Scholarlink

Abstract <p>Few-shot Biomedical Named Entity Recognition (BioNER) presents significant challenges due to limited training data and the presence of nested and discontinuous entities. To tackle these issues, a novel approach GKP-BioNER, Generation-based Few-Shot BioNER via Local Knowledge Index and Dual Prompts, is proposed. It redefines BioNER as a generation task by integrating hard and soft prompts. Specifically, GKP-BioNER constructs a localized knowledge index using a Wikipedia dump, facilitating the retrieval of semantically relevant texts to the original sentence. These texts are then reordered to prioritize the most semantically relevant content to the input data, serving as hard prompts. This helps the model to address challenges demanding domain-specific insights. Simultaneously, GKP-BioNER preserves the integrity of the pre-trained models while introducing learnable parameters as soft prompts to guide the self-attention layer, allowing the model to adapt to the context. Moreover, a soft prompt mechanism is designed to support knowledge transfer across domains. Extensive experiments on five datasets demonstrate that GKP-BioNER significantly outperforms eight state-of-the-art methods. It shows robust performance in low-resource and complex scenarios across various domains, highlighting its strength in knowledge transfer and broad applicability.</p> Graphical abstract <p>GKP-BioNER framework. It contains four modules. <b>a</b> The&#xa0;knowledge retrieval module for obtaining relevant knowledge in the target domain. <b>b</b> Encoder for encoding the input and domain knowledge into hidden representations. <b>c</b> Decoder for decoding the output of Encoder. <b>d</b> The index generation module for producing NER results. Soft prompt parameters that are trainable are introduced into the attention layers of Encoder and Decoder modules, respectively.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Generation-Based Few-Shot BioNER via Local Knowledge Index and Dual Prompts

  • 期刊论文

  • 发表日期：

    2025年5月10日

查看全文
  • Weixin Li,
  • Hong Wang,
  • Wei Li,
  • Jun Zhao,
  • Yanshen Sun

摘要

Abstract

Few-shot Biomedical Named Entity Recognition (BioNER) presents significant challenges due to limited training data and the presence of nested and discontinuous entities. To tackle these issues, a novel approach GKP-BioNER, Generation-based Few-Shot BioNER via Local Knowledge Index and Dual Prompts, is proposed. It redefines BioNER as a generation task by integrating hard and soft prompts. Specifically, GKP-BioNER constructs a localized knowledge index using a Wikipedia dump, facilitating the retrieval of semantically relevant texts to the original sentence. These texts are then reordered to prioritize the most semantically relevant content to the input data, serving as hard prompts. This helps the model to address challenges demanding domain-specific insights. Simultaneously, GKP-BioNER preserves the integrity of the pre-trained models while introducing learnable parameters as soft prompts to guide the self-attention layer, allowing the model to adapt to the context. Moreover, a soft prompt mechanism is designed to support knowledge transfer across domains. Extensive experiments on five datasets demonstrate that GKP-BioNER significantly outperforms eight state-of-the-art methods. It shows robust performance in low-resource and complex scenarios across various domains, highlighting its strength in knowledge transfer and broad applicability.

Graphical abstract

GKP-BioNER framework. It contains four modules. a The knowledge retrieval module for obtaining relevant knowledge in the target domain. b Encoder for encoding the input and domain knowledge into hidden representations. c Decoder for decoding the output of Encoder. d The index generation module for producing NER results. Soft prompt parameters that are trainable are introduced into the attention layers of Encoder and Decoder modules, respectively.