论文标题
$ k $ nn-ner：带有最近邻居搜索的命名实体识别
$k$NN-NER: Named Entity Recognition with Nearest Neighbor Search
论文作者
论文摘要
受到NLP〜 \ citep {Khandelwal2019Generalitized的最新进展的启发，Khandelwal2020neart，meng2021gnn}的启发，在本文中，我们引入了$ k $ thegright Negright Negright Ner（$ K $ nn-ner）框架，从$ k $ nn-nn-ner框架中提出了分配$ $ k的分配。该策略使该模型更有能力处理长尾案例，以及更好的几次学习能力。 $ k $ nn-ner在培训阶段不需要额外的操作，并且通过将$ k $近的邻居搜索插入Vanilla ner型号中，$ k $ nn-nn-nn-ner始终优于其香草的范围：我们实现了一种新的最先进的F1得分评分，在杂质上使用了72.03（+1.25）的72.03（+1.25），并在杂质上使用了杂货。此外，我们表明，$ k $ nn-ner可以与培训数据少40％\％\％\％的培训模型获得可比的结果。可在\ url {https://github.com/shannonai/knn-ner}上获得代码。
Inspired by recent advances in retrieval augmented methods in NLP~\citep{khandelwal2019generalization,khandelwal2020nearest,meng2021gnn}, in this paper, we introduce a $k$ nearest neighbor NER ($k$NN-NER) framework, which augments the distribution of entity labels by assigning $k$ nearest neighbors retrieved from the training set. This strategy makes the model more capable of handling long-tail cases, along with better few-shot learning abilities. $k$NN-NER requires no additional operation during the training phase, and by interpolating $k$ nearest neighbors search into the vanilla NER model, $k$NN-NER consistently outperforms its vanilla counterparts: we achieve a new state-of-the-art F1-score of 72.03 (+1.25) on the Chinese Weibo dataset and improved results on a variety of widely used NER benchmarks. Additionally, we show that $k$NN-NER can achieve comparable results to the vanilla NER model with 40\% less amount of training data. Code available at \url{https://github.com/ShannonAI/KNN-NER}.
