@inproceedings{mueller-etal-2024-multi,
title = "Multi-Task Transfer Matters During Instruction-Tuning",
author = "Mueller, David and
Dredze, Mark and
Andrews, Nicholas",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.883",
doi = "10.18653/v1/2024.findings-acl.883",
pages = "14880--14891",
abstract = "Instruction-tuning trains a language model on hundreds of tasks jointly to improve a model{'}s ability to learn in-context;however, the mechanisms that drive in-context learning are poorly understood and, as a result, the role of instruction-tuning on in-context generalization is poorly understood as well.In this work, we study the impact of instruction-tuning on multi-task transfer: how well a model{'}s parameters adapt to an unseen task via fine-tuning.We find that instruction-tuning negatively impacts a model{'}s transfer to unseen tasks, and that model transfer and in-context generalization are highly correlated, suggesting that this catastrophic forgetting may impact in-context learning.We study methods to improve model transfer, finding that multi-task training{---}how well the training tasks are optimized{---}can significantly impact ICL generalization; additionally, we find that continual training on unsupervised pre-training data can mitigate forgetting and improve ICL generalization as well.Finally, we demonstrate that, early into training, the impact of instruction-tuning on model transfer to tasks impacts in-context generalization on that task.Overall, we provide significant evidence that multi-task transfer is deeply connected to a model{'}s ability to learn a task in-context.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mueller-etal-2024-multi">
<titleInfo>
<title>Multi-Task Transfer Matters During Instruction-Tuning</title>
</titleInfo>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Mueller</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mark</namePart>
<namePart type="family">Dredze</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicholas</namePart>
<namePart type="family">Andrews</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Instruction-tuning trains a language model on hundreds of tasks jointly to improve a model’s ability to learn in-context;however, the mechanisms that drive in-context learning are poorly understood and, as a result, the role of instruction-tuning on in-context generalization is poorly understood as well.In this work, we study the impact of instruction-tuning on multi-task transfer: how well a model’s parameters adapt to an unseen task via fine-tuning.We find that instruction-tuning negatively impacts a model’s transfer to unseen tasks, and that model transfer and in-context generalization are highly correlated, suggesting that this catastrophic forgetting may impact in-context learning.We study methods to improve model transfer, finding that multi-task training—how well the training tasks are optimized—can significantly impact ICL generalization; additionally, we find that continual training on unsupervised pre-training data can mitigate forgetting and improve ICL generalization as well.Finally, we demonstrate that, early into training, the impact of instruction-tuning on model transfer to tasks impacts in-context generalization on that task.Overall, we provide significant evidence that multi-task transfer is deeply connected to a model’s ability to learn a task in-context.</abstract>
<identifier type="citekey">mueller-etal-2024-multi</identifier>
<identifier type="doi">10.18653/v1/2024.findings-acl.883</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.883</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>14880</start>
<end>14891</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Multi-Task Transfer Matters During Instruction-Tuning
%A Mueller, David
%A Dredze, Mark
%A Andrews, Nicholas
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics: ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F mueller-etal-2024-multi
%X Instruction-tuning trains a language model on hundreds of tasks jointly to improve a model’s ability to learn in-context;however, the mechanisms that drive in-context learning are poorly understood and, as a result, the role of instruction-tuning on in-context generalization is poorly understood as well.In this work, we study the impact of instruction-tuning on multi-task transfer: how well a model’s parameters adapt to an unseen task via fine-tuning.We find that instruction-tuning negatively impacts a model’s transfer to unseen tasks, and that model transfer and in-context generalization are highly correlated, suggesting that this catastrophic forgetting may impact in-context learning.We study methods to improve model transfer, finding that multi-task training—how well the training tasks are optimized—can significantly impact ICL generalization; additionally, we find that continual training on unsupervised pre-training data can mitigate forgetting and improve ICL generalization as well.Finally, we demonstrate that, early into training, the impact of instruction-tuning on model transfer to tasks impacts in-context generalization on that task.Overall, we provide significant evidence that multi-task transfer is deeply connected to a model’s ability to learn a task in-context.
%R 10.18653/v1/2024.findings-acl.883
%U https://aclanthology.org/2024.findings-acl.883
%U https://doi.org/10.18653/v1/2024.findings-acl.883
%P 14880-14891
Markdown (Informal)
[Multi-Task Transfer Matters During Instruction-Tuning](https://aclanthology.org/2024.findings-acl.883) (Mueller et al., Findings 2024)
ACL