@inproceedings{zhang-etal-2022-clle,
title = "{CLLE}: A Benchmark for Continual Language Learning Evaluation in Multilingual Machine Translation",
author = "Zhang, Han and
Zhang, Sheng and
Xiang, Yang and
Liang, Bin and
Su, Jinsong and
Miao, Zhongjian and
Wang, Hui and
Xu, Ruifeng",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.30",
doi = "10.18653/v1/2022.findings-emnlp.30",
pages = "428--443",
abstract = "Continual Language Learning (CLL) in multilingual translation is inevitable when new languages are required to be translated. Due to the lack of unified and generalized benchmarks, the evaluation of existing methods is greatly influenced by experimental design which usually has a big gap from the industrial demands. In this work, we propose the first Continual Language Learning Evaluation benchmark CLLE in multilingual translation. CLLE consists of a Chinese-centric corpus {---} CN-25 and two CLL tasks {---} the close-distance language continual learning task and the language family continual learning task designed for real and disparate demands. Different from existing translation benchmarks, CLLE considers several restrictions for CLL, including domain distribution alignment, content overlap, language diversity, and the balance of corpus. Furthermore, we propose a novel framework COMETA based on Constrained Optimization and META-learning to alleviate catastrophic forgetting and dependency on history training data by using a meta-model to retain the important parameters for old languages. Our experiments prove that CLLE is a challenging CLL benchmark and that our proposed method is effective when compared with other strong baselines. Due to the construction of the corpus, the task designing and the evaluation method are independent of the centric language, we also construct and release the English-centric corpus EN-25 to facilitate academic research.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2022-clle">
<titleInfo>
<title>CLLE: A Benchmark for Continual Language Learning Evaluation in Multilingual Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Han</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sheng</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Xiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bin</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jinsong</namePart>
<namePart type="family">Su</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhongjian</namePart>
<namePart type="family">Miao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hui</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</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>Findings of the Association for Computational Linguistics: EMNLP 2022</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>Continual Language Learning (CLL) in multilingual translation is inevitable when new languages are required to be translated. Due to the lack of unified and generalized benchmarks, the evaluation of existing methods is greatly influenced by experimental design which usually has a big gap from the industrial demands. In this work, we propose the first Continual Language Learning Evaluation benchmark CLLE in multilingual translation. CLLE consists of a Chinese-centric corpus — CN-25 and two CLL tasks — the close-distance language continual learning task and the language family continual learning task designed for real and disparate demands. Different from existing translation benchmarks, CLLE considers several restrictions for CLL, including domain distribution alignment, content overlap, language diversity, and the balance of corpus. Furthermore, we propose a novel framework COMETA based on Constrained Optimization and META-learning to alleviate catastrophic forgetting and dependency on history training data by using a meta-model to retain the important parameters for old languages. Our experiments prove that CLLE is a challenging CLL benchmark and that our proposed method is effective when compared with other strong baselines. Due to the construction of the corpus, the task designing and the evaluation method are independent of the centric language, we also construct and release the English-centric corpus EN-25 to facilitate academic research.</abstract>
<identifier type="citekey">zhang-etal-2022-clle</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.30</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.30</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>428</start>
<end>443</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CLLE: A Benchmark for Continual Language Learning Evaluation in Multilingual Machine Translation
%A Zhang, Han
%A Zhang, Sheng
%A Xiang, Yang
%A Liang, Bin
%A Su, Jinsong
%A Miao, Zhongjian
%A Wang, Hui
%A Xu, Ruifeng
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F zhang-etal-2022-clle
%X Continual Language Learning (CLL) in multilingual translation is inevitable when new languages are required to be translated. Due to the lack of unified and generalized benchmarks, the evaluation of existing methods is greatly influenced by experimental design which usually has a big gap from the industrial demands. In this work, we propose the first Continual Language Learning Evaluation benchmark CLLE in multilingual translation. CLLE consists of a Chinese-centric corpus — CN-25 and two CLL tasks — the close-distance language continual learning task and the language family continual learning task designed for real and disparate demands. Different from existing translation benchmarks, CLLE considers several restrictions for CLL, including domain distribution alignment, content overlap, language diversity, and the balance of corpus. Furthermore, we propose a novel framework COMETA based on Constrained Optimization and META-learning to alleviate catastrophic forgetting and dependency on history training data by using a meta-model to retain the important parameters for old languages. Our experiments prove that CLLE is a challenging CLL benchmark and that our proposed method is effective when compared with other strong baselines. Due to the construction of the corpus, the task designing and the evaluation method are independent of the centric language, we also construct and release the English-centric corpus EN-25 to facilitate academic research.
%R 10.18653/v1/2022.findings-emnlp.30
%U https://aclanthology.org/2022.findings-emnlp.30
%U https://doi.org/10.18653/v1/2022.findings-emnlp.30
%P 428-443
Markdown (Informal)
[CLLE: A Benchmark for Continual Language Learning Evaluation in Multilingual Machine Translation](https://aclanthology.org/2022.findings-emnlp.30) (Zhang et al., Findings 2022)
ACL
- Han Zhang, Sheng Zhang, Yang Xiang, Bin Liang, Jinsong Su, Zhongjian Miao, Hui Wang, and Ruifeng Xu. 2022. CLLE: A Benchmark for Continual Language Learning Evaluation in Multilingual Machine Translation. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 428–443, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.