Xiaoshi Zhong

Xiaoshi Zhong, Xiang Yu, Erik Cambria et al.

Entities involve important concepts with concrete meanings and play important roles in numerous linguistic tasks. Entities have different forms in different linguistic tasks and researchers treat those different forms as different concepts. In this paper, we are curious to know whether there are some common characteristics that connect those different forms of entities. Specifically, we investigate the underlying distributions of entities from different types and different languages, trying to figure out some common characteristics behind those diverse entities. After analyzing twelve datasets about different types of entities and eighteen datasets about entities in different languages, we find that while these entities are dramatically diverse from each other in many aspects, their length-frequencies can be well characterized by a family of Marshall-Olkin power-law (MOPL) distributions. We conduct experiments on those thirty datasets about entities in different types and different languages, and experimental results demonstrate that MOPL models characterize the length-frequencies of entities much better than two state-of-the-art power-law models and an alternative log-normal model. Experimental results also demonstrate that MOPL models are scalable to the length-frequency of entities in large-scale real-world datasets.

Yufei Zhao, Xiaoshi Zhong, Erik Cambria et al.

Named entity recognition (NER) is a fundamental task in numerous downstream applications. Recently, researchers have employed pre-trained language models (PLMs) and large language models (LLMs) to address this task. However, fully leveraging the capabilities of PLMs and LLMs with minimal human effort remains challenging. In this paper, we propose GPT4NER, a method that prompts LLMs to resolve the few-shot NER task. GPT4NER constructs effective prompts using three key components: entity definition, few-shot examples, and chain-of-thought. By prompting LLMs with these effective prompts, GPT4NER transforms few-shot NER, which is traditionally considered as a sequence-labeling problem, into a sequence-generation problem. We conduct experiments on two benchmark datasets, CoNLL2003 and OntoNotes5.0, and compare the performance of GPT4NER to representative state-of-the-art models in both few-shot and fully supervised settings. Experimental results demonstrate that GPT4NER achieves the $F_1$ of 83.15\% on CoNLL2003 and 70.37\% on OntoNotes5.0, significantly outperforming few-shot baselines by an average margin of 7 points. Compared to fully-supervised baselines, GPT4NER achieves 87.9\% of their best performance on CoNLL2003 and 76.4\% of their best performance on OntoNotes5.0. We also utilize a relaxed-match metric for evaluation and report performance in the sub-task of named entity extraction (NEE), and experiments demonstrate their usefulness to help better understand model behaviors in the NER task.