is competitive with the state-of-the-art, and yields a new best average performance.
%R 10.18653/v1/2022.findings-emnlp.288
%U https://aclanthology.org/2022.findings-emnlp.288/
%U https://doi.org/10.18653/v1/2022.findings-emnlp.288
%P 3918-3929
Markdown (Informal)
[SepLL: Separating Latent Class Labels from Weak Supervision Noise](https://aclanthology.org/2022.findings-emnlp.288/) (Stephan et al., Findings 2022)
ACL