发表机构
Yunnan University（云南大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
DE-NER是一个对话引导框架，利用大型语言模型的对话能力解决零样本命名实体识别的提示与演示工程局限，在多基准零样本设置下平均F1值提升3.75%。
AI 中文摘要
零样本命名实体识别（NER）的最新进展通过将序列标注转化为问答任务，构建了强大的基准，大型语言模型（LLM）可自然应用于该任务。然而，现有基于LLM的零样本NER方法存在提示和演示工程的局限。为以最少人工干预解决这些问题，我们提出DE-NER，这是一个对话引导框架，利用LLM的对话能力充分挖掘其编码知识。实验表明，该方法在零样本设置下的多个基准中优于竞争基线，平均F1值提升3.75%。代码已在该https URL发布。
英文摘要
Recent advancements of zero-shot Named Entity Recognition (NER) establish strong baselines by formulating sequence labeling into question answering where Large Language Models (LLMs) can be naturally adopted. However, existing LLM-based zero-shot NER methods suffer from the limitations of prompt and demonstration engineering. To address these issues with minimal human interventions, we introduce DE-NER, a dialogue elicitation framework which elicits the chatting ability of LLMs to fully extract the knowledge encoded in LLMs. Our experiments demonstrate that the proposed method outperform the competitive baselines in zero-shot settings across multiple benchmarks, with an average improvement of 3.75\% F1 points. Codes are released in https://github.com/kkkenshi/DE-NER.
