GCDT：一种全局上下文增强的序列标记深度转换结构
GCDT: A Global Context Enhanced Deep Transition Architecture for Sequence Labeling
摘要：
当前用于序列标记的最先进系统通常基于递归神经网络（RNN）家族。然而，RNN的连续隐藏状态之间的浅连接和全局信息的建模不足限制了这些模型的潜在性能。在本文中，我们试图解决这些问题，从而提出了一个全局上下文增强的深度转换架构的序列标记命名为GCDT。我们在句子中的每个位置深化状态转换路径，并进一步为每个标记分配从整个句子中学习到的全局表示。在两个标准的序列标记任务上的实验表明，在仅给定训练数据和普遍存在的词嵌入（Glove）的情况下，我们的GCDT在CoNLL 03 NER任务上达到91.96 F1，在CoNLL 2000 Chunking任务上达到95.43 F1，这优于相同设置下的最佳结果。此外，通过利用BERT作为额外的资源，我们建立了新的最先进的结果，NER上的F1为93.47，组块上的F1为97.30。
英文摘要：
Current state-of-the-art systems for sequence labeling are typically based on the family of Recurrent Neural Networks (RNNs). However, the shallow connections between consecutive hidden states of RNNs and insufficient modeling of global information restrict the potential performance of those models. In this paper, we try to address these issues, and thus propose a Global Context enhanced Deep Transition architecture for sequence labeling named GCDT. We deepen the state transition path at each position in a sentence, and further assign every token with a global representation learned from the entire sentence. Experiments on two standard sequence labeling tasks show that, given only training data and the ubiquitous word embeddings (Glove), our GCDT achieves 91.96 F1 on the CoNLL03 NER task and 95.43 F1 on the CoNLL2000 Chunking task, which outperforms the best reported results under the same settings. Furthermore, by leveraging BERT as an additional resource, we establish new state-of-the-art results with 93.47 F1 on NER and 97.30 F1 on Chunking.
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。
