DynaGen: Unifying Temporal Knowledge Graph Reasoning with Dynamic Subgraphs and Generative Regularization

2025-12-14

AI Summary

Existing time-aware knowledge graph reasoning (TKGR) methods suffer from insufficient temporal context modeling for fact interpolation (historical completion) and generalization bias for extrapolation (future prediction). This paper proposes a unified framework: first, it constructs dynamic entity-centric subgraphs and employs a collaborative dual-branch GNN to enable fine-grained temporal contextual modeling; second, it introduces a conditional diffusion-based generative regularization mechanism that compels the model to learn intrinsic event evolution laws rather than spurious statistical patterns, thereby unifying interpolation and extrapolation in a principled manner. The approach integrates dynamic subgraph sampling, snapshot sequence modeling, and temporal embedding. Evaluated on six benchmark datasets, it achieves state-of-the-art performance—improving average MRR by 2.61 for interpolation and 1.45 for extrapolation—demonstrating significantly enhanced long-horizon temporal generalization and zero-shot event prediction capability.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Temporal Planning
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion

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

Temporal Knowledge Graph Reasoning (TKGR) aims to complete missing factual elements along the timeline. Depending on the temporal position of the query, the task is categorized into interpolation and extrapolation. Existing interpolation methods typically embed temporal information into individual facts to complete missing historical knowledge, while extrapolation techniques often leverage sequence models over graph snapshots to identify recurring patterns for future event prediction. These methods face two critical challenges: limited contextual modeling in interpolation and cognitive generalization bias in extrapolation. To address these, we propose a unified method for TKGR, dubbed DynaGen. For interpolation, DynaGen dynamically constructs entity-centric subgraphs and processes them with a synergistic dual-branch GNN encoder to capture evolving structural context. For extrapolation, it applies a conditional diffusion process, which forces the model to learn underlying evolutionary principles rather than just superficial patterns, enhancing its ability to predict unseen future events. Extensive experiments on six benchmark datasets show DynaGen achieves state-of-the-art performance. On average, compared to the second-best models, DynaGen improves the Mean Reciprocal Rank (MRR) score by 2.61 points for interpolation and 1.45 points for extrapolation.

Problem

• Unifies interpolation and extrapolation in temporal knowledge graph reasoning
• Addresses limited contextual modeling in interpolation tasks
• Mitigates cognitive generalization bias in extrapolation tasks

Innovation

• Dynamic entity-centric subgraphs with dual-branch GNN encoder
• Conditional diffusion process for learning evolutionary principles
• Unified method for both interpolation and extrapolation tasks