<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ormazabal-etal-2023-comblm">
    <titleInfo>
        <title>CombLM: Adapting Black-Box Language Models through Small Fine-Tuned Models</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Aitor</namePart>
        <namePart type="family">Ormazabal</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Mikel</namePart>
        <namePart type="family">Artetxe</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Eneko</namePart>
        <namePart type="family">Agirre</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Houda</namePart>
            <namePart type="family">Bouamor</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Juan</namePart>
            <namePart type="family">Pino</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kalika</namePart>
            <namePart type="family">Bali</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Singapore</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Methods for adapting language models (LMs) to new tasks and domains have traditionally assumed white-box access to the model, and work by modifying its parameters. However, this is incompatible with a recent trend in the field, where the highest quality models are only available as black-boxes through inference APIs. Even when the model weights are available, the computational cost of fine-tuning large LMs can be prohibitive for most practitioners. In this work, we present a lightweight method for adapting large LMs to new domains and tasks, assuming no access to their weights or intermediate activations. Our approach fine-tunes a small white-box LM and combines it with the large black-box LM at the probability level through a small network, learned on a small validation set. We validate our approach by adapting a large LM (OPT-30B) to several domains and a downstream task (machine translation), observing improved performance in all cases, of up to 9%, while using a domain expert 23x smaller.</abstract>
    <identifier type="citekey">ormazabal-etal-2023-comblm</identifier>
    <identifier type="doi">10.18653/v1/2023.emnlp-main.180</identifier>
    <location>
        <url>https://aclanthology.org/2023.emnlp-main.180</url>
    </location>
    <part>
        <date>2023-12</date>
        <extent unit="page">
            <start>2961</start>
            <end>2974</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T CombLM: Adapting Black-Box Language Models through Small Fine-Tuned Models
%A Ormazabal, Aitor
%A Artetxe, Mikel
%A Agirre, Eneko
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F ormazabal-etal-2023-comblm
%X Methods for adapting language models (LMs) to new tasks and domains have traditionally assumed white-box access to the model, and work by modifying its parameters. However, this is incompatible with a recent trend in the field, where the highest quality models are only available as black-boxes through inference APIs. Even when the model weights are available, the computational cost of fine-tuning large LMs can be prohibitive for most practitioners. In this work, we present a lightweight method for adapting large LMs to new domains and tasks, assuming no access to their weights or intermediate activations. Our approach fine-tunes a small white-box LM and combines it with the large black-box LM at the probability level through a small network, learned on a small validation set. We validate our approach by adapting a large LM (OPT-30B) to several domains and a downstream task (machine translation), observing improved performance in all cases, of up to 9%, while using a domain expert 23x smaller.
%R 10.18653/v1/2023.emnlp-main.180
%U https://aclanthology.org/2023.emnlp-main.180
%U https://doi.org/10.18653/v1/2023.emnlp-main.180
%P 2961-2974

Markdown (Informal)

[CombLM: Adapting Black-Box Language Models through Small Fine-Tuned Models](https://aclanthology.org/2023.emnlp-main.180) (Ormazabal et al., EMNLP 2023)

• CombLM: Adapting Black-Box Language Models through Small Fine-Tuned Models (Ormazabal et al., EMNLP 2023)

ACL

• Aitor Ormazabal, Mikel Artetxe, and Eneko Agirre. 2023. CombLM: Adapting Black-Box Language Models through Small Fine-Tuned Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 2961–2974, Singapore. Association for Computational Linguistics.