arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2009.07032cs.CL
收藏

基于噪声自知识蒸馏的文本自动文摘

Noisy Self-Knowledge Distillation for Text Summarization

Microsoft Cognitive Services Research(微软认知服务研究院);University of California, Berkeley(加州大学伯克利分校);University of Edinburgh(爱丁堡大学)

Yang Liu, Sheng Shen, Mirella Lapata

摘要:

在本文中,我们将自知识蒸馏应用到文本摘要中,我们认为这可以缓解在单一参考和噪声数据集上进行最大似然训练的问题。我们的学生摘要模型不是依赖于一个热门的注释标签,而是在老师的指导下进行训练,老师会生成平滑的标签来帮助规范训练。此外,为了在训练过程中改善模型的不确定性,我们为教师和学生模型引入了多个噪声信号。我们在三个基准测试上通过实验证明,我们的框架可以提高预训练和非预训练摘要器的性能,从而获得最先进的结果。

英文摘要:

In this paper we apply self-knowledge distillation to text summarization which we argue can alleviate problems with maximum-likelihood training on single reference and noisy datasets. Instead of relying on one-hot annotation labels, our student summarization model is trained with guidance from a teacher which generates smoothed labels to help regularize training. Furthermore, to better model uncertainty during training, we introduce multiple noise signals for both teacher and student models. We demonstrate experimentally on three benchmarks that our framework boosts the performance of both pretrained and non-pretrained summarizers achieving state-of-the-art results.

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。

↑