Label Alignment and Reassignment with Generalist Large Language Model for Enhanced Cross-Domain Named Entity Recognition

TL;DR

This paper introduces LAR, a label alignment and reassignment method leveraging large language models like ChatGPT to improve cross-domain named entity recognition, especially in zero-shot scenarios.

Contribution

The study proposes a novel label alignment and reassignment approach that effectively addresses label conflicts in cross-domain NER using large language models.

Findings

LAR outperforms state-of-the-art methods in supervised cross-domain NER.

LAR achieves significant improvements in zero-shot NER performance.

Experimental results validate the effectiveness of integrating large language models for label reassignment.

Abstract

Named entity recognition on the in-domain supervised and few-shot settings have been extensively discussed in the NLP community and made significant progress. However, cross-domain NER, a more common task in practical scenarios, still poses a challenge for most NER methods. Previous research efforts in that area primarily focus on knowledge transfer such as correlate label information from source to target domains but few works pay attention to the problem of label conflict. In this study, we introduce a label alignment and reassignment approach, namely LAR, to address this issue for enhanced cross-domain named entity recognition, which includes two core procedures: label alignment between source and target domains and label reassignment for type inference. The process of label reassignment can significantly be enhanced by integrating with an advanced large-scale language model such as ChatGPT. We conduct an extensive range of experiments on NER datasets involving both supervised and zero-shot scenarios. Empirical experimental results demonstrate the validation of our method with remarkable performance under the supervised and zero-shot out-of-domain settings compared to SOTA methods.