<?xml version="1.0" encoding="utf-8"?>
<TEI xmlns="http://www.tei-c.org/ns/1.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:hal="http://hal.archives-ouvertes.fr/" xmlns:gml="http://www.opengis.net/gml/3.3/" xmlns:gmlce="http://www.opengis.net/gml/3.3/ce" version="1.1" xsi:schemaLocation="http://www.tei-c.org/ns/1.0 http://api.archives-ouvertes.fr/documents/aofr-sword.xsd">
  <teiHeader>
    <fileDesc>
      <titleStmt>
        <title>HAL TEI export of hal-04666458</title>
      </titleStmt>
      <publicationStmt>
        <distributor>CCSD</distributor>
        <availability status="restricted">
          <licence target="https://creativecommons.org/publicdomain/zero/1.0/">CC0 1.0 - Universal</licence>
        </availability>
        <date when="2026-05-22T02:51:30+02:00"/>
      </publicationStmt>
      <sourceDesc>
        <p part="N">HAL API Platform</p>
      </sourceDesc>
    </fileDesc>
  </teiHeader>
  <text>
    <body>
      <listBibl>
        <biblFull>
          <titleStmt>
            <title xml:lang="en">Improved Transformer-Based Implicit Latent GAN with Multi-headed Self-attention for Unconditional Text Generation</title>
            <author role="aut">
              <persName>
                <forename type="first">Fuji</forename>
                <surname>Ren</surname>
              </persName>
              <email type="md5">c2999f8d9969ce8c2bdb6cd32d68d033</email>
              <email type="domain">uestc.edu.cn</email>
              <idno type="idhal" notation="numeric">1405993</idno>
              <idno type="halauthorid" notation="string">2876012-1405993</idno>
              <affiliation ref="#struct-373746"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Ziyun</forename>
                <surname>Jiao</surname>
              </persName>
              <email type="md5">539e89f165799e623c14bf0efa123dca</email>
              <email type="domain">tokushima-u.ac.jp</email>
              <idno type="idhal" notation="numeric">1405994</idno>
              <idno type="halauthorid" notation="string">3209401-1405994</idno>
              <affiliation ref="#struct-303688"/>
            </author>
            <author role="aut">
              <persName>
                <forename type="first">Xin</forename>
                <surname>Kang</surname>
              </persName>
              <email type="md5">6fd9d1151265b61c491b3f5fcc3bd256</email>
              <email type="domain">is.tokushima-u.ac.jp</email>
              <idno type="idhal" notation="numeric">1275130</idno>
              <idno type="halauthorid" notation="string">630754-1275130</idno>
              <affiliation ref="#struct-303688"/>
            </author>
            <editor role="depositor">
              <persName>
                <forename>Hal</forename>
                <surname>Ifip</surname>
              </persName>
              <email type="md5">2073ac78024b6e13f2714db96e9b1e63</email>
              <email type="domain">inria.fr</email>
            </editor>
          </titleStmt>
          <editionStmt>
            <edition n="v1" type="current">
              <date type="whenSubmitted">2024-08-01 15:05:25</date>
              <date type="whenModified">2024-08-01 15:12:49</date>
              <date type="whenReleased">2024-08-01 15:12:49</date>
              <date type="whenProduced">2022-10-28</date>
              <date type="whenEndEmbargoed">2025-01-01</date>
              <ref type="file" target="https://hal.science/hal-04666458v1/document">
                <date notBefore="2025-01-01"/>
              </ref>
              <ref type="file" subtype="author" n="1" target="https://hal.science/hal-04666458v1/file/537972_1_En_18_Chapter.pdf" id="file-4666458-4052574">
                <date notBefore="2025-01-01"/>
              </ref>
            </edition>
            <respStmt>
              <resp>contributor</resp>
              <name key="200187">
                <persName>
                  <forename>Hal</forename>
                  <surname>Ifip</surname>
                </persName>
                <email type="md5">2073ac78024b6e13f2714db96e9b1e63</email>
                <email type="domain">inria.fr</email>
              </name>
            </respStmt>
          </editionStmt>
          <publicationStmt>
            <distributor>CCSD</distributor>
            <idno type="halId">hal-04666458</idno>
            <idno type="halUri">https://hal.science/hal-04666458</idno>
            <idno type="halBibtex">ren:hal-04666458</idno>
            <idno type="halRefHtml">&lt;i&gt;5th International Conference on Intelligence Science (ICIS)&lt;/i&gt;, Oct 2022, Xi'an, China. pp.166-173, &lt;a target="_blank" href="https://dx.doi.org/10.1007/978-3-031-14903-0_18"&gt;&amp;#x27E8;10.1007/978-3-031-14903-0_18&amp;#x27E9;&lt;/a&gt;</idno>
            <idno type="halRef">5th International Conference on Intelligence Science (ICIS), Oct 2022, Xi'an, China. pp.166-173, &amp;#x27E8;10.1007/978-3-031-14903-0_18&amp;#x27E9;</idno>
            <availability status="restricted">
              <licence target="https://creativecommons.org/licenses/by/4.0/">CC BY 4.0 - Attribution<ref corresp="#file-4666458-4052574"/></licence>
            </availability>
          </publicationStmt>
          <seriesStmt>
            <idno type="stamp" n="IFIP">IFIP - International Federation for Information Processing</idno>
            <idno type="stamp" n="IFIP-AICT" corresp="IFIP">IFIP Advances in Information and Communication Technology</idno>
            <idno type="stamp" n="IFIP-TC" corresp="IFIP">IFIP Technical Committees </idno>
            <idno type="stamp" n="IFIP-TC12" corresp="IFIP-TC">TC12 - Artificial Intelligence</idno>
            <idno type="stamp" n="IFIP-ICIS" corresp="IFIP">International Conference on Intelligence Science</idno>
            <idno type="stamp" n="IFIP-AICT-659" corresp="IFIP-AICT">Intelligence Science IV</idno>
          </seriesStmt>
          <notesStmt>
            <note type="commentary">Part 4: Language Cognition</note>
            <note type="audience" n="2">International</note>
            <note type="invited" n="0">No</note>
            <note type="popular" n="0">No</note>
            <note type="peer" n="1">Yes</note>
            <note type="proceedings" n="1">Yes</note>
          </notesStmt>
          <sourceDesc>
            <biblStruct>
              <analytic>
                <title xml:lang="en">Improved Transformer-Based Implicit Latent GAN with Multi-headed Self-attention for Unconditional Text Generation</title>
                <author role="aut">
                  <persName>
                    <forename type="first">Fuji</forename>
                    <surname>Ren</surname>
                  </persName>
                  <email type="md5">c2999f8d9969ce8c2bdb6cd32d68d033</email>
                  <email type="domain">uestc.edu.cn</email>
                  <idno type="idhal" notation="numeric">1405993</idno>
                  <idno type="halauthorid" notation="string">2876012-1405993</idno>
                  <affiliation ref="#struct-373746"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Ziyun</forename>
                    <surname>Jiao</surname>
                  </persName>
                  <email type="md5">539e89f165799e623c14bf0efa123dca</email>
                  <email type="domain">tokushima-u.ac.jp</email>
                  <idno type="idhal" notation="numeric">1405994</idno>
                  <idno type="halauthorid" notation="string">3209401-1405994</idno>
                  <affiliation ref="#struct-303688"/>
                </author>
                <author role="aut">
                  <persName>
                    <forename type="first">Xin</forename>
                    <surname>Kang</surname>
                  </persName>
                  <email type="md5">6fd9d1151265b61c491b3f5fcc3bd256</email>
                  <email type="domain">is.tokushima-u.ac.jp</email>
                  <idno type="idhal" notation="numeric">1275130</idno>
                  <idno type="halauthorid" notation="string">630754-1275130</idno>
                  <affiliation ref="#struct-303688"/>
                </author>
              </analytic>
              <monogr>
                <idno type="isbn">978-3-031-14902-3</idno>
                <title level="m">IFIP Advances in Information and Communication Technology</title>
                <meeting>
                  <title>5th International Conference on Intelligence Science (ICIS)</title>
                  <date type="start">2022-10-28</date>
                  <date type="end">2022-10-31</date>
                  <settlement>Xi'an</settlement>
                  <country key="CN">China</country>
                </meeting>
                <editor>Zhongzhi Shi</editor>
                <imprint>
                  <publisher>Springer International Publishing</publisher>
                  <biblScope unit="serie">Intelligence Science IV</biblScope>
                  <biblScope unit="volume">AICT-659</biblScope>
                  <biblScope unit="pp">166-173</biblScope>
                  <date type="datePub">2022</date>
                </imprint>
              </monogr>
              <idno type="doi">10.1007/978-3-031-14903-0_18</idno>
            </biblStruct>
          </sourceDesc>
          <profileDesc>
            <langUsage>
              <language ident="en">English</language>
            </langUsage>
            <textClass>
              <keywords scheme="author">
                <term xml:lang="en">Unconditional text generation</term>
                <term xml:lang="en">GAN</term>
                <term xml:lang="en">Self-attention</term>
                <term xml:lang="en">TILGAN</term>
              </keywords>
              <classCode scheme="halDomain" n="info">Computer Science [cs]</classCode>
              <classCode scheme="halTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halOldTypology" n="COMM">Conference papers</classCode>
              <classCode scheme="halTreeTypology" n="COMM">Conference papers</classCode>
            </textClass>
            <abstract xml:lang="en">
              <p>Generative Adversarial Network (GAN) is widely used in computer vision, such as image generation and other tasks. In recent years, GAN has also been developed in the field of unconditional text generation. In this work, we improve TILGAN for unconditional text generation by refactoring the generator. In short, we use Multi-headed Self-attention to replace the Linear layer and BN layer to endow the generator with better text generation capabilities. Our model consists of three components: a transformer autoencoder, a Multi-headed Self attention based generator and a linear based discriminator. The encoder in transformer autoencoder is used to generate the distribution of real samples, and the decoder is used to decode real or generated sentence vector into text. The loss functions for autoencoder and GAN are cross entropy and KL divergence, respectively. On the MS COCO dataset, the proposed model has achieved a better BLEU score than TILGAN. Our ablation experiments also proved the effectiveness of the proposed generator network for unconditional text generation.</p>
            </abstract>
            <particDesc>
              <org type="consortium">TC 12</org>
            </particDesc>
          </profileDesc>
        </biblFull>
      </listBibl>
    </body>
    <back>
      <listOrg type="structures">
        <org type="institution" xml:id="struct-373746" status="VALID">
          <idno type="ROR">https://ror.org/04qr3zq92</idno>
          <orgName>University of Electronic Science and Technology of China [Chengdu]</orgName>
          <orgName type="acronym">UESTC</orgName>
          <desc>
            <address>
              <addrLine>4 Jianshe North Rd 2nd Section, Chenghua, Chengdu, Sichuan, Chine, 610051</addrLine>
              <country key="CN"/>
            </address>
            <ref type="url">https://en.uestc.edu.cn/</ref>
          </desc>
        </org>
        <org type="institution" xml:id="struct-303688" status="VALID">
          <idno type="ROR">https://ror.org/044vy1d05</idno>
          <orgName>Tokushima University</orgName>
          <desc>
            <address>
              <addrLine>2-24, Shinkura-cho, Tokushima 770-8501</addrLine>
              <country key="JP"/>
            </address>
            <ref type="url">http://www.tokushima-u.ac.jp/english/</ref>
          </desc>
        </org>
      </listOrg>
    </back>
  </text>
</TEI>