• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Computer speech and language >Adversarial subsequences for unconditional text generation
【24h】

Adversarial subsequences for unconditional text generation

机译：对无条件文本生成的对抗子条断

获取原文
获取原文并翻译 | 示例
       
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Generative Adversarial Nets (GAN) has been successfully introduced to unconditional generating text to alleviate exposure bias. However, the discriminator in this model only evaluates the entire sequence, which causes feedback sparsity and mode collapse. To tackle these problems, we propose a novel mechanism. The mechanism first segments the entire sequence into several subsequences. Then, these subsequences, together with the entire sequence, are evaluated individually by the discriminator. Finally, these feedback signals are all used to guide the learning of GAN. This mechanism learns the generation of both the entire sequence and the subsequences simultaneously. Learning to generate subsequences is easy and is helpful in generating an entire sequence. It is easy to improve the existing GAN-based models with this mechanism. Although Li et al. (2017) segments the generated responses in a conditional text generation task, i.e., a dialogue system, they conclude it is weaker than the Monte Carlo search. However, for unconditional text generation, we observe that adversarial learning on subsequences works well. We rebuild three previous models with our mechanism, and the experimental results on two benchmark datasets show these models are improved greatly and outperform the state-of-the-art model.
机译：已成功引入生成的对抗网（GAN）以无条件的生成文本来缓解曝光偏见。然而，该模型中的鉴别器仅评估整个序列，这会导致反馈稀疏性和模式崩溃。为了解决这些问题，我们提出了一种新的机制。该机制首先将整个序列分成几个子序列。然后，通过鉴别器单独评估这些子序列与整个序列一起评估。最后，这些反馈信号都是用于指导GaN的学习。该机制同时学习整个序列和子序列的生成。学习生成子序列很容易，并且有助于生成整个序列。通过这种机制，易于改善现有的基于GaN的模型。虽然李等人。 （2017）分段在条件文本生成任务中产生的响应，即对话系统，他们总结它比蒙特卡罗搜索弱。然而，对于无条件的文本生成，我们观察到对局部的对抗学习运作良好。我们使用我们的机制重建三种以前的模型，并且两个基准数据集上的实验结果显示了这些模型的大大提高，优于最先进的模型。

著录项

  • 来源
    《Computer speech and language》 |2021年第11期|101242.1-101242.8|共8页
  • 作者

    Xingyuan Chen; Peng Jin; Yanzhe Li; Jiuhua Zhang; Xinyu Dai; Jiajun Chen;

  • 作者单位

    Department of Computer Science and Technology Nanjing University China;

    School of Artificial Intelligence Leshan Normal University China;

    School of Artificial Intelligence Leshan Normal University China;

    School of Artificial Intelligence Leshan Normal University China;

    Department of Computer Science and Technology Nanjing University China;

    Department of Computer Science and Technology Nanjing University China;

  • 收录信息 美国《科学引文索引》(SCI);美国《工程索引》(EI);
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    Unconditional text generation; GAN; Subsequences;

    机译：无条件文本生成;甘;续工;

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号