<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="miyano-etal-2023-self">
    <titleInfo>
        <title>Self-Ensemble of N-best Generation Hypotheses by Lexically Constrained Decoding</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Ryota</namePart>
        <namePart type="family">Miyano</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Tomoyuki</namePart>
        <namePart type="family">Kajiwara</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Yuki</namePart>
        <namePart type="family">Arase</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2023-12</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Houda</namePart>
            <namePart type="family">Bouamor</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Juan</namePart>
            <namePart type="family">Pino</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Kalika</namePart>
            <namePart type="family">Bali</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Singapore</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>We propose a method that ensembles N-best hypotheses to improve natural language generation. Previous studies have achieved notable improvements in generation quality by explicitly reranking N-best candidates. These studies assume that there exists a hypothesis of higher quality. We expand the assumption to be more practical as there exist partly higher quality hypotheses in the N-best yet they may be imperfect as the entire sentences. By merging these high-quality fragments, we can obtain a higher-quality output than the single-best sentence. Specifically, we first obtain N-best hypotheses and conduct token-level quality estimation. We then apply tokens that should or should not be present in the final output as lexical constraints in decoding. Empirical experiments on paraphrase generation, summarisation, and constrained text generation confirm that our method outperforms the strong N-best reranking methods.</abstract>
    <identifier type="citekey">miyano-etal-2023-self</identifier>
    <identifier type="doi">10.18653/v1/2023.emnlp-main.905</identifier>
    <location>
        <url>https://aclanthology.org/2023.emnlp-main.905</url>
    </location>
    <part>
        <date>2023-12</date>
        <extent unit="page">
            <start>14653</start>
            <end>14661</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Self-Ensemble of N-best Generation Hypotheses by Lexically Constrained Decoding
%A Miyano, Ryota
%A Kajiwara, Tomoyuki
%A Arase, Yuki
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F miyano-etal-2023-self
%X We propose a method that ensembles N-best hypotheses to improve natural language generation. Previous studies have achieved notable improvements in generation quality by explicitly reranking N-best candidates. These studies assume that there exists a hypothesis of higher quality. We expand the assumption to be more practical as there exist partly higher quality hypotheses in the N-best yet they may be imperfect as the entire sentences. By merging these high-quality fragments, we can obtain a higher-quality output than the single-best sentence. Specifically, we first obtain N-best hypotheses and conduct token-level quality estimation. We then apply tokens that should or should not be present in the final output as lexical constraints in decoding. Empirical experiments on paraphrase generation, summarisation, and constrained text generation confirm that our method outperforms the strong N-best reranking methods.
%R 10.18653/v1/2023.emnlp-main.905
%U https://aclanthology.org/2023.emnlp-main.905
%U https://doi.org/10.18653/v1/2023.emnlp-main.905
%P 14653-14661

Markdown (Informal)

[Self-Ensemble of N-best Generation Hypotheses by Lexically Constrained Decoding](https://aclanthology.org/2023.emnlp-main.905) (Miyano et al., EMNLP 2023)

• Self-Ensemble of N-best Generation Hypotheses by Lexically Constrained Decoding (Miyano et al., EMNLP 2023)

ACL

• Ryota Miyano, Tomoyuki Kajiwara, and Yuki Arase. 2023. Self-Ensemble of N-best Generation Hypotheses by Lexically Constrained Decoding. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 14653–14661, Singapore. Association for Computational Linguistics.