LinkNER: Linking Local Named Entity Recognition Models to Large Language Models using Uncertainty

arXiv:2402.10573v1 Announce Type: new
Abstract: Named Entity Recognition (NER) serves as a fundamental task in natural language understanding, bearing direct implications for web content analysis, search engines, and information retrieval systems. Fine-tuned NER models exhibit satisfactory performance on standard NER benchmarks. However, due to limited fine-tuning data and lack of knowledge, it performs poorly on unseen entity recognition. As a result, the usability and reliability of NER models in web-related applications are compromised. Instead, Large Language Models (LLMs) like GPT-4 …

abstract analysis arxiv benchmarks cs.cl data fine-tuning information language language models language understanding large language large language models natural natural language ner performance recognition retrieval search standard systems type uncertainty understanding web