@inproceedings{clark-etal-2022-meta,
title = "Meta-Learning Fast Weight Language Models",
author = "Clark, Kevin and
Guu, Kelvin and
Chang, Ming-Wei and
Pasupat, Panupong and
Hinton, Geoffrey and
Norouzi, Mohammad",
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.661",
doi = "10.18653/v1/2022.emnlp-main.661",
pages = "9751--9757",
abstract = "Dynamic evaluation of language models (LMs) adapts model parameters at test time using gradient information from previous tokens and substantially improves LM performance. However, it requires over 3x more compute than standard inference. We present Fast Weight Layers (FWLs), a neural component that provides the benefits of dynamic evaluation much more efficiently by expressing gradient updates as linear attention. A key improvement over dynamic evaluation is that FWLs can also be applied at training time, so the model learns to make good use of gradient updates. FWLs can easily be added on top of existing transformer models, require relatively little extra compute or memory to run, and significantly improve language modeling perplexity.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="clark-etal-2022-meta">
<titleInfo>
<title>Meta-Learning Fast Weight Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kevin</namePart>
<namePart type="family">Clark</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kelvin</namePart>
<namePart type="family">Guu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming-Wei</namePart>
<namePart type="family">Chang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Panupong</namePart>
<namePart type="family">Pasupat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Geoffrey</namePart>
<namePart type="family">Hinton</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="family">Norouzi</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>Dynamic evaluation of language models (LMs) adapts model parameters at test time using gradient information from previous tokens and substantially improves LM performance. However, it requires over 3x more compute than standard inference. We present Fast Weight Layers (FWLs), a neural component that provides the benefits of dynamic evaluation much more efficiently by expressing gradient updates as linear attention. A key improvement over dynamic evaluation is that FWLs can also be applied at training time, so the model learns to make good use of gradient updates. FWLs can easily be added on top of existing transformer models, require relatively little extra compute or memory to run, and significantly improve language modeling perplexity.</abstract>
<identifier type="citekey">clark-etal-2022-meta</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.661</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.661</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>9751</start>
<end>9757</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Meta-Learning Fast Weight Language Models
%A Clark, Kevin
%A Guu, Kelvin
%A Chang, Ming-Wei
%A Pasupat, Panupong
%A Hinton, Geoffrey
%A Norouzi, Mohammad
%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 clark-etal-2022-meta
%X Dynamic evaluation of language models (LMs) adapts model parameters at test time using gradient information from previous tokens and substantially improves LM performance. However, it requires over 3x more compute than standard inference. We present Fast Weight Layers (FWLs), a neural component that provides the benefits of dynamic evaluation much more efficiently by expressing gradient updates as linear attention. A key improvement over dynamic evaluation is that FWLs can also be applied at training time, so the model learns to make good use of gradient updates. FWLs can easily be added on top of existing transformer models, require relatively little extra compute or memory to run, and significantly improve language modeling perplexity.
%R 10.18653/v1/2022.emnlp-main.661
%U https://aclanthology.org/2022.emnlp-main.661
%U https://doi.org/10.18653/v1/2022.emnlp-main.661
%P 9751-9757
Markdown (Informal)
[Meta-Learning Fast Weight Language Models](https://aclanthology.org/2022.emnlp-main.661) (Clark et al., EMNLP 2022)
ACL
- Kevin Clark, Kelvin Guu, Ming-Wei Chang, Panupong Pasupat, Geoffrey Hinton, and Mohammad Norouzi. 2022. Meta-Learning Fast Weight Language Models. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 9751–9757, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.