@inproceedings{white-etal-2022-mixed,
title = "Mixed-effects transformers for hierarchical adaptation",
author = "White, Julia and
Goodman, Noah and
Hawkins, Robert",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.261",
doi = "10.18653/v1/2022.emnlp-main.261",
pages = "3944--3954",
abstract = "Language differs dramatically from context to context. To some degree, large language models like GPT-3 account for such variation by conditioning on strings of initial input text, or prompts. However, prompting can be ineffective when contexts are sparse, out-of-sample, or extra-textual. In this paper, we introduce the mixed-effects transformer (MET), a novel approach for learning hierarchically-structured prefixes{---} lightweight modules prepended to an input sequence{---} to account for structured variation in language use. Specifically, we show how the popular class of mixed-effects regression models may be extended to transformer-based architectures using a regularized prefix-tuning procedure with dropout. We evaluate this approach on several domain-adaptation benchmarks, finding that it learns contextual variation from minimal data while generalizing well to unseen contexts.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="white-etal-2022-mixed">
<titleInfo>
<title>Mixed-effects transformers for hierarchical adaptation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Julia</namePart>
<namePart type="family">White</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Noah</namePart>
<namePart type="family">Goodman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Robert</namePart>
<namePart type="family">Hawkins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Language differs dramatically from context to context. To some degree, large language models like GPT-3 account for such variation by conditioning on strings of initial input text, or prompts. However, prompting can be ineffective when contexts are sparse, out-of-sample, or extra-textual. In this paper, we introduce the mixed-effects transformer (MET), a novel approach for learning hierarchically-structured prefixes— lightweight modules prepended to an input sequence— to account for structured variation in language use. Specifically, we show how the popular class of mixed-effects regression models may be extended to transformer-based architectures using a regularized prefix-tuning procedure with dropout. We evaluate this approach on several domain-adaptation benchmarks, finding that it learns contextual variation from minimal data while generalizing well to unseen contexts.</abstract>
<identifier type="citekey">white-etal-2022-mixed</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.261</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.261</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>3944</start>
<end>3954</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Mixed-effects transformers for hierarchical adaptation
%A White, Julia
%A Goodman, Noah
%A Hawkins, Robert
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F white-etal-2022-mixed
%X Language differs dramatically from context to context. To some degree, large language models like GPT-3 account for such variation by conditioning on strings of initial input text, or prompts. However, prompting can be ineffective when contexts are sparse, out-of-sample, or extra-textual. In this paper, we introduce the mixed-effects transformer (MET), a novel approach for learning hierarchically-structured prefixes— lightweight modules prepended to an input sequence— to account for structured variation in language use. Specifically, we show how the popular class of mixed-effects regression models may be extended to transformer-based architectures using a regularized prefix-tuning procedure with dropout. We evaluate this approach on several domain-adaptation benchmarks, finding that it learns contextual variation from minimal data while generalizing well to unseen contexts.
%R 10.18653/v1/2022.emnlp-main.261
%U https://aclanthology.org/2022.emnlp-main.261
%U https://doi.org/10.18653/v1/2022.emnlp-main.261
%P 3944-3954
Markdown (Informal)
[Mixed-effects transformers for hierarchical adaptation](https://aclanthology.org/2022.emnlp-main.261) (White et al., EMNLP 2022)
ACL
- Julia White, Noah Goodman, and Robert Hawkins. 2022. Mixed-effects transformers for hierarchical adaptation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 3944–3954, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.