@inproceedings{liang-etal-2023-prompts,
title = "Prompts Can Play Lottery Tickets Well: Achieving Lifelong Information Extraction via Lottery Prompt Tuning",
author = "Liang, Zujie and
Wei, Feng and
Jie, Yin and
Qian, Yuxi and
Hao, Zhenghong and
Han, Bing",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.16",
doi = "10.18653/v1/2023.acl-long.16",
pages = "277--292",
abstract = "Thanks to the recent success of Pre-trained Language Models (PLMs), it has become a promising research direction to develop a universal model (UIE) that can solve all typical information extraction tasks within one generative framework. Nonetheless, in real-world scenarios of UIE applications, new data of different IE tasks and domains usually come in a stream over time. A desirable UIE system should be capable of continually learning new tasks without forgetting old ones, thereby allowing knowledge and functionalities expansion without re-training the whole system. In this paper, we study the UIE system under a more challenging yet practical scenario, i.e., {``}lifelong learning{''} settings, to evaluate its abilities in three aspects, including knowledge sharing and expansion, catastrophic forgetting prevention, and rapid generalization on few-shot and unseen tasks. To achieve these three goals, we present a novel parameter- and deployment-efficient prompt tuning method namely Lottery Prompt Tuning (LPT).LPT freezes the PLM{'}s parameters and sequentially learns compact pruned prompt vectors for each task leveraging a binary prompt mask, while keeping the prompt parameters selected by the previous tasks insusceptible. Furthermore, we use a simple yet effective method to perform mask selection and show the powerful transferability of Lottery Prompts to novel tasks. Extensive experiments demonstrate that LPT consistently sets state-of-the-art performance on multiple lifelong learning settings of UIE, including task-incremental setting on seen tasks, few-shot adaptation, and zero-shot generalization on novel tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liang-etal-2023-prompts">
<titleInfo>
<title>Prompts Can Play Lottery Tickets Well: Achieving Lifelong Information Extraction via Lottery Prompt Tuning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zujie</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Feng</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yin</namePart>
<namePart type="family">Jie</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxi</namePart>
<namePart type="family">Qian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhenghong</namePart>
<namePart type="family">Hao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bing</namePart>
<namePart type="family">Han</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>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</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>Thanks to the recent success of Pre-trained Language Models (PLMs), it has become a promising research direction to develop a universal model (UIE) that can solve all typical information extraction tasks within one generative framework. Nonetheless, in real-world scenarios of UIE applications, new data of different IE tasks and domains usually come in a stream over time. A desirable UIE system should be capable of continually learning new tasks without forgetting old ones, thereby allowing knowledge and functionalities expansion without re-training the whole system. In this paper, we study the UIE system under a more challenging yet practical scenario, i.e., “lifelong learning” settings, to evaluate its abilities in three aspects, including knowledge sharing and expansion, catastrophic forgetting prevention, and rapid generalization on few-shot and unseen tasks. To achieve these three goals, we present a novel parameter- and deployment-efficient prompt tuning method namely Lottery Prompt Tuning (LPT).LPT freezes the PLM’s parameters and sequentially learns compact pruned prompt vectors for each task leveraging a binary prompt mask, while keeping the prompt parameters selected by the previous tasks insusceptible. Furthermore, we use a simple yet effective method to perform mask selection and show the powerful transferability of Lottery Prompts to novel tasks. Extensive experiments demonstrate that LPT consistently sets state-of-the-art performance on multiple lifelong learning settings of UIE, including task-incremental setting on seen tasks, few-shot adaptation, and zero-shot generalization on novel tasks.</abstract>
<identifier type="citekey">liang-etal-2023-prompts</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.16</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.16</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>277</start>
<end>292</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Prompts Can Play Lottery Tickets Well: Achieving Lifelong Information Extraction via Lottery Prompt Tuning
%A Liang, Zujie
%A Wei, Feng
%A Jie, Yin
%A Qian, Yuxi
%A Hao, Zhenghong
%A Han, Bing
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F liang-etal-2023-prompts
%X Thanks to the recent success of Pre-trained Language Models (PLMs), it has become a promising research direction to develop a universal model (UIE) that can solve all typical information extraction tasks within one generative framework. Nonetheless, in real-world scenarios of UIE applications, new data of different IE tasks and domains usually come in a stream over time. A desirable UIE system should be capable of continually learning new tasks without forgetting old ones, thereby allowing knowledge and functionalities expansion without re-training the whole system. In this paper, we study the UIE system under a more challenging yet practical scenario, i.e., “lifelong learning” settings, to evaluate its abilities in three aspects, including knowledge sharing and expansion, catastrophic forgetting prevention, and rapid generalization on few-shot and unseen tasks. To achieve these three goals, we present a novel parameter- and deployment-efficient prompt tuning method namely Lottery Prompt Tuning (LPT).LPT freezes the PLM’s parameters and sequentially learns compact pruned prompt vectors for each task leveraging a binary prompt mask, while keeping the prompt parameters selected by the previous tasks insusceptible. Furthermore, we use a simple yet effective method to perform mask selection and show the powerful transferability of Lottery Prompts to novel tasks. Extensive experiments demonstrate that LPT consistently sets state-of-the-art performance on multiple lifelong learning settings of UIE, including task-incremental setting on seen tasks, few-shot adaptation, and zero-shot generalization on novel tasks.
%R 10.18653/v1/2023.acl-long.16
%U https://aclanthology.org/2023.acl-long.16
%U https://doi.org/10.18653/v1/2023.acl-long.16
%P 277-292
Markdown (Informal)
[Prompts Can Play Lottery Tickets Well: Achieving Lifelong Information Extraction via Lottery Prompt Tuning](https://aclanthology.org/2023.acl-long.16) (Liang et al., ACL 2023)
ACL