Debate to Align: Reliable Entity Alignment through Two-Stage Multi-Agent Debate

2026-04-15

AI Summary

This work addresses the challenges of unreliable candidate sets and limited reasoning capabilities of large language models in cross-knowledge graph entity alignment by proposing AgentEA, a multi-agent debate framework. The approach enhances embedding quality through optimized entity representation preferences and introduces a novel two-stage debate mechanism featuring lightweight verification followed by deep alignment among multiple specialized agents. This iterative process progressively improves the reliability, efficiency, and interpretability of alignment decisions. Extensive experiments across diverse public benchmarks—including cross-lingual, sparse, large-scale, and heterogeneous settings—demonstrate that AgentEA significantly outperforms state-of-the-art methods, confirming its effectiveness and strong generalization capability.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Multiagent Systems: Agreement, Argumentation & Negotiation
• Knowledge Representation and Reasoning: Argumentation

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Entity alignment (EA) aims to identify entities referring to the same real-world object across different knowledge graphs (KGs). Recent approaches based on large language models (LLMs) typically obtain entity embeddings through knowledge representation learning and use embedding similarity to identify an alignment-uncertain entity set. For each uncertain entity, a candidate entity set (CES) is then retrieved based on embedding similarity to support subsequent alignment reasoning and decision making. However, the reliability of the CES and the reasoning capability of LLMs critically affect the effectiveness of subsequent alignment decisions. To address this issue, we propose AgentEA, a reliable EA framework based on multi-agent debate. AgentEA first improves embedding quality through entity representation preference optimization, and then introduces a two-stage multi-role debate mechanism consisting of lightweight debate verification and deep debate alignment to progressively enhance the reliability of alignment decisions while enabling more efficient debate-based reasoning. Extensive experiments on public benchmarks under cross-lingual, sparse, large-scale, and heterogeneous settings demonstrate the effectiveness of AgentEA.

Problem

• Entity Alignment
• Large Language Models
• Candidate Entity Set
• Reliability
• Knowledge Graphs

Innovation

• entity alignment
• multi-agent debate
• large language models
• two-stage reasoning
• knowledge graph