Breaking the Reasoning Horizon in Entity Alignment Foundation Models

TL;DR

This paper introduces a novel entity alignment foundation model that effectively captures long-range dependencies in knowledge graphs by using parallel encoding and local anchors, improving transferability to unseen graphs.

Contribution

It proposes a parallel encoding strategy with anchor-guided message passing and a merged relation graph to address the reasoning horizon gap in entity alignment models.

Findings

Significantly improves alignment accuracy on benchmark datasets.

Demonstrates strong generalizability to unseen knowledge graphs.

Reduces inference complexity by leveraging local structural proximity.

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.