arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.14465cs.CLcs.AI
收藏

基于transformer的句子分类器的Zero-shot序列标注

Zero-shot Sequence Labeling for Transformer-based Sentence Classifiers

Imperial College London(帝国理工学院);King’s College London(伦敦国王学院)

Kamil Bujel, Helen Yannakoudakis, Marek Rei

摘要:

我们研究如何在没有任何直接监督的情况下,在令牌级别将并行级Transformers修改为有效的序列标记器。当应用于基于变换器的架构时,用于zero-shot序列标记的现有方法不能很好地执行。由于Transformers包含多层多头自我注意,句子中的信息会分布在许多标记之间,从而对zero-shot标记级性能产生负面影响。我们发现,一个软的注意力模块,明确鼓励关注权重的锐度可以显着优于现有的方法。

英文摘要:

We investigate how sentence-level transformers can be modified into effective sequence labelers at the token level without any direct supervision. Existing approaches to zero-shot sequence labeling do not perform well when applied on transformer-based architectures. As transformers contain multiple layers of multi-head self-attention, information in the sentence gets distributed between many tokens, negatively affecting zero-shot token-level performance. We find that a soft attention module which explicitly encourages sharpness of attention weights can significantly outperform existing methods.

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。

↑