@inproceedings{razumovskaia-etal-2022-data,
title = "Data Augmentation and Learned Layer Aggregation for Improved Multilingual Language Understanding in Dialogue",
author = "Razumovskaia, Evgeniia and
Vuli{\'c}, Ivan and
Korhonen, Anna",
editor = "Muresan, Smaranda and
Nakov, Preslav and
Villavicencio, Aline",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2022",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-acl.160",
doi = "10.18653/v1/2022.findings-acl.160",
pages = "2017--2033",
abstract = "Scaling dialogue systems to a multitude of domains, tasks and languages relies on costly and time-consuming data annotation for different domain-task-language configurations. The annotation efforts might be substantially reduced by the methods that generalise well in zero- and few-shot scenarios, and also effectively leverage external unannotated data sources (e.g., Web-scale corpora). We propose two methods to this aim, offering improved dialogue natural language understanding (NLU) across multiple languages: 1) Multi-SentAugment, and 2) LayerAgg. Multi-SentAugment is a self-training method which augments available (typically few-shot) training data with similar (automatically labelled) in-domain sentences from large monolingual Web-scale corpora. LayerAgg learns to select and combine useful semantic information scattered across different layers of a Transformer model (e.g., mBERT); it is especially suited for zero-shot scenarios as semantically richer representations should strengthen the model{'}s cross-lingual capabilities. Applying the two methods with state-of-the-art NLU models obtains consistent improvements across two standard multilingual NLU datasets covering 16 diverse languages. The gains are observed in zero-shot, few-shot, and even in full-data scenarios. The results also suggest that the two methods achieve a synergistic effect: the best overall performance in few-shot setups is attained when the methods are used together.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="razumovskaia-etal-2022-data">
<titleInfo>
<title>Data Augmentation and Learned Layer Aggregation for Improved Multilingual Language Understanding in Dialogue</title>
</titleInfo>
<name type="personal">
<namePart type="given">Evgeniia</namePart>
<namePart type="family">Razumovskaia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="family">Vulić</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Korhonen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Smaranda</namePart>
<namePart type="family">Muresan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Preslav</namePart>
<namePart type="family">Nakov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aline</namePart>
<namePart type="family">Villavicencio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Scaling dialogue systems to a multitude of domains, tasks and languages relies on costly and time-consuming data annotation for different domain-task-language configurations. The annotation efforts might be substantially reduced by the methods that generalise well in zero- and few-shot scenarios, and also effectively leverage external unannotated data sources (e.g., Web-scale corpora). We propose two methods to this aim, offering improved dialogue natural language understanding (NLU) across multiple languages: 1) Multi-SentAugment, and 2) LayerAgg. Multi-SentAugment is a self-training method which augments available (typically few-shot) training data with similar (automatically labelled) in-domain sentences from large monolingual Web-scale corpora. LayerAgg learns to select and combine useful semantic information scattered across different layers of a Transformer model (e.g., mBERT); it is especially suited for zero-shot scenarios as semantically richer representations should strengthen the model’s cross-lingual capabilities. Applying the two methods with state-of-the-art NLU models obtains consistent improvements across two standard multilingual NLU datasets covering 16 diverse languages. The gains are observed in zero-shot, few-shot, and even in full-data scenarios. The results also suggest that the two methods achieve a synergistic effect: the best overall performance in few-shot setups is attained when the methods are used together.</abstract>
<identifier type="citekey">razumovskaia-etal-2022-data</identifier>
<identifier type="doi">10.18653/v1/2022.findings-acl.160</identifier>
<location>
<url>https://aclanthology.org/2022.findings-acl.160</url>
</location>
<part>
<date>2022-05</date>
<extent unit="page">
<start>2017</start>
<end>2033</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Data Augmentation and Learned Layer Aggregation for Improved Multilingual Language Understanding in Dialogue
%A Razumovskaia, Evgeniia
%A Vulić, Ivan
%A Korhonen, Anna
%Y Muresan, Smaranda
%Y Nakov, Preslav
%Y Villavicencio, Aline
%S Findings of the Association for Computational Linguistics: ACL 2022
%D 2022
%8 May
%I Association for Computational Linguistics
%C Dublin, Ireland
%F razumovskaia-etal-2022-data
%X Scaling dialogue systems to a multitude of domains, tasks and languages relies on costly and time-consuming data annotation for different domain-task-language configurations. The annotation efforts might be substantially reduced by the methods that generalise well in zero- and few-shot scenarios, and also effectively leverage external unannotated data sources (e.g., Web-scale corpora). We propose two methods to this aim, offering improved dialogue natural language understanding (NLU) across multiple languages: 1) Multi-SentAugment, and 2) LayerAgg. Multi-SentAugment is a self-training method which augments available (typically few-shot) training data with similar (automatically labelled) in-domain sentences from large monolingual Web-scale corpora. LayerAgg learns to select and combine useful semantic information scattered across different layers of a Transformer model (e.g., mBERT); it is especially suited for zero-shot scenarios as semantically richer representations should strengthen the model’s cross-lingual capabilities. Applying the two methods with state-of-the-art NLU models obtains consistent improvements across two standard multilingual NLU datasets covering 16 diverse languages. The gains are observed in zero-shot, few-shot, and even in full-data scenarios. The results also suggest that the two methods achieve a synergistic effect: the best overall performance in few-shot setups is attained when the methods are used together.
%R 10.18653/v1/2022.findings-acl.160
%U https://aclanthology.org/2022.findings-acl.160
%U https://doi.org/10.18653/v1/2022.findings-acl.160
%P 2017-2033
Markdown (Informal)
[Data Augmentation and Learned Layer Aggregation for Improved Multilingual Language Understanding in Dialogue](https://aclanthology.org/2022.findings-acl.160) (Razumovskaia et al., Findings 2022)
ACL