将嵌套实体识别转化为单次序列标注，简化复杂度且效果不减。
Nested Named Entity Recognition as Single-Pass Sequence Labeling
- 通过线性化句法结构，把嵌套实体识别转为简单词元分类。
- 仅用 n 次标记操作即可识别所有嵌套实体，性能媲美复杂方法。
- 兼容主流序列标注工具，可快速部署于各类NLP系统。
我们将嵌套命名实体识别（NNER）建模为序列标注任务，利用已有句法结构线性化方法，有效将这一结构预测问题的复杂度降低为简单的词元分类。结合这些句法线性化与预训练编码器，我们的方法在仅执行恰好 n 次标记操作的前提下，成功捕获嵌套实体。该方法在性能上达到与更复杂系统相当的水平，同时可使用任意现成的序列标注库进行训练。
原文摘要 · Abstract (English)
We cast nested named entity recognition (NNER) as a sequence labeling task by leveraging prior work that linearizes constituency structures, effectively reducing the complexity of this structured prediction problem to straightforward token classification. By combining these constituency linearizations with pretrained encoders, our method captures nested entities while performing exactly n tagging actions. Our approach achieves competitive performance compared to less efficient systems, and it can be trained using any off-the-shelf sequence labeling library.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。