@inproceedings{9f771d3fd9fa4187aeba99fb4680028e,
title = "ConvD: Attention Enhanced Dynamic Convolutional Embeddings for Knowledge Graph Completion (Extended Abstract)",
abstract = "We propose ConvD, a dynamic convolutional embedding model that addresses knowledge graph incompleteness by enhancing relation-entity feature interactions. Unlike existing convolution-based KGC models that rely on fixed external kernels, ConvD reshapes relation embeddings into multiple internal dynamic kernels and integrates a knowledge-guided attention mechanism to adaptively weight their contributions. Experiments on multiple benchmarks demonstrate that ConvD achieves consistent performance gains of 3. 2 8 \% - 1 4. 6 9 \% over state-of-the-art methods while reducing parameters by 5 0. 6 6 \% - 8 5. 4 0 \%. This abstract is based on our TKDE paper which was accepted on June 10, 2025, and the e-copy is available at the DOI link: https://doi.org/10.1109/TKDE.2025.3582243.",
keywords = "Knowledge graph, knowledge graph completion, knowledge representation learning, link prediction",
author = "Wenbin Guo and Zhao Li and Xin Wang and Zirui Chen and Jun Zhao and Jianxin Li and Ye Yuan",
note = "Publisher Copyright: {\textcopyright} 2026 IEEE.; 42nd International Conference on Data Engineering, ICDE 2026 ; Conference date: 04-05-2026",
year = "2026",
doi = "10.1109/ICDE65706.2026.00347",
language = "English",
series = "Proceedings - 2026 IEEE 42nd International Conference on Data Engineering, ICDE 2026",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "4303--4304",
editor = "Khuzaima Daudjee and Qiong Luo and Alkis Simitsis",
booktitle = "Proceedings - 2026 IEEE 42nd International Conference on Data Engineering, ICDE 2026",
address = "United States",
}