Sep 2026· Proceedings of the 35th ACM International Conference on Information and Knowledge Management· pp. 3305-3316· 0 citations· 28 references
Computer Science
TL;DR
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.
Abstract
Extrapolative temporal knowledge graph reasoning (TKGR) predicts future facts from historical snapshots. Most existing methods train once on an early prefix of the timeline and then use a frozen model for all future timestamps. We argue that this fixed-prefix protocol is misaligned with extrapolation. It learns from a static prefix, whereas the target stream is non-stationary: new entities and facts emerge, temporal dependencies shift across regimes, and recurring historical signals must be refreshed online. As a result, models trained only on early snapshots become outdated and degrade over long horizons. We address this mismatch by formulating extrapolative TKGR as continual learning over streaming snapshots. Under this view, effective extrapolation must jointly handle current dynamics, stable knowledge, and recurring historical evidence. Based on these requirements, we propose History-enhanced Two-Step Continual Learning (HiTS-CL), a backbone-agnostic continual learning framework for extrapolative TKGR. HiTS-CL tracks current dynamics via continual fine-tuning, preserves stable knowledge via multi-teacher adaptive distillation, and retains recurring historical evidence via a selective memory of recent and frequent facts. We integrate HiTS-CL into five representative TKGR backbones and evaluate it on four benchmark datasets. HiTS-CL consistently improves extrapolation accuracy, reduces long-horizon degradation, and outperforms strong continual-learning baselines, including a recent method for temporal knowledge graphs. Source code and data are available at https://github.com/liuyansong98/HiTS-CL.
DyHist is proposed, a query-aware historical dynamics modeling framework for TKG extrapolation that captures local dynamics by separating recent evidence from medium-range contexts and explicitly modeling their query-relevant deviations as predictive features.
Pei-Yuan Jiang, Yao Liu, Yu-Bo Wen et al.· Conference on Empirical Meth...· 0 citations
Temporal knowledge graph (TKG) reasoning involves predicting future facts by identifying dynamic evolutionary patterns from historical data. The foundation of effective reasoning lies in the comprehensive identification and analysis of the temporal regularities inherent in the given facts. However, existing approache...
Jia-Bin Zhang, Min Wang, Guo-Qiang Xie et al.· PeerJ Computer Science· 0 citations
Temporal Evidence-Parameter Co-Alignment (TEPCA), a framework for LLM reasoning over TKGs that aligns LLM reasoning with temporal knowledge at both the evidence and parameter levels, and enables LLMs to better capture fine-grained entity evolution and time-dependent relation semantics while preserving parameter efficie...
Event knowledge graphs (EKGs) must evolve continuously as new events emerge, while preserving temporal integrity and structural consistency. However, existing triple-completion and embedding approaches assume closed-world settings and lack mechanisms for open-world alignment and interpretable updates. Furthermore, real...
Chun-Yan An, Zhi-Hong Hao, Hui-Cheng Yin et al.· Proceedings of the 35th ACM...· 0 citations
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
MIT News · Artificial Intelligence· news.mit.eduSep 29, 2026
Professor Sherry Turkle’s new book, “Artificial Intimacy,” offers a withering critique of chatbots and the antisocial dynamics she believes they encourage.