A Multi-Expert Structural-Semantic Hybrid Framework for Unveiling Historical Patterns in Temporal Knowledge Graphs

2025-06-17

AI Summary

Existing temporal knowledge graph (TKG) reasoning methods struggle to jointly optimize structural learning and semantic reasoning, while overlooking the intrinsic differences in temporal patterns between historical and non-historical events—leading to limited generalization. This paper proposes the first structure–semantics collaborative mixture-of-experts framework, dynamically integrating dual perspectives via three specialized experts: a structural expert (graph neural network), a semantic expert (temporal attention), and a historical-pattern expert (event-type-aware modeling). Crucially, it explicitly models the temporal pattern specificity of historical events for the first time. Evaluated on three standard benchmarks, our method achieves an average 9.2% improvement in historical event prediction accuracy and demonstrates significantly enhanced cross-granularity temporal generalization, consistently outperforming state-of-the-art approaches.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Mixture of Experts (MoE)
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

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 neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Temporal knowledge graph reasoning aims to predict future events with knowledge of existing facts and plays a key role in various downstream tasks. Previous methods focused on either graph structure learning or semantic reasoning, failing to integrate dual reasoning perspectives to handle different prediction scenarios. Moreover, they lack the capability to capture the inherent differences between historical and non-historical events, which limits their generalization across different temporal contexts. To this end, we propose a Multi-Expert Structural-Semantic Hybrid (MESH) framework that employs three kinds of expert modules to integrate both structural and semantic information, guiding the reasoning process for different events. Extensive experiments on three datasets demonstrate the effectiveness of our approach.

Problem

• Predict future events in temporal knowledge graphs
• Integrate structural and semantic reasoning perspectives
• Capture differences between historical and non-historical events

Innovation

• Multi-expert hybrid framework for TKG
• Integrates structural and semantic reasoning
• Captures historical event differences