HiTS-CL: A Continual Learning Framework for Long-Horizon Temporal Knowledge Graph Extrapolation
History-enhanced Two-Step Continual Learning (HiTS-CL), a backbone-agnostic continual learning framework for extrapolative TKGR that consistently improves extrapolation accuracy, reduces long-horizon degradation, and outperforms strong continual-learning baselines.
Yan-Song Liu, Rui Liu, Yuan Zuo et al.
· Proceedings of the 35th ACM... · 0 citations