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

2026/01/31 by Samy Haffoudhi, Fabian M. Suchanek, Nils Holzenberger · 3 citations
Computer Science · #cs.CL

paper · pdf · doi:10.48550/arxiv.2601.05192

Accepted at ISWC 2026. Extended version with appendices

arxiv created 2026/08/06 · arxiv updated 2026/08/07

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.

Citations

• Leveraging the Power of Large Language Models in Entity Linking via Adaptive Routing and Targeted Reasoning
• Contextual Augmentation for Entity Linking using Large Language Models
• LongCodeZip: Compress Long Context for Code Language Models
• jina-reranker-v3: Last but Not Late Interaction for Listwise Document Reranking
• zELO: ELO-inspired Training Method for Rerankers and Embedding Models
• DeAR: Dual-Stage Document Reranking with Reasoning Agents via LLM Distillation
• DeepMEL: A Multi-Agent Collaboration Framework for Multimodal Entity Linking
• ReasonRank: Empowering Passage Ranking with Strong Reasoning Ability
• JointRank: Rank Large Set with Single Pass
• Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models
• Don't "Overthink" Passage Reranking: Is Reasoning Truly Necessary?
• Evaluating Design Decisions for Dual Encoder-based Entity Disambiguation
• Between Underthinking and Overthinking: An Empirical Study of Reasoning Length and correctness in LLMs
• CoRanking: Collaborative Ranking with Small and Large Ranking Agents
• Rank-R1: Enhancing Reasoning in LLM-based Document Rerankers via Reinforcement Learning
• OneNet: A Fine-Tuning Free Framework for Few-Shot Entity Linking via Large Language Model Prompting
• Attention in Large Language Models Yields Efficient Zero-Shot Re-Rankers
• UniMEL: A Unified Framework for Multimodal Entity Linking with Large Language Models
• LLMAEL: Large Language Models are Good Context Augmenters for Entity Linking
• TourRank: Utilizing Large Language Models for Documents Ranking with a Tournament-Inspired Strategy
• Match, Compare, or Select? An Investigation of Large Language Models for Entity Matching
• Entity Disambiguation via Fusion Entity Decoding
• Entity Linking in the Job Market Domain
• The Faiss library
• A Read-and-Select Framework for Zero-shot Entity Linking
• Efficient Memory Management for Large Language Model Serving with PagedAttention
• YAGO 4.5: A Large and Clean Knowledge Base with a Rich Taxonomy
• Lost in the Middle: How Language Models Use Long Contexts
• Large Language Models are Effective Text Rankers with Pairwise Ranking Prompting
• Generative Multimodal Entity Linking
• GLADIS: A General and Large Acronym Disambiguation Benchmark
• Design of Negative Sampling Strategies for Distantly Supervised Skill Extraction
• ELEVANT: A Fully Automatic Fine-Grained Entity Linking Evaluation and Analysis Tool
• ReFinED: An Efficient Zero-shot-capable Approach to End-to-End Entity Linking
• Effective Few-Shot Named Entity Linking by Meta-Learning
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• Highly Parallel Autoregressive Entity Linking with Discriminative Correction
• LoRA: Low-Rank Adaptation of Large Language Models
• Multilingual Autoregressive Entity Linking
• Autoregressive Entity Retrieval
• Machine Knowledge: Creation and Curation of Comprehensive Knowledge Bases
• Language Models are Few-Shot Learners
• Empirical Evaluation of Pretraining Strategies for Supervised Entity Linking
• Fine-Grained Entity Typing for Domain Independent Entity Linking
• Global Entity Disambiguation with BERT
• Zero-Shot Entity Linking by Reading Entity Descriptions
• Named Entity Disambiguation for Noisy Text
• Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer
• Sequence-Level Knowledge Distillation
• Distilling the Knowledge in a Neural Network
• ESCO: Boosting Job Matching in Europe with Semantic Interoperability
• Relevance weighting of search terms