Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognition

TL;DR

This paper introduces B2NERD, a curated dataset that standardizes entity definitions across multiple datasets and languages, significantly improving Large Language Models' ability to perform open NER across diverse domains.

Contribution

The creation of B2NERD, a universal entity taxonomy dataset that enhances LLMs' generalization in open NER by resolving inconsistencies and redundancies in existing datasets.

Findings

B2NERD improves LLMs' open NER performance by 6.8-12.0 F1 points.

Models trained on B2NERD outperform GPT-4 in out-of-domain benchmarks.

B2NERD covers over 400 entity types across 15 datasets and 6 languages.

Abstract

Open Named Entity Recognition (NER), which involves identifying arbitrary types of entities from arbitrary domains, remains challenging for Large Language Models (LLMs). Recent studies suggest that fine-tuning LLMs on extensive NER data can boost their performance. However, training directly on existing datasets neglects their inconsistent entity definitions and redundant data, limiting LLMs to dataset-specific learning and hindering out-of-domain adaptation. To address this, we present B2NERD, a compact dataset designed to guide LLMs' generalization in Open NER under a universal entity taxonomy. B2NERD is refined from 54 existing English and Chinese datasets using a two-step process. First, we detect inconsistent entity definitions across datasets and clarify them by distinguishable label names to construct a universal taxonomy of 400+ entity types. Second, we address redundancy using a data pruning strategy that selects fewer samples with greater category and semantic diversity. Comprehensive evaluation shows that B2NERD significantly enhances LLMs' Open NER capabilities. Our B2NER models, trained on B2NERD, outperform GPT-4 by 6.8-12.0 F1 points and surpass previous methods in 3 out-of-domain benchmarks across 15 datasets and 6 languages. The data, models, and code are publicly available at https://github.com/UmeanNever/B2NER.