# Open-World Knowledge Graph Completion

arXiv: https://arxiv.org/abs/1711.03438 · HTML: https://mlsanity.com/paper/1711.03438/

## Key facts

- Cited by 64 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Nov 2017.

## Built on by (most cited first, 30 of 64)

- [A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multimodal](https://mlsanity.com/paper/2212.05767/) (2022)
- [Knowledge Graphs: Opportunities and Challenges](https://mlsanity.com/paper/2303.13948/) (2023)
- [NormBank: A Knowledge Bank of Situational Social Norms](https://mlsanity.com/paper/2305.17008/) (2023)
- [OntoZSL: Ontology-enhanced Zero-shot Learning](https://mlsanity.com/paper/2102.07339/) (2021)
- [Knowledge-aware Zero-Shot Learning: Survey and Perspective](https://mlsanity.com/paper/2103.00070/) (2021)
- [T2G-Former: Organizing Tabular Features into Relation Graphs Promotes Heterogeneous Feature Interaction](https://mlsanity.com/paper/2211.16887/) (2022)
- [A Decade of Knowledge Graphs in Natural Language Processing: A Survey](https://mlsanity.com/paper/2210.00105/) (2022)
- [A Review of Biomedical Datasets Relating to Drug Discovery: A Knowledge Graph Perspective](https://mlsanity.com/paper/2102.10062/) (2021)
- [Open-world Machine Learning: Applications, Challenges, and Opportunities](https://mlsanity.com/paper/2105.13448/) (2021)
- [Contrastive Language-Image Pre-Training with Knowledge Graphs](https://mlsanity.com/paper/2210.08901/) (2022)
- [Zero-shot and Few-shot Learning with Knowledge Graphs: A Comprehensive Survey](https://mlsanity.com/paper/2112.10006/) (2021)
- [Knowledge Graph Self-Supervised Rationalization for Recommendation](https://mlsanity.com/paper/2307.02759/) (2023)
- [Learning to Sample and Aggregate: Few-shot Reasoning over Temporal Knowledge Graphs](https://mlsanity.com/paper/2210.08654/) (2022)
- [Graph Learning under Distribution Shifts: A Comprehensive Survey on Domain Adaptation, Out-of-distribution, and Continual Learning](https://mlsanity.com/paper/2402.16374/) (2024)
- [Can Language Models Act as Knowledge Bases at Scale?](https://mlsanity.com/paper/2402.14273/) (2024)
- [Generalizing to Unseen Elements: A Survey on Knowledge Extrapolation for Knowledge Graphs](https://mlsanity.com/paper/2302.01859/) (2023)
- [A Review on Knowledge Graphs for Healthcare: Resources, Applications, and Promises](https://mlsanity.com/paper/2306.04802/) (2023)
- [Reasoning Through Memorization: Nearest Neighbor Knowledge Graph Embeddings](https://mlsanity.com/paper/2201.05575/) (2022)
- [Neural Graph Databases](https://mlsanity.com/paper/2209.09732/) (2022)
- [Active Relation Discovery: Towards General and Label-aware Open Relation Extraction](https://mlsanity.com/paper/2211.04215/) (2022)
- [Are Missing Links Predictable? An Inferential Benchmark for Knowledge Graph Completion](https://mlsanity.com/paper/2108.01387/) (2021)
- [Few-Shot Inductive Learning on Temporal Knowledge Graphs using Concept-Aware Information](https://mlsanity.com/paper/2211.08169/) (2022)
- [Text-Augmented Open Knowledge Graph Completion via Pre-Trained Language Models](https://mlsanity.com/paper/2305.15597/) (2023)
- [Reddit Entity Linking Dataset](https://mlsanity.com/paper/2101.01228/) (2021)
- [Disconnected Emerging Knowledge Graph Oriented Inductive Link Prediction](https://mlsanity.com/paper/2209.01397/) (2022)
- [GraphOracle: Efficient Fully-Inductive Knowledge Graph Reasoning via Relation-Dependency Graphs](https://mlsanity.com/paper/2505.11125/) (2025)
- [Effective Few-Shot Named Entity Linking by Meta-Learning](https://mlsanity.com/paper/2207.05280/) (2022)
- [Inductive Knowledge Graph Reasoning for Multi-batch Emerging Entities](https://mlsanity.com/paper/2208.10378/) (2022)
- [A Survey On Few-shot Knowledge Graph Completion with Structural and Commonsense Knowledge](https://mlsanity.com/paper/2301.01172/) (2023)
- [Increasing Coverage and Precision of Textual Information in Multilingual Knowledge Graphs](https://mlsanity.com/paper/2311.15781/) (2023)

## Abstract (arXiv)

Knowledge Graphs (KGs) have been applied to many tasks including Web search, link prediction, recommendation, natural language processing, and entity linking. However, most KGs are far from complete and are growing at a rapid pace. To address these problems, Knowledge Graph Completion (KGC) has been proposed to improve KGs by filling in its missing connections. Unlike existing methods which hold a closed-world assumption, i.e., where KGs are fixed and new entities cannot be easily added, in the present work we relax this assumption and propose a new open-world KGC task. As a first attempt to solve this task we introduce an open-world KGC model called ConMask. This model learns embeddings of the entity's name and parts of its text-description to connect unseen entities to the KG. To mitigate the presence of noisy text descriptions, ConMask uses a relationship-dependent content masking to extract relevant snippets and then trains a fully convolutional neural network to fuse the extracted snippets with entities in the KG. Experiments on large data sets, both old and new, show that ConMask performs well in the open-world KGC task and even outperforms existing KGC models on the standard closed-world KGC task.