<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kambhatla-etal-2023-quantifying">
    <titleInfo>
        <title>Quantifying Train-Evaluation Overlap with Nearest Neighbors</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Gauri</namePart>
        <namePart type="family">Kambhatla</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Thuy</namePart>
        <namePart type="family">Nguyen</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Eunsol</namePart>
        <namePart type="family">Choi</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>Characterizing benchmark datasets is crucial to interpreting model performance. In this work, we study train-evaluation overlap as a measure of an individual dataset’s adequacy to evaluate model generalization over a wide range of datasets. We quantify the overlap with a simple novel metric based on a nearest neighbors approach between the training and evaluation sets. We identify nearest training examples for each evaluation example by mapping instances with generic and task-specific embedding methods. Our study on eleven classification and extractive QA tasks reveals a wide range of train-evaluation overlap, and we show that the data collection method of the dataset and the difficulty of the task may play a role in the amount of overlap. Lastly, we use our nearest neighbor analysis to identify challenging or potentially mislabeled examples. Our analysis quantifies train-evaluation overlap, providing insights for constructing datasets to study generalization.</abstract>
    <identifier type="citekey">kambhatla-etal-2023-quantifying</identifier>
    <identifier type="doi">10.18653/v1/2023.findings-acl.183</identifier>
    <location>
        <url>https://aclanthology.org/2023.findings-acl.183</url>
    </location>
    <part>
        <date>2023-07</date>
        <extent unit="page">
            <start>2905</start>
            <end>2920</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Quantifying Train-Evaluation Overlap with Nearest Neighbors
%A Kambhatla, Gauri
%A Nguyen, Thuy
%A Choi, Eunsol
%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 kambhatla-etal-2023-quantifying
%X Characterizing benchmark datasets is crucial to interpreting model performance. In this work, we study train-evaluation overlap as a measure of an individual dataset’s adequacy to evaluate model generalization over a wide range of datasets. We quantify the overlap with a simple novel metric based on a nearest neighbors approach between the training and evaluation sets. We identify nearest training examples for each evaluation example by mapping instances with generic and task-specific embedding methods. Our study on eleven classification and extractive QA tasks reveals a wide range of train-evaluation overlap, and we show that the data collection method of the dataset and the difficulty of the task may play a role in the amount of overlap. Lastly, we use our nearest neighbor analysis to identify challenging or potentially mislabeled examples. Our analysis quantifies train-evaluation overlap, providing insights for constructing datasets to study generalization.
%R 10.18653/v1/2023.findings-acl.183
%U https://aclanthology.org/2023.findings-acl.183
%U https://doi.org/10.18653/v1/2023.findings-acl.183
%P 2905-2920

Markdown (Informal)

[Quantifying Train-Evaluation Overlap with Nearest Neighbors](https://aclanthology.org/2023.findings-acl.183) (Kambhatla et al., Findings 2023)

• Quantifying Train-Evaluation Overlap with Nearest Neighbors (Kambhatla et al., Findings 2023)

ACL

• Gauri Kambhatla, Thuy Nguyen, and Eunsol Choi. 2023. Quantifying Train-Evaluation Overlap with Nearest Neighbors. In Findings of the Association for Computational Linguistics: ACL 2023, pages 2905–2920, Toronto, Canada. Association for Computational Linguistics.