GAttNHP: Group Attention Neural Hawkes Process for Extrapolation Reasoning in Temporal Knowledge Graphs
For researchers working on temporal knowledge graph reasoning, GAttNHP provides a unified solution that significantly improves both entity and time prediction, especially for rare events.
GAttNHP addresses three key challenges in temporal knowledge graph forecasting—long-range dependencies, cross-chain excitation, and heavy-tailed inter-arrival times—by combining a self-attention encoder, a semantic soft-grouping module, and a Non-Crossing Quantile regression head. It achieves state-of-the-art results on six benchmarks, with largest gains on long-tail event chains.
Temporal Knowledge Graphs (TKGs) record how facts evolve over time, but forecasting future events on a TKG remains difficult for three reasons: (i) long-range temporal dependencies are hard to encode; (ii) events on different chains mutually excite or inhibit one another in ways that snapshot-level models cannot express; and (iii) inter-arrival times are heavy-tailed and statistically sparse, so deterministic time predictors are unreliable. We address these three issues with a single framework, the \textbf{Group Attention Neural Hawkes Process (GAttNHP)}, built around three matched components. First, a self-attention encoder casts each subject--relation chain as a continuous-time point process and captures the lingering excitation of distant history. Second, a semantic soft-grouping module turns globally learnable Hawkes priors into an analytical cross-attention mask, so chains share excitation patterns through their latent group memberships rather than through exhaustive pairwise computation. Third, a Non-Crossing Quantile (NCQ) regression head replaces mean-based time prediction, providing calibrated, monotonically ordered quantile estimates that remain stable under heavy-tailed inter-arrival distributions. On six benchmark TKG datasets, GAttNHP improves over state-of-the-art baselines on both entity prediction and time prediction, and ablations confirm that its largest gains arise on the long-tail event chains where existing models fail most severely.