ReverseNER: A Self-Generated Example-Driven Framework for Zero-Shot Named Entity Recognition with Large Language Models

2024-11-01

arXiv.org

AI Summary

Zero-shot named entity recognition (NER) with large language models (LLMs) suffers from heavy reliance on manually crafted demonstrations, poor cross-domain generalization, and high computational overhead. Method: We propose the first NER reverse-generation paradigm: (1) generating prototypical entities directly from entity type definitions; (2) extracting highly discriminative syntactic patterns via semantic clustering to automatically construct a high-quality labeled sentence corpus; and (3) introducing an entity-level self-consistency scoring mechanism to enhance sample reliability. Our approach integrates sentence-structure copying, example retrieval, and self-consistent reasoning—requiring no human annotation or model fine-tuning. Contribution/Results: Experiments demonstrate significant improvements over state-of-the-art LLM-based zero-shot NER baselines across multiple domains, achieving stronger domain generalization while substantially reducing inference resource consumption.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Large Multimodal Models (LMMs)
• Knowledge Representation and Reasoning: Common-Sense Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

This paper presents ReverseNER, a method aimed at overcoming the limitation of large language models (LLMs) in zero-shot named entity recognition (NER) tasks, arising from their reliance on pre-provided demonstrations. ReverseNER tackles this challenge by constructing a reliable example library composed of dozens of entity-labeled sentences, generated through the reverse process of NER. Specifically, while conventional NER methods label entities in a sentence, ReverseNER features reversing the process by using an LLM to generate entities from their definitions and subsequently expand them into full sentences. During the entity expansion process, the LLM is guided to generate sentences by replicating the structures of a set of specific extsl{feature sentences}, extracted from the task sentences by clustering. This expansion process produces dozens of entity-labeled task-relevant sentences. After constructing the example library, the method selects several semantically similar entity-labeled examples for each task sentence as references to facilitate the LLM's entity recognition. We also propose an entity-level self-consistency scoring mechanism to improve NER performance with LLMs. Experiments show that ReverseNER significantly outperforms other zero-shot NER methods with LLMs, marking a notable improvement in NER for domains without labeled data, while declining computational resource consumption.

Problem

• Large Language Models
• Entity Recognition
• Zero-shot Learning

Innovation

• ReverseNER
• Zero-shot Learning
• Large-scale Language Model