LELA: an LLM-based Entity Linking Approach with Zero-Shot Domain Adaptation

2026-01-08

arXiv.org

AI Summary

This work addresses the challenge of zero-shot entity linking in cross-domain scenarios without fine-tuning by proposing a modular coarse-to-fine reasoning framework based on large language models (LLMs). The approach enables plug-and-play adaptation across diverse domains, knowledge bases, and LLMs entirely in a zero-shot manner, eliminating the need for any model fine-tuning. Evaluated on multiple standard entity linking benchmarks, the method substantially outperforms existing non-fine-tuned approaches and achieves performance comparable to—or even surpassing—that of state-of-the-art models requiring fine-tuning, thereby demonstrating strong generalization capabilities and practical utility.

Technology Category

Machine Learning: Large Multimodal Models (LMMs)Natural Language Processing: Language Grounding & Multi-modal NLPComputer Vision: Large Vision Models

Application Category

Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMsGraph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphsUser Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Entity linking (mapping ambiguous mentions in text to entities in a knowledge base) is a foundational step in tasks such as knowledge graph construction, question-answering, and information extraction. Our method, LELA, is a modular coarse-to-fine approach that leverages the capabilities of large language models (LLMs), and works with different target domains, knowledge bases and LLMs, without any fine-tuning phase. Our experiments across various entity linking settings show that LELA is highly competitive with fine-tuned approaches, and substantially outperforms the non-fine-tuned ones.

Problem

• Entity Linking
• Zero-Shot Domain Adaptation
• Large Language Models
• Knowledge Base
• Ambiguous Mentions

Innovation

• Entity Linking
• Large Language Models
• Zero-Shot Domain Adaptation
• Coarse-to-Fine
• Knowledge Base