HiTS-CL: A Continual Learning Framework for Long-Horizon Temporal Knowledge Graph Extrapolation

Yansong Liu; Rui Liu; Yuan Zuo; Hongwei Zhao; Da Fu; Fuwei Zhang; Fuzhen Zhuang; Yong Chen

cs.LGSep 29, 2026

HiTS-CL: A Continual Learning Framework for Long-Horizon Temporal Knowledge Graph Extrapolation

Authors: Yansong Liu, Rui Liu, Yuan Zuo, Hongwei Zhao, Da Fu, Fuwei Zhang, Fuzhen Zhuang, Yong Chen, +1 more

Organizations: Beihang University Beijing, China · Beijing University of Posts and Telecommunications Beijing, China · Hubei Engineering University Xiaogan, Hubei, China

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 As a result, models trained only on early snapshots become outdated and degrade over long horizons.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.

Figures & tables

Panel of Figure 1 — Figure 1 . (a) Long-horizon performance decay of a fixed-prefix backbone versus HiTS-CL on the test stream (MRR averaged over 100 snapshots). Dashed/solid curves denote the backbone and its HiTS-CL variant. (b) Fixed-prefix training versus continual updating evaluated on the immediately following test snapshot after the training period (snapshot 1001; train: 1–1000).

Figure 2 . Overview of HiTS-CL: continual fine-tuning captures current dynamics, multi-teacher distillation preserves stable knowledge while incorporating time-varying patterns, and selective-memory history enhancement provides recurring historical evidence.

Dataset	Entities	Relations	Snapshots	Train	Valid	Test	Test Snapshots	New Entities	Granularity
ICEWS14	7,218	230	365	27,391	4,514	58,825	231	2,149	24 hours
ICEWS18	23,033	256	304	140,714	23,320	304,524	203	5,082	24 hours
ICEWS05-15	10,488	251	4,017	138,576	22,959	299,794	2,725	3,289	24 hours
GDELT	7,691	240	2,751	683,646	114,076	1,480,683	1,688	1,304	15 mins

Table 1 . Statistics of the datasets. All facts are chronologically split into training, validation, and test sets.

Model	ICEWS14				ICEWS18				ICEWS05-15				GDELT
Model	MRR	H@1	H@3	H@10	MRR	H@1	H@3	H@10	MRR	H@1	H@3	H@10	MRR	H@1	H@3	H@10
RE-GCN (2021)	33.81	25.66	37.48	49.47	28.41	19.33	32.00	46.23	44.29	34.76	49.07	62.65	18.41	11.56	19.45	31.76
RE-GCN+FT	40.00	30.91	44.61	57.13	31.16	21.48	35.22	50.12	49.75	39.51	55.35	69.18	21.79	13.94	23.56	37.13
RE-GCN+Replay	41.29	32.16	45.86	58.61	32.48	22.53	36.75	51.93	51.64	41.21	57.42	71.31	21.75	13.86	23.46	37.21
RE-GCN+Reg	40.10	31.10	44.55	57.02	31.67	21.75	35.77	51.23	52.38	41.99	58.09	72.12	21.10	13.42	22.72	36.09
RE-GCN+HiTS-CL	43.61	34.61	48.48	60.13	34.34	24.44	38.78	53.52	52.53	42.45	58.18	71.56	25.22	16.51	27.83	42.28

Table 2 . Time-aware filtered MRR and Hits@K on ICEWS14/18, ICEWS05-15, and GDELT. We report each backbone under fixed-prefix training and continual learning variants (+FT, +Replay, +Reg, +HiTS-CL).

Panel of Figure 3 — Figure 3 . MRR degradation trends over time of ICEWS05-15.

Figure 4 . Comparison of DGAR and LogCL+HiTS-CL.

Panel of Figure 5 — Figure 5 . Fixed-prefix training versus continual updating under the same evaluation snapshot (MRR).

Model	ICEWS14				ICEWS18				ICEWS05-15				GDELT
Model	MRR	H@1	H@3	H@10	MRR	H@1	H@3	H@10	MRR	H@1	H@3	H@10	MRR	H@1	H@3	H@10
RE-GCN w/o St.1	42.11	33.46	46.61	58.04	33.69	24.01	38.05	52.45	51.23	41.55	56.71	69.28	23.91	15.44	26.36	40.51
RE-GCN w/o St.2	42.47	33.60	47.33	58.74	33.54	23.83	37.91	52.32	51.59	41.51	57.24	70.58	23.09	16.43	27.69	42.03
RE-GCN w/o His.	41.52	32.36	46.19	58.60	32.20	22.36	36.31	51.54	51.19	41.03	56.69	70.46	22.07	14.14	23.85	37.59
RE-GCN+HiTS-CL	43.61	34.61	48.48	60.13	34.34	24.44	38.78	53.52	52.53	42.45	58.18	71.56	25.22	16.51	27.83	42.28
CEN w/o St.1	39.58	31.10	44.33	54.94	31.95	22.20	36.33	50.93	46.80	37.38	52.24	64.22	21.07	13.05	22.93	36.86

Table 3 . Ablation study on ICEWS14, ICEWS18, ICEWS05-15 and GDELT. All metrics are time-aware filtered.

Backbone	Validation	ICEWS14	ICEWS05-15	GDELT	ICEWS18
DiMNet	$G_{t-1}$	41.50	53.08	24.61	32.82
DiMNet	$G_{t}$ (15%)	41.03	52.49	24.36	32.69
RE-GCN	$G_{t-1}$	43.61	52.53	25.22	34.34
RE-GCN	$G_{t}$ (15%)	43.29	51.48	25.14	33.43

Table 4 . Effect of validation strategy on two backbone models (MRR). Using

G_{t-1}

for validation consistently performs slightly better than holding out 15%

G_{t}

	ICEWS14	ICEWS05-15	ICEWS18	GDELT
Runtime Per snapshot (s)	0.064	0.032	0.408	0.297
Memory usage (MB)	134.18	339.35	535.66	965.93

Table 5 . Inference-time computational cost of the History Enhancement (His) module. We report the average runtime per snapshot and the memory usage of the historical cache on four datasets.

Figure 6 . Runtime comparison (in seconds) between full training and continual learning.

Model	Pretrain	Fine-Tune	Distillation	Overall
DiMNet	620	279	492	1,391
LogCL	1,618	736	1,153	3,507
RE-GCN	526	183	527	1,236

Table 6 . Training time breakdown of HiTS-CL (seconds) on ICEWS14, including pretraining, continual fine-tuning, and distillation stages.

Query	(Media_(China), Make_statement, ?, 299)
HiTS-CL w/o His. (Top-5)	Japan, South_Korea, Kim_Jong-Un, China , North_Korea
HiTS-CL (Top-5)	China , Japan, South_Korea, Kim_Jong-Un, North_Korea
Historical Entities	China (freq=11, latest=261) , Japan (freq=1, latest=222)

Table 7 . Case studies illustrating the effect of the History Enhancement module. Entities in bold denote the ground-truth entity.

Panel of Figure 7 — Figure 7 . Sensitivity analysis of $\alpha$ and $\beta$ hyper-parameters.

Explore similar work

OntoTKGE: Ontology-Enhanced Temporal Knowledge Graph Extrapolation
Apr 7, 2026Dongying Lin, Yinan Liu, Shengwei tang +2Temporal Knowledge Graphs Extrapolation
Northeastern University, Shenyang, China
GAttNHP: Group Attention Neural Hawkes Process for Extrapolation Reasoning in Temporal Knowledge Graphs
Jul 16, 2026Xiangni Tian, Kaixian Yu, Runpeng Dai +2Temporal Knowledge Graphs Long-Range Temporal Dependencies
Yunnan Key Laboratory of Statistical Modeling and Data Analysis, Yunnan University, Kunming, China · Insilicom LLC, Peabody, MA, USA · University of North Carolina at Chapel Hill, North Carolina, USA
AdaTKG: Adaptive Memory for Temporal Knowledge Graph Reasoning
May 8, 2026Seunghan Lee, Jun Seo, Jaehoon Lee +7Temporal Knowledge Graphs Graph-Structured Memories
LG AI Research Seoul, South Korea

Apr 7, 2026, cs.AIJ/K move · Enter open · S save

OntoTKGE: Ontology-Enhanced Temporal Knowledge Graph Extrapolation

Dongying Lin, Yinan Liu, Shengwei tang, Bin Wang+1

Northeastern University, Shenyang, China

Temporal Knowledge Graphs Extrapolation Text Embeddings Interaction History Decoder Layers

Open paper AskPDF arXiv

Temporal knowledge graph (TKG) extrapolation is an important task that aims to predict future facts through historical interaction information within KG snapshots. A key challenge for most existing TKG extrapolation models is handling entities with sparse historical interaction. The ontological knowledge is beneficial for alleviating this sparsity issue by enabling these entities to inherit behavioral patterns from other entities with the same concept, which is ignored by previous studies. In this paper, we propose a novel encoder-decoder framework OntoTKGE that leverages the ontological knowledge from the ontology-view KG (i.e., a KG modeling hierarchical relations among abstract concepts as well as the connections between concepts and entities) to guide the TKG extrapolation model's learning process through the effective integration of the ontological and temporal knowledge, thereby enhancing entity embeddings. OntoTKGE is flexible enough to adapt to many TKG extrapolation models. Extensive experiments on five data sets demonstrate that OntoTKGE not only significantly improves the performance of many TKG extrapolation models but also surpasses many state-of-the-art(SOTA) baseline methods.

Abstract

Explore similar work

OntoTKGE: Ontology-Enhanced Temporal Knowledge Graph Extrapolation

GAttNHP: Group Attention Neural Hawkes Process for Extrapolation Reasoning in Temporal Knowledge Graphs

AdaTKG: Adaptive Memory for Temporal Knowledge Graph Reasoning