<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="goyal-etal-2016-natural">
    <titleInfo>
        <title>Natural Language Generation through Character-based RNNs with Finite-state Prior Knowledge</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Raghav</namePart>
        <namePart type="family">Goyal</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Marc</namePart>
        <namePart type="family">Dymetman</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Eric</namePart>
        <namePart type="family">Gaussier</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2016-dec</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers</title>
        </titleInfo>
        <originInfo>
            <publisher>The COLING 2016 Organizing Committee</publisher>
            <place>
                <placeTerm type="text">Osaka, Japan</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>Recently Wen et al. (2015) have proposed a Recurrent Neural Network (RNN) approach to the generation of utterances from dialog acts, and shown that although their model requires less effort to develop than a rule-based system, it is able to improve certain aspects of the utterances, in particular their naturalness. However their system employs generation at the word-level, which requires one to pre-process the data by substituting named entities with placeholders. This pre-processing prevents the model from handling some contextual effects and from managing multiple occurrences of the same attribute. Our approach uses a character-level model, which unlike the word-level model makes it possible to learn to “copy” information from the dialog act to the target without having to pre-process the input. In order to avoid generating non-words and inventing information not present in the input, we propose a method for incorporating prior knowledge into the RNN in the form of a weighted finite-state automaton over character sequences. Automatic and human evaluations show improved performance over baselines on several evaluation criteria.</abstract>
    <identifier type="citekey">goyal-etal-2016-natural</identifier>
    <location>
        <url>https://aclanthology.org/C16-1103</url>
    </location>
    <part>
        <date>2016-dec</date>
        <extent unit="page">
            <start>1083</start>
            <end>1092</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Natural Language Generation through Character-based RNNs with Finite-state Prior Knowledge
%A Goyal, Raghav
%A Dymetman, Marc
%A Gaussier, Eric
%S Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers
%D 2016
%8 dec
%I The COLING 2016 Organizing Committee
%C Osaka, Japan
%F goyal-etal-2016-natural
%X Recently Wen et al. (2015) have proposed a Recurrent Neural Network (RNN) approach to the generation of utterances from dialog acts, and shown that although their model requires less effort to develop than a rule-based system, it is able to improve certain aspects of the utterances, in particular their naturalness. However their system employs generation at the word-level, which requires one to pre-process the data by substituting named entities with placeholders. This pre-processing prevents the model from handling some contextual effects and from managing multiple occurrences of the same attribute. Our approach uses a character-level model, which unlike the word-level model makes it possible to learn to “copy” information from the dialog act to the target without having to pre-process the input. In order to avoid generating non-words and inventing information not present in the input, we propose a method for incorporating prior knowledge into the RNN in the form of a weighted finite-state automaton over character sequences. Automatic and human evaluations show improved performance over baselines on several evaluation criteria.
%U https://aclanthology.org/C16-1103
%P 1083-1092

Markdown (Informal)

[Natural Language Generation through Character-based RNNs with Finite-state Prior Knowledge](https://aclanthology.org/C16-1103) (Goyal et al., COLING 2016)

• Natural Language Generation through Character-based RNNs with Finite-state Prior Knowledge (Goyal et al., COLING 2016)

ACL

• Raghav Goyal, Marc Dymetman, and Eric Gaussier. 2016. Natural Language Generation through Character-based RNNs with Finite-state Prior Knowledge. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 1083–1092, Osaka, Japan. The COLING 2016 Organizing Committee.