<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="dasgupta-etal-2023-cost">
    <titleInfo>
        <title>Cost-effective Distillation of Large Language Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Sayantan</namePart>
        <namePart type="family">Dasgupta</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Trevor</namePart>
        <namePart type="family">Cohn</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Timothy</namePart>
        <namePart type="family">Baldwin</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>Knowledge distillation (KD) involves training a small “student” model to replicate the strong performance of a high-capacity “teacher” model, enabling efficient deployment in resource-constrained settings. Top-performing methods tend to be task- or architecture-specific and lack generalizability. Several existing approaches require pretraining of the teacher on task-specific datasets, which can be costly for large and unstable for small datasets. Here we propose an approach for improving KD through a novel distillation loss agnostic to the task and model architecture. We successfully apply our method to the distillation of the BERT-base and achieve highly competitive results from the distilled student across a range of GLUE tasks, especially for tasks with smaller datasets.</abstract>
    <identifier type="citekey">dasgupta-etal-2023-cost</identifier>
    <identifier type="doi">10.18653/v1/2023.findings-acl.463</identifier>
    <location>
        <url>https://aclanthology.org/2023.findings-acl.463/</url>
    </location>
    <part>
        <date>2023-07</date>
        <extent unit="page">
            <start>7346</start>
            <end>7354</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Cost-effective Distillation of Large Language Models
%A Dasgupta, Sayantan
%A Cohn, Trevor
%A Baldwin, Timothy
%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 dasgupta-etal-2023-cost
%X Knowledge distillation (KD) involves training a small “student” model to replicate the strong performance of a high-capacity “teacher” model, enabling efficient deployment in resource-constrained settings. Top-performing methods tend to be task- or architecture-specific and lack generalizability. Several existing approaches require pretraining of the teacher on task-specific datasets, which can be costly for large and unstable for small datasets. Here we propose an approach for improving KD through a novel distillation loss agnostic to the task and model architecture. We successfully apply our method to the distillation of the BERT-base and achieve highly competitive results from the distilled student across a range of GLUE tasks, especially for tasks with smaller datasets.
%R 10.18653/v1/2023.findings-acl.463
%U https://aclanthology.org/2023.findings-acl.463/
%U https://doi.org/10.18653/v1/2023.findings-acl.463
%P 7346-7354

Markdown (Informal)

[Cost-effective Distillation of Large Language Models](https://aclanthology.org/2023.findings-acl.463/) (Dasgupta et al., Findings 2023)

• Cost-effective Distillation of Large Language Models (Dasgupta et al., Findings 2023)

ACL

• Sayantan Dasgupta, Trevor Cohn, and Timothy Baldwin. 2023. Cost-effective Distillation of Large Language Models. In Findings of the Association for Computational Linguistics: ACL 2023, pages 7346–7354, Toronto, Canada. Association for Computational Linguistics.