@inproceedings{weerasooriya-etal-2023-disagreement,
title = "Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with {D}is{C}o",
author = "Weerasooriya, Tharindu Cyril and
Ororbia, Alexander and
Bhensadadia, Raj and
KhudaBukhsh, Ashiqur and
Homan, Christopher",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.287",
doi = "10.18653/v1/2023.findings-acl.287",
pages = "4679--4695",
abstract = "Annotator disagreement is common whenever human judgment is needed for supervised learning. It is conventional to assume that one label per item represents ground truth. However, this obscures minority opinions, if present. We regard {``}ground truth{''} as the distribution of all labels that a population of annotators could produce, if asked (and of which we only have a small sample). We next introduce DisCo (Distribution from Context), a simple neural model that learns to predict this distribution. The model takes annotator-item pairs, rather than items alone, as input, and performs inference by aggregating over all annotators. Despite its simplicity, our experiments show that, on six benchmark datasets, our model is competitive with, and frequently outperforms, other, more complex models that either do not model specific annotators or were not designed for label distribution learning.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="weerasooriya-etal-2023-disagreement">
<titleInfo>
<title>Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with DisCo</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tharindu</namePart>
<namePart type="given">Cyril</namePart>
<namePart type="family">Weerasooriya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Ororbia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raj</namePart>
<namePart type="family">Bhensadadia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ashiqur</namePart>
<namePart type="family">KhudaBukhsh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Homan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Annotator disagreement is common whenever human judgment is needed for supervised learning. It is conventional to assume that one label per item represents ground truth. However, this obscures minority opinions, if present. We regard “ground truth” as the distribution of all labels that a population of annotators could produce, if asked (and of which we only have a small sample). We next introduce DisCo (Distribution from Context), a simple neural model that learns to predict this distribution. The model takes annotator-item pairs, rather than items alone, as input, and performs inference by aggregating over all annotators. Despite its simplicity, our experiments show that, on six benchmark datasets, our model is competitive with, and frequently outperforms, other, more complex models that either do not model specific annotators or were not designed for label distribution learning.</abstract>
<identifier type="citekey">weerasooriya-etal-2023-disagreement</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.287</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.287</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>4679</start>
<end>4695</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with DisCo
%A Weerasooriya, Tharindu Cyril
%A Ororbia, Alexander
%A Bhensadadia, Raj
%A KhudaBukhsh, Ashiqur
%A Homan, Christopher
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F weerasooriya-etal-2023-disagreement
%X Annotator disagreement is common whenever human judgment is needed for supervised learning. It is conventional to assume that one label per item represents ground truth. However, this obscures minority opinions, if present. We regard “ground truth” as the distribution of all labels that a population of annotators could produce, if asked (and of which we only have a small sample). We next introduce DisCo (Distribution from Context), a simple neural model that learns to predict this distribution. The model takes annotator-item pairs, rather than items alone, as input, and performs inference by aggregating over all annotators. Despite its simplicity, our experiments show that, on six benchmark datasets, our model is competitive with, and frequently outperforms, other, more complex models that either do not model specific annotators or were not designed for label distribution learning.
%R 10.18653/v1/2023.findings-acl.287
%U https://aclanthology.org/2023.findings-acl.287
%U https://doi.org/10.18653/v1/2023.findings-acl.287
%P 4679-4695
Markdown (Informal)
[Disagreement Matters: Preserving Label Diversity by Jointly Modeling Item and Annotator Label Distributions with DisCo](https://aclanthology.org/2023.findings-acl.287) (Weerasooriya et al., Findings 2023)
ACL