Datasets › FB15k-237

FB15k-237

1 Jul 2015 archive 2025-07-28

FB15k-237 is a link prediction dataset created from FB15k. While FB15k consists of 1,345 relations, 14,951 entities, and 592,213 triples, many triples are inverses that cause leakage from the training to testing and validation splits. FB15k-237 was created by Toutanova and Chen (2015) to ensure that the testing and evaluation datasets do not have inverse relation test leakage. In summary, FB15k-237 dataset contains 310,116 triples with 14,541 entities and 237 relation types.

Homepage Benchmarks Papers JSON

Benchmarks archive 2025-07-28

All 3 leaderboards whose dataset resolves to this page shown (sort by any header). "First row" is the archive's own first row at snapshot, in the archive's row order; nothing here re-ranks and metric direction is not asserted.

		First row (archive order)	Paper	Code
Link Prediction	FB15k-237	NBFNet Hits@1 0.321	Neural Bellman-Ford Networks: A General Graph Neural...	DeepGraphLearning/NBFNet +2	75	Compare
Complex Query Answering	FB15k-237	QTO MRR 1p 0.490	Answering Complex Logical Queries on Knowledge Graphs...	bys0318/qto	9	Compare
Knowledge Graph Completion	FB15k-237	KBGAT Hits@10 62.6	Learning Attention-based Embeddings for Relation...	deepakn97/relationPrediction +1	4	Compare

Papers archive 2025-07-28

30 shown of 75 papers with a leaderboard row on this dataset's benchmarks, newest first. The archive's own "papers using this dataset" list was never published, so this is the benchmark-backed subset; the archive's count for this dataset is 452. The Syntology column is from Syntology's graph (read 2026-09-24), stated per sample; it is not part of any archive number.

			Date	Samples run Syntology
Distance-Adaptive Quaternion Knowledge Graph Embedding with Bidirectional Rotation	1	1	5 Dec 2024	not harvested
MoCoKGC: Momentum Contrast Entity Encoding for Knowledge Graph Completion	0	1	12 Nov 2024	not harvested
KERMIT: Knowledge Graph Completion of Enhanced Relation Modeling with Inverse Transformation	1	1	26 Sep 2023	not harvested
MoCoSA: Momentum Contrast for Knowledge Graph Completion with Structure-Augmented Pre-trained Language Models	0	1	16 Aug 2023	not harvested
Self-Distillation with Meta Learning for Knowledge Graph Completion	1	1	20 May 2023	not harvested
Adapting Neural Link Predictors for Data-Efficient Complex Query Answering	0	1	29 Jan 2023	not harvested
Answering Complex Logical Queries on Knowledge Graphs via Query Computation Tree Optimization	1	1	19 Dec 2022	ran 1 of 7 samples (6 unverified; 7 pointer-only for licence)
A Retrieve-and-Read Framework for Knowledge Graph Link Prediction	1	2	19 Dec 2022	not harvested
KGLM: Integrating Knowledge Graph Structure in Language Models for Link Prediction	1	1	4 Nov 2022	ran 0 of 5 samples (5 unverified)
PALT: Parameter-Lite Transfer of Language Models for Knowledge Graph Completion	1	1	25 Oct 2022	not harvested
MEIM: Multi-partition Embedding Interaction Beyond Block Term Format for Efficient and Expressive Link Prediction	4	1	30 Sep 2022	ran 6 of 7 samples (1 unverified; 7 pointer-only for licence)
Joint Language Semantic and Structure Embedding for Knowledge Graph Completion	1	1	19 Sep 2022	not harvested
GreenKGC: A Lightweight Knowledge Graph Completion Method	1	2	19 Aug 2022	not harvested
Relphormer: Relational Graph Transformer for Knowledge Graph Representations	1	1	22 May 2022	ran 3 of 3 samples (0 unverified)
Neural-Symbolic Models for Logical Queries on Knowledge Graphs	1	1	16 May 2022	ran 0 of 4 samples (4 unverified)
SimKGC: Simple Contrastive Knowledge Graph Completion with Pre-trained Language Models	2	1	4 Mar 2022	ran 3 of 6 samples (3 unverified; 3 pointer-only for licence)
R-GCN: The R Could Stand for Random	1	1	4 Mar 2022	not harvested
From Discrimination to Generation: Knowledge Graph Completion with Generative Transformer	1	1	4 Feb 2022	not harvested
Reasoning Through Memorization: Nearest Neighbor Knowledge Graph Embeddings	1	1	14 Jan 2022	not harvested
Multi-task Pre-training Language Model for Semantic Network Completion	1	1	13 Jan 2022	not harvested
Self-attention Presents Low-dimensional Knowledge Graph Embeddings for Link Prediction	1	1	20 Dec 2021	not harvested
Do Pre-trained Models Benefit Knowledge Graph Completion? A Reliable Evaluation and a Reasonable Approach	0	1	16 Nov 2021	not harvested
Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones	1	1	28 Oct 2021	ran 2 of 2 samples (0 unverified)
Rot-Pro: Modeling Transitivity by Projection in Knowledge Graph Embedding	2	1	27 Oct 2021	not harvested
Relation Prediction as an Auxiliary Training Objective for Improving Multi-Relational Graph Representations	1	3	6 Oct 2021	not harvested
How Does Knowledge Graph Embedding Extrapolate to Unseen Data: A Semantic Evidence View	1	1	24 Sep 2021	not harvested
SAFRAN: An interpretable, rule-based link prediction method outperforming embedding models	1	1	16 Sep 2021	not harvested
KGRefiner: Knowledge Graph Refinement for Improving Accuracy of Translational Link Prediction Methods	0	1	27 Jun 2021	not harvested
Unified Interpretation of Softmax Cross-Entropy and Negative Sampling: With Case Study for Knowledge Graph Embedding	1	2	14 Jun 2021	not harvested
Neural Bellman-Ford Networks: A General Graph Neural Network Framework for Link Prediction	3	1	13 Jun 2021	ran 2 of 15 samples (13 unverified)

The full list of 75 is in the JSON twin.

Dataset loaders archive 2025-07-28

pykeen.readthedocs.io/en/latest/api/pykeen.datasets.FB15k237.html pykeen/pykeenpytorch

bi-graph/emgraphtf

2 loaders as listed in the archive; links are outbound and not re-checked here.

Tasks archive 2025-07-28

Link Prediction Knowledge Graph Completion Complex Query Answering

License archive 2025-07-28

No licence recorded in the archive. Absence here is not a statement about the dataset's terms.

Modalities archive 2025-07-28

Graphs

Languages archive 2025-07-28

English Chinese

Variants archive 2025-07-28

FB15k-237

1 variant name, as the archive lists them.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections