Conditional Diffusion Guided Knowledge Transfer for Multi-Domain Knowledge Graph Completion
Quick Answer
The DMKGC framework introduces a conditional diffusion-guided approach for multi-domain knowledge graph completion, achieving a 4.3% average MRR improvement in tail entity prediction across 14 KGs.
Quick Take
The DMKGC framework introduces a conditional diffusion-guided approach for multi-domain knowledge graph completion, achieving a 4.3% average MRR improvement in tail entity prediction across 14 KGs. This method preserves domain-specific information while enhancing knowledge transfer, particularly beneficial in low-resource scenarios.
Key Points
- DMKGC leverages conditional diffusion models for knowledge transfer across KGs.
- Initial domain-agnostic embeddings are refined using support KGs for better predictions.
- Achieves a 4.3% MRR improvement over state-of-the-art methods.
- Demonstrates sustained performance gains in low-resource data settings.
- Addresses limitations of existing methods by preserving contextual information.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Multi-domain knowledge graph completion (MKGC) aims to improve missing triple prediction in a target KG by transferring knowledge from other support KGs. Existing methods typically enforce consistency constraints on equivalent entities across KGs to transfer knowledge, which risks suppressing domain-specific contextual information of entities. This design can also compromise entity representation information from all KG domains, impeding performance improvements, especially in low-resource data scenarios. To address this, we pioneer a generation-based paradigm for MKGC and propose DMKGC, a conditional diffusion-guided knowledge transfer framework. Our key insight is to treat each KG as a partial view of the entity entire information, and generate informative domain-general entity embeddings through diffusion models conditioned on support KGs. Particularly, we first initialize domain-agnostic entity embeddings as prior entity embeddings, and then encode them within individual KGs. Afterward, we fuse equivalent entities from support KGs as the conditional diffusion generation guidance. We leverage the prior entity embeddings as the proxy generation objective, which ensures this conditional generation to be unbiased towards any conditioned KGs. Simultaneously, we also train the generated embeddings to be predictive across KGs, thus preserving domain-specific information. Extensive experiments on 14 KGs in 3 benchmarks demonstrate a 4.3\% average MRR improvement in tail entity prediction over state-of-the-art methods, with sustained gains in low-resource data settings.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.03154 [cs.CL] |
| (or arXiv:2607.03154v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.03154 arXiv-issued DOI via DataCite (pending registration) |
|
| Related DOI: | https://doi.org/10.1145/3774904.3792252
DOI(s) linking to related resources |
Submission history
From: Jiawei Sheng [view email]
[v1]
Fri, 3 Jul 2026 09:51:49 UTC (339 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →Quantifying Prior Dominance in Systems
The study introduces the Normalized Context Utilization (NCU) metric to evaluate Retrieval-Augmented Generation (RAG) systems, revealing that Small Language Models (SLMs) outperform larger models in factual extraction. The findings indicate that traditional scaling laws yield diminishing returns, with a commercial API frequently failing against adversarial evidence due to systemic confidence collapse.