# Multilingual Knowledge Graph Completion from Pretrained Language Models with Knowledge Constraints

arXiv: https://arxiv.org/abs/2406.18085 · HTML: https://mlsanity.com/paper/2406.18085/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 33 references that are in the map.
- First posted on arXiv 26 Jun 2024.

## Builds on (most cited first, 17 of 17)

- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [How Can We Know What Language Models Know?](https://mlsanity.com/paper/1911.12543/) (2019)
- [Complex Embeddings for Simple Link Prediction](https://mlsanity.com/paper/1606.06357/) (2016)
- [RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space](https://mlsanity.com/paper/1902.10197/) (2019)
- [MINE: Mutual Information Neural Estimation](https://mlsanity.com/paper/1801.04062/) (2018)
- [COMET: Commonsense Transformers for Automatic Knowledge Graph Construction](https://mlsanity.com/paper/1906.05317/) (2019)
- [On Mutual Information Maximization for Representation Learning](https://mlsanity.com/paper/1907.13625/) (2019)
- [Pretrained Transformers for Text Ranking: BERT and Beyond](https://mlsanity.com/paper/2010.06467/) (2020)
- [Holographic Embeddings of Knowledge Graphs](https://mlsanity.com/paper/1510.04935/) (2015)
- [Knowledge-aware Graph Neural Networks with Label Smoothness Regularization for Recommender Systems](https://mlsanity.com/paper/1905.04413/) (2019)
- [Pretrained Language Models for Text Generation: A Survey](https://mlsanity.com/paper/2105.10311/) (2021)
- [An Unsupervised Sentence Embedding Method by Mutual Information Maximization](https://mlsanity.com/paper/2009.12061/) (2020)
- [Common Sense Beyond English: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning](https://mlsanity.com/paper/2106.06937/) (2021)
- [Generative Knowledge Graph Construction: A Review](https://mlsanity.com/paper/2210.12714/) (2022)
- [From Discrimination to Generation: Knowledge Graph Completion with Generative Transformer](https://mlsanity.com/paper/2202.02113/) (2022)
- [Prix-LM: Pretraining for Multilingual Knowledge Base Construction](https://mlsanity.com/paper/2110.08443/) (2021)
- [Leveraging Knowledge in Multilingual Commonsense Reasoning](https://mlsanity.com/paper/2110.08462/) (2021)

## Built on by (most cited first, 4 of 4)

- [Multilingual Generative Retrieval via Cross-lingual Semantic Compression](https://mlsanity.com/paper/2510.07812/) (2025)
- [Multilingual Knowledge Graph Completion via Efficient Multilingual Knowledge Sharing](https://mlsanity.com/paper/2510.07736/) (2025)
- [In Data or Invisible: Toward a Better Digital Representation of Low-Resource Languages with Knowledge Graphs](https://mlsanity.com/paper/2605.05931/) (2026)
- [A Study of Temporal Fusion Strategies for Named Entity Recognition in Historical Texts](https://mlsanity.com/paper/2606.27881/) (2026)

## Abstract (arXiv)

Multilingual Knowledge Graph Completion (mKGC) aim at solving queries like (h, r, ?) in different languages by reasoning a tail entity t thus improving multilingual knowledge graphs. Previous studies leverage multilingual pretrained language models (PLMs) and the generative paradigm to achieve mKGC. Although multilingual pretrained language models contain extensive knowledge of different languages, its pretraining tasks cannot be directly aligned with the mKGC tasks. Moreover, the majority of KGs and PLMs currently available exhibit a pronounced English-centric bias. This makes it difficult for mKGC to achieve good results, particularly in the context of low-resource languages. To overcome previous problems, this paper introduces global and local knowledge constraints for mKGC. The former is used to constrain the reasoning of answer entities, while the latter is used to enhance the representation of query contexts. The proposed method makes the pretrained model better adapt to the mKGC task. Experimental results on public datasets demonstrate that our method outperforms the previous SOTA on Hits@1 and Hits@10 by an average of 12.32% and 16.03%, which indicates that our proposed method has significant enhancement on mKGC.