arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.07207cs.CL
收藏

直接到梯度:学习使用新颖的令牌进行神经文本生成

Straight to the Gradient: Learning to Use Novel Tokens for Neural Text Generation

Nanyang Technological University(南洋理工大学);Salesforce Research Asia(赛富时亚洲研究院)

Xiang Lin, Simeng Han, Shafiq Joty

摘要:

先进的大规模神经语言模型在许多语言生成任务中取得了重大成功。然而,最常用的训练目标,最大似然估计(MLE),已经被证明是有问题的,其中训练的模型更喜欢使用枯燥和重复的短语。在这项工作中,我们引入ScaleGrad,直接修改的梯度的损失函数,以弥补退化问题的标准MLE目标。通过直接操纵梯度信息,ScaleGrad使模型学习使用新的令牌。实验结果表明,我们的方法不仅在开放式生成的有效性,而且在定向生成任务。由于架构简单,我们的方法可以作为一个通用的训练目标,适用于大多数神经文本生成任务。

英文摘要:

Advanced large-scale neural language models have led to significant success in many language generation tasks. However, the most commonly used training objective, Maximum Likelihood Estimation (MLE), has been shown problematic, where the trained model prefers using dull and repetitive phrases. In this work, we introduce ScaleGrad, a modification straight to the gradient of the loss function, to remedy the degeneration issue of the standard MLE objective. By directly maneuvering the gradient information, ScaleGrad makes the model learn to use novel tokens. Empirical results show the effectiveness of our method not only in open-ended generation, but also in directed generation tasks. With the simplicity in architecture, our method can serve as a general training objective that is applicable to most of the neural text generation tasks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑