@inproceedings{zheng-etal-2024-learn,
title = "Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models",
author = "Zheng, Junhao and
Qiu, Shengjie and
Ma, Qianli",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.luhme-long.794/",
doi = "10.18653/v1/2024.acl-long.794",
pages = "14848--14877",
abstract = "Incremental Learning (IL) has been a long-standing problem in both vision and Natural Language Processing (NLP) communities.In recent years, as Pre-trained Language Models (PLMs) have achieved remarkable progress in various NLP downstream tasks, utilizing PLMs as backbones has become a common practice in recent research of IL in NLP.Most assume that catastrophic forgetting is the biggest obstacle to achieving superior IL performance and propose various techniques to overcome this issue.However, we find that this assumption is problematic.Specifically, we revisit more than 20 methods on four classification tasks (Text Classification, Intent Classification, Relation Extraction, and Named Entity Recognition) under the two most popular IL settings (Class-Incremental and Task-Incremental) and reveal that most of them severely underestimate the inherent anti-forgetting ability of PLMs.Based on the observation, we propose a frustratingly easy method called SEQ* for IL with PLMs.The results show that SEQ* has competitive or superior performance compared with state-of-the-art (SOTA) IL methods yet requires considerably less trainable parameters and training time.These findings urge us to revisit the IL with PLMs and encourage future studies to have a fundamental understanding of the catastrophic forgetting in PLMs."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zheng-etal-2024-learn">
<titleInfo>
<title>Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Junhao</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shengjie</namePart>
<namePart type="family">Qiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qianli</namePart>
<namePart type="family">Ma</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>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</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>Incremental Learning (IL) has been a long-standing problem in both vision and Natural Language Processing (NLP) communities.In recent years, as Pre-trained Language Models (PLMs) have achieved remarkable progress in various NLP downstream tasks, utilizing PLMs as backbones has become a common practice in recent research of IL in NLP.Most assume that catastrophic forgetting is the biggest obstacle to achieving superior IL performance and propose various techniques to overcome this issue.However, we find that this assumption is problematic.Specifically, we revisit more than 20 methods on four classification tasks (Text Classification, Intent Classification, Relation Extraction, and Named Entity Recognition) under the two most popular IL settings (Class-Incremental and Task-Incremental) and reveal that most of them severely underestimate the inherent anti-forgetting ability of PLMs.Based on the observation, we propose a frustratingly easy method called SEQ* for IL with PLMs.The results show that SEQ* has competitive or superior performance compared with state-of-the-art (SOTA) IL methods yet requires considerably less trainable parameters and training time.These findings urge us to revisit the IL with PLMs and encourage future studies to have a fundamental understanding of the catastrophic forgetting in PLMs.</abstract>
<identifier type="citekey">zheng-etal-2024-learn</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.794</identifier>
<location>
<url>https://aclanthology.org/2024.luhme-long.794/</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>14848</start>
<end>14877</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models
%A Zheng, Junhao
%A Qiu, Shengjie
%A Ma, Qianli
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F zheng-etal-2024-learn
%X Incremental Learning (IL) has been a long-standing problem in both vision and Natural Language Processing (NLP) communities.In recent years, as Pre-trained Language Models (PLMs) have achieved remarkable progress in various NLP downstream tasks, utilizing PLMs as backbones has become a common practice in recent research of IL in NLP.Most assume that catastrophic forgetting is the biggest obstacle to achieving superior IL performance and propose various techniques to overcome this issue.However, we find that this assumption is problematic.Specifically, we revisit more than 20 methods on four classification tasks (Text Classification, Intent Classification, Relation Extraction, and Named Entity Recognition) under the two most popular IL settings (Class-Incremental and Task-Incremental) and reveal that most of them severely underestimate the inherent anti-forgetting ability of PLMs.Based on the observation, we propose a frustratingly easy method called SEQ* for IL with PLMs.The results show that SEQ* has competitive or superior performance compared with state-of-the-art (SOTA) IL methods yet requires considerably less trainable parameters and training time.These findings urge us to revisit the IL with PLMs and encourage future studies to have a fundamental understanding of the catastrophic forgetting in PLMs.
%R 10.18653/v1/2024.acl-long.794
%U https://aclanthology.org/2024.luhme-long.794/
%U https://doi.org/10.18653/v1/2024.acl-long.794
%P 14848-14877
Markdown (Informal)
[Learn or Recall? Revisiting Incremental Learning with Pre-trained Language Models](https://aclanthology.org/2024.luhme-long.794/) (Zheng et al., ACL 2024)
ACL