Historically Relevant Event Structuring for Temporal Knowledge Graph Reasoning

TL;DR

This paper tackles extrapolatory reasoning in Temporal Knowledge Graphs by explicitly modeling historically relevant events. It introduces HisRES, an encoder-decoder framework with a multi-granularity evolutionary encoder for recent snapshots and a global relevance encoder powered by ConvGAT to capture query-relevant historical signals, merged through a self-gating mechanism. Through extensive experiments on ICEWS and GDELT datasets, HisRES achieves state-of-the-art results and demonstrates that both local historical evolution and distant global relevance are crucial for accurate future event prediction. The work highlights the practical impact of structuring historical relevance in TKG reasoning and offers a scalable approach to incorporate rich historical dependencies.

Abstract

Temporal Knowledge Graph (TKG) reasoning focuses on predicting events through historical information within snapshots distributed on a timeline. Existing studies mainly concentrate on two perspectives of leveraging the history of TKGs, including capturing evolution of each recent snapshot or correlations among global historical facts. Despite the achieved significant accomplishments, these models still fall short of I) investigating the impact of multi-granular interactions across recent snapshots, and II) harnessing the expressive semantics of significant links accorded with queries throughout the entire history, particularly events exerting a profound impact on the future. These inadequacies restrict representation ability to reflect historical dependencies and future trends thoroughly. To overcome these drawbacks, we propose an innovative TKG reasoning approach towards \textbf{His}torically \textbf{R}elevant \textbf{E}vents \textbf{S}tructuring (HisRES). Concretely, HisRES comprises two distinctive modules excelling in structuring historically relevant events within TKGs, including a multi-granularity evolutionary encoder that captures structural and temporal dependencies of the most recent snapshots, and a global relevance encoder that concentrates on crucial correlations among events relevant to queries from the entire history. Furthermore, HisRES incorporates a self-gating mechanism for adaptively merging multi-granularity recent and historically relevant structuring representations. Extensive experiments on four event-based benchmarks demonstrate the state-of-the-art performance of HisRES and indicate the superiority and effectiveness of structuring historical relevance for TKG reasoning.

Figures (11)

• Figure 1: Example of extrapolation over TKGs. Left: original historical snapshots containing timestamped facts. Right: derived graphs structured through multi-granularity correlation (red elements) and global relevance (blue elements) perspectives.
• Figure 2: An illustration of HisRES model architecture.
• Figure 3: The achitecture of proposed multi-granularity evolutionary encoder.
• Figure 4: An illustrative construction of the globally relevant graph. In the historical TKGs, query-specific facts are framed by green ovals, while the globally relevant graph is structured by all those facts.
• Figure 5: The influence of granularity spans from each single snapshot to every three adjacent snapshots.