Breaking the Reasoning Horizon in Entity Alignment Foundation Models

2026-01-29

AI Summary

This work addresses the limited generalization of existing entity alignment methods on unseen knowledge graphs, primarily due to their inability to capture long-range dependencies in sparse and heterogeneous graph structures—a challenge termed the “reasoning horizon gap.” To bridge this gap, the authors propose a local anchor-driven parallel encoding foundation model. Leveraging seed alignments as anchors, the model employs dual-stream parallel encoding with anchor-conditioned message passing and integrates a unified relation graph to model global dependencies. A learnable interaction module further refines entity matching. This design substantially shortens reasoning paths and circumvents the limitations of conventional global search strategies. Extensive experiments demonstrate that the proposed method significantly outperforms state-of-the-art approaches across multiple benchmark datasets, exhibiting strong cross-graph generalization capability.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Search and Optimization: Metareasoning and Metaheuristics
• Machine Learning: Graph-based Machine Learning

Application Category

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

Abstract

Entity alignment (EA) is critical for knowledge graph (KG) fusion. Existing EA models lack transferability and are incapable of aligning unseen KGs without retraining. While using graph foundation models (GFMs) offer a solution, we find that directly adapting GFMs to EA remains largely ineffective. This stems from a critical"reasoning horizon gap": unlike link prediction in GFMs, EA necessitates capturing long-range dependencies across sparse and heterogeneous KG structuresTo address this challenge, we propose a EA foundation model driven by a parallel encoding strategy. We utilize seed EA pairs as local anchors to guide the information flow, initializing and encoding two parallel streams simultaneously. This facilitates anchor-conditioned message passing and significantly shortens the inference trajectory by leveraging local structural proximity instead of global search. Additionally, we incorporate a merged relation graph to model global dependencies and a learnable interaction module for precise matching. Extensive experiments verify the effectiveness of our framework, highlighting its strong generalizability to unseen KGs.

Problem

• Entity Alignment
• Knowledge Graph
• Reasoning Horizon
• Graph Foundation Models
• Long-range Dependencies

Innovation

• Entity Alignment
• Graph Foundation Models
• Reasoning Horizon
• Parallel Encoding
• Anchor-Conditioned Message Passing