Saved in:
Bibliographic Details
Main Authors: Si, Yuehang, Zeng, Zefan, Huang, Jincai, Cheng, Qing
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.01597
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911034925645824
author Si, Yuehang
Zeng, Zefan
Huang, Jincai
Cheng, Qing
author_facet Si, Yuehang
Zeng, Zefan
Huang, Jincai
Cheng, Qing
contents Temporal Knowledge Graph (TKG) is an efficient method for describing the dynamic development of facts along a timeline. Most research on TKG reasoning (TKGR) focuses on modelling the repetition of global facts and designing patterns of local historical facts. However, they face two significant challenges: inadequate modeling of the event distribution shift between training and test samples, and reliance on random entity substitution for generating negative samples, which often results in low-quality sampling. To this end, we propose a novel distributional feature modeling approach for training TKGR models, Test-Time Training-guided Distribution shift Modelling (T3DM), to adjust the model based on distribution shift and ensure the global consistency of model reasoning. In addition, we design a negative-sampling strategy to generate higher-quality negative quadruples based on adversarial training. Extensive experiments show that T3DM provides better and more robust results than the state-of-the-art baselines in most cases.
format Preprint
id arxiv_https___arxiv_org_abs_2507_01597
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle T3DM: Test-Time Training-Guided Distribution Shift Modelling for Temporal Knowledge Graph Reasoning
Si, Yuehang
Zeng, Zefan
Huang, Jincai
Cheng, Qing
Artificial Intelligence
Computation and Language
Temporal Knowledge Graph (TKG) is an efficient method for describing the dynamic development of facts along a timeline. Most research on TKG reasoning (TKGR) focuses on modelling the repetition of global facts and designing patterns of local historical facts. However, they face two significant challenges: inadequate modeling of the event distribution shift between training and test samples, and reliance on random entity substitution for generating negative samples, which often results in low-quality sampling. To this end, we propose a novel distributional feature modeling approach for training TKGR models, Test-Time Training-guided Distribution shift Modelling (T3DM), to adjust the model based on distribution shift and ensure the global consistency of model reasoning. In addition, we design a negative-sampling strategy to generate higher-quality negative quadruples based on adversarial training. Extensive experiments show that T3DM provides better and more robust results than the state-of-the-art baselines in most cases.
title T3DM: Test-Time Training-Guided Distribution Shift Modelling for Temporal Knowledge Graph Reasoning
topic Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2507.01597