中国邮电高校学报(英文) ›› 2023, Vol. 30 ›› Issue (5): 32-41.doi: 10.19682/j.cnki.1005-8885.2023.0011

所属专题: Special Topic on Digital Human

• Special Topic : Digital Human • 上一篇    下一篇

Pointer-prototype fusion network for few-shot named entity recognition

赵海英, 郭轩   

  1. 北京邮电大学
  • 收稿日期:2023-07-20 修回日期:2023-09-02 出版日期:2023-10-31 发布日期:2023-10-30
  • 通讯作者: 赵海英 E-mail:zhaohaiying@bupt.edu.cn
  • 基金资助:
    the National Key Research and Development Project (2021YFF0901701).

Pointer-prototype fusion network for few-shot named entity recognition

  Hai-Ying ZHAO2 , Xuan GUO1   

  1. 1. School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing 100876, China  2. School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China
  • Received:2023-07-20 Revised:2023-09-02 Online:2023-10-31 Published:2023-10-30
  • Contact: Hai-Ying ZHAO E-mail:zhaohaiying@bupt.edu.cn
  • Supported by:
    the National Key Research and Development Project (2021YFF0901701).

摘要:

   Few-shot named entity recognition (NER) aims to identify named entities in new domains using a limited amount of annotated data. Previous methods divided this task into entity span detection and entity classification, achieving good results. However these methods are limited by the imbalance between the entity and non-entity categories due to the use of sequence labeling for entity span detection. To this end, a point-proto network ( PPN) combining pointer and prototypical networks was proposed. Specifically, the pointer network generates the position of entities in sentences in the entity span detection stage. The prototypical network builds semantic prototypes of entity types and classifies entities based on their distance from these prototypes in the entity classification stage. Moreover, the low-rank adaptation ( LoRA) fine-tuning method, which involves freezing the pre-trained weights and injecting a trainable decomposition matrix, reduces the parameters that need to be trained and saved. Extensive experiments on the few-shot NER Dataset (Few-NERD) and Cross-Dataset demonstrate the superiority of PPN in this domain.

关键词: few-shot named entity recognition (NER), pointer network, prototypical network, low-rank adaptation

Abstract:

   Few-shot named entity recognition (NER) aims to identify named entities in new domains using a limited amount of annotated data. Previous methods divided this task into entity span detection and entity classification, achieving good results. However these methods are limited by the imbalance between the entity and non-entity categories due to the use of sequence labeling for entity span detection. To this end, a point-proto network ( PPN) combining pointer and prototypical networks was proposed. Specifically, the pointer network generates the position of entities in sentences in the entity span detection stage. The prototypical network builds semantic prototypes of entity types and classifies entities based on their distance from these prototypes in the entity classification stage. Moreover, the low-rank adaptation ( LoRA) fine-tuning method, which involves freezing the pre-trained weights and injecting a trainable decomposition matrix, reduces the parameters that need to be trained and saved. Extensive experiments on the few-shot NER Dataset (Few-NERD) and Cross-Dataset demonstrate the superiority of PPN in this domain.

Key words:

few-shot named entity recognition (NER), pointer network, prototypical network, low-rank adaptation