This is the official code release of the following paper:

Hao Dong, Ziyue Qiao, Zhiyuan Ning, Qi Hao, Yi Du, Pengyang Wang and Yuanchun Zhou. "Disentangled Multi-span Evolutionary Network against Temporal Knowledge Graph Reasoning." ACL 2025 Findings.

conda create -n dimnet python=3.8.15
conda activate dimnet
conda install pytorch==1.10.0 torchvision==0.11.0 torchaudio==0.10.0 cudatoolkit=11.3  -c pytorch  -c conda-forge
conda install pytorch-geometric==2.0.4 -c rusty1s -c conda-forge

❗️Note: This work also requires the wandb environment to be set up in advance and logged in. (Ref: W&B Quickstart)

1. Switch to src/ folder

cd src/

2. Run scripts

python main.py -d ICEWS18 --history_len 10 --num_head 4 --num_ly 3 --topk 50 --decay 1e-4 --gpu 0

To get the optimal result reported in the paper, change the hyperparameters and other setting according to the Implementation Details section in the paper.

If you find the resource in this repository helpful, please cite

@article{dong2025disentangled,
  title={Disentangled Multi-span Evolutionary Network against Temporal Knowledge Graph Reasoning},
  author={Dong, Hao and Qiao, Ziyue and Ning, Zhiyuan and Hao, Qi and Du, Yi and Wang, Pengyang and Zhou, Yuanchun},
  journal={arXiv preprint arXiv:2505.14020},
  year={2025}
}