<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sharma-etal-2024-linear">
    <titleInfo>
        <title>Linear Layer Extrapolation for Fine-Grained Emotion Classification</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Mayukh</namePart>
        <namePart type="family">Sharma</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sean</namePart>
        <namePart type="family">O’Brien</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Julian</namePart>
        <namePart type="family">McAuley</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2024-11</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Yaser</namePart>
            <namePart type="family">Al-Onaizan</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Mohit</namePart>
            <namePart type="family">Bansal</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Yun-Nung</namePart>
            <namePart type="family">Chen</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Miami, Florida, USA</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Certain abilities of Transformer-based language models consistently emerge in their later layers. Previous research has leveraged this phenomenon to improve factual accuracy through self-contrast, penalizing early-exit predictions based on the premise that later-layer updates are more factually reliable than earlier-layer associations. We observe a similar pattern for fine-grained emotion classification in text, demonstrating that self-contrast can enhance encoder-based text classifiers. Additionally, we reinterpret self-contrast as a form of linear extrapolation, which motivates a refined approach that dynamically adjusts the contrastive strength based on the selected intermediate layer. Experiments across multiple models and emotion classification datasets show that our method outperforms standard classification techniques in fine-grained emotion classification tasks.</abstract>
    <identifier type="citekey">sharma-etal-2024-linear</identifier>
    <identifier type="doi">10.18653/v1/2024.emnlp-main.1161</identifier>
    <location>
        <url>https://aclanthology.org/2024.emnlp-main.1161</url>
    </location>
    <part>
        <date>2024-11</date>
        <extent unit="page">
            <start>20880</start>
            <end>20888</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Linear Layer Extrapolation for Fine-Grained Emotion Classification
%A Sharma, Mayukh
%A O’Brien, Sean
%A McAuley, Julian
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F sharma-etal-2024-linear
%X Certain abilities of Transformer-based language models consistently emerge in their later layers. Previous research has leveraged this phenomenon to improve factual accuracy through self-contrast, penalizing early-exit predictions based on the premise that later-layer updates are more factually reliable than earlier-layer associations. We observe a similar pattern for fine-grained emotion classification in text, demonstrating that self-contrast can enhance encoder-based text classifiers. Additionally, we reinterpret self-contrast as a form of linear extrapolation, which motivates a refined approach that dynamically adjusts the contrastive strength based on the selected intermediate layer. Experiments across multiple models and emotion classification datasets show that our method outperforms standard classification techniques in fine-grained emotion classification tasks.
%R 10.18653/v1/2024.emnlp-main.1161
%U https://aclanthology.org/2024.emnlp-main.1161
%U https://doi.org/10.18653/v1/2024.emnlp-main.1161
%P 20880-20888

Markdown (Informal)

[Linear Layer Extrapolation for Fine-Grained Emotion Classification](https://aclanthology.org/2024.emnlp-main.1161) (Sharma et al., EMNLP 2024)

• Linear Layer Extrapolation for Fine-Grained Emotion Classification (Sharma et al., EMNLP 2024)

ACL

• Mayukh Sharma, Sean O’Brien, and Julian McAuley. 2024. Linear Layer Extrapolation for Fine-Grained Emotion Classification. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 20880–20888, Miami, Florida, USA. Association for Computational Linguistics.