Multi-Faceted Continual Knowledge Graph Embedding for Semantic-Aware Link Prediction

2026-04-12

AI Summary

Existing continual knowledge graph embedding methods conflate old and new knowledge within a shared embedding space, struggling to capture the multifaceted semantic evolution of entities over time. This often leads to catastrophic forgetting and degraded link prediction performance. To address these limitations, this work proposes MF-CKGE, a multifaceted continual knowledge graph embedding framework that introduces, for the first time, a multifaceted semantic modeling mechanism. During the offline phase, it achieves independent storage of temporal knowledge through multi-space embedding separation and semantic disentanglement; during the online phase, it adaptively selects relevant semantic embeddings based on query-aware cues. This approach effectively mitigates knowledge entanglement and semantic redundancy, yielding consistent improvements across eight benchmark datasets—with average gains of 1.7% in MRR and 1.4% in Hits@10, and peak improvements reaching 2.7% and 3.8%, respectively.

Technology Category

• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• Machine Learning: Life-Long and Continual Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

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: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Continual Knowledge Graph Embedding (CKGE) aims to continually learn embeddings for new knowledge, i.e., entities and relations, while retaining previously acquired knowledge. Most existing CKGE methods mitigate catastrophic forgetting via regularization or replaying old knowledge. They conflate new and old knowledge of an entity within the same embedding space to seek a balance between them. However, entities inherently exhibit multi-faceted semantics that evolve dynamically as their relational contexts change over time. A shared embedding fails to capture and distinguish these temporal semantic variations, degrading lifelong link prediction accuracy across snapshots. To address this, we propose a Multi-Faceted CKGE framework (MF-CKGE) for semantic-aware link prediction. During offline learning, MF-CKGE separates temporal old and new knowledge into distinct embedding spaces to prevent knowledge entanglement and employs semantic decoupling to reduce semantic redundancy, thereby improving space efficiency. During online inference, MF-CKGE adaptively identifies semantically query-relevant entity embeddings by quantifying their semantic importance, reducing interference from query-irrelevant noise. Experiments on eight datasets show that MF-CKGE achieves an average (maximum) improvement of 1.7% (2.7%) and 1.4% (3.8%) in MRR and Hits@10, respectively, over the best baseline. Our source code and datasets are available at: https://anonymous.4open.science/r/MF-CKGE-04E5.

Problem

• Continual Knowledge Graph Embedding
• Multi-Faceted Semantics
• Catastrophic Forgetting
• Semantic-Aware Link Prediction
• Temporal Knowledge Graph

Innovation

• Continual Knowledge Graph Embedding
• Multi-Faceted Semantics
• Semantic Decoupling
• Link Prediction
• Catastrophic Forgetting