The DOI matches its Crossref record; 23 of the 23 references it cites were found in a registry or an index; 5 of the 14 sources its claims cite were checked for support.
TuckER: Tensor Factorization for Knowledge Graph Completion
Open accessAbstract
Ivana Balazevic, Carl Allen, Timothy Hospedales. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
TuckER is a linear link-prediction model based on Tucker decomposition of the binary knowledge-graph tensor, using a shared core tensor and entity/relation embeddings. It achieves state-of-the-art results on standard datasets, is fully expressive, and subsumes several prior linear models as special cases.
Key findings
- TuckER reaches state-of-the-art link prediction results on WN18RR (MRR .470, Hits@10 .526, Hits@3 .482, Hits@1 .443) and FB15k-237 (MRR .358, Hits@10 .544, Hits@3 .394, Hits@1 .266), and on WN18 (MRR .953, Hits@10 .958, Hits@3 .955, Hits@1 .949) and FB15k (MRR .795, Hits@10 .892, Hits@3 .833, Hits@1 .741), apart from Hits@10 on WN18 where R-GCN does better.
- TuckER is fully expressive: Theorem 1 gives a sufficient bound of entity embedding dimensionality \(d_e = n_e\) and relation embedding dimensionality \(d_r = n_r\).
- RESCAL, DistMult, ComplEx, and SimplE can be viewed as special cases of TuckER.
- Parameter sharing through the core tensor helps: on WN18RR TuckER uses ∼9.4M parameters versus ∼16.4M for ComplEx and SimplE at \(d_e=200, d_r=30\); at embedding size 20 (∼300k parameters), TuckER is almost as good as ComplEx and SimplE at embedding size 200 (∼6M parameters), and on FB15k it improves Hits@1 by +14% over ComplEx and +8% over SimplE.
Methods
- TuckER factorizes the third-order binary tensor of triples with Tucker decomposition, scoring \(\phi(e_s, r, e_o) = W \times_1 e_s \times_2 w_r \times_3 e_o\), where \(W\) is a core tensor and subject/object entity embeddings are shared; a logistic sigmoid converts scores to probabilities.
- Training uses reciprocal relations, 1-N scoring, Bernoulli negative log-likelihood, Adam with batch size 128, and random-search hyperparameters; evaluation uses the filtered setting with MRR and Hits@k on FB15k, FB15k-237, WN18, and WN18RR.
- Theoretical analysis proves full expressiveness, derives the dimensionality bound, and shows the relationship between TuckER and prior linear tensor-factorization models.
Limitations
- The paper states future work might include exploring how to incorporate background knowledge on individual relation properties; it does not present a dedicated limitations section.
- Evident from the text: evaluation is limited to four standard link-prediction datasets and MRR/Hits@k metrics, with no broader task or computational-efficiency analysis beyond parameter counts.
- Evident from the text: the full-expressiveness bound \(d_e = n_e, d_r = n_r\) is theoretical and may be much larger than practical embedding sizes; the authors expect smaller dimensionalities to suffice in practice.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
5 of 14 cited sources checked · 3 supported · 0 flagged · accuracy not yet measured
Every source the paper cites for this claim was checked and supports it.
In the paper
“The authors show that convolution is a way of introducing sparsity and parameter tying and that HypER can be understood in terms of tensor factorization up to a non-linearity, thus placing HypER closer to the well established family of factorization models.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 1 Introduction
What the sources say
- Supportedcites Balažević et al., 2019 source
Section 4.2 (p. 6) describes HypER as tensor multiplication with a constrained core tensor and notes that ReLU is applied before the final dot product. The Related Work section (p. 3) explicitly identifies that ReLU non-linearity as the key difference from existing factorization approaches and describes convolution as multiplication by a sparse tensor with tied weights. (LLM, conf=0.94)
Strength:boosted· not yet calibratedCausal:correlational· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“They further prove other models are not fully expressive: DistMult, because it cannot model asymmetric relations; and transitive models such as TransE (Bordes et al., 2013) and its variants FTransE (Feng et al., 2016) and STransE (Nguyen et al., 2016), because of certain contradictions that they impose between different relation types.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.1 Full Expressiveness and Embedding Dimensionality
What the sources say
- Not matchedcites Bordes et al., 2013
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Not matchedcites Feng et al., 2016
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Supportedcites Nguyen et al., 2016 source
The citation immediately following “STransE” identifies the model, rather than attributing the citing authors’ expressiveness proof to Nguyen et al. The source’s model description (SE section; page unavailable) explicitly introduces STransE. It does not establish the claimed contradictions, but that proof is framed as the citing authors’ own. (LLM, conf=0.91)
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes- Not matched
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“To our knowledge, ComplEx-N3 (Lacroix et al., 2018) is the only other linear link prediction model that benefits from multitask learning.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
What the sources say
- Partly supportedcites Lacroix et al., 2018 sourceExperimental — accuracy not yet measured
The Experimental Setup (p. 7) identifies ComplEx-N3, and the Introduction (pp. 1–2) reports performance gains for ComplEx from modeling inverse predicates and, on harder datasets, from N3 regularization. Our copy does not establish that ComplEx-N3 benefits from multitask learning or that it is the only other linear link-prediction model to do so. (LLM, conf=0.78)
Strength:hedged· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“There, rank regularization of the embedding matrices is used to encourage a lowrank factorization, thus forcing parameter sharing between relations.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
What the sources say
- Partly supportedcites Lacroix et al., 2018 sourceExperimental — accuracy not yet measured
Section 4 (p. 4) confirms that the paper proposes nuclear 3-norm regularization for tensor-factorization models, including ComplEx. But it does not say that this regularizer forces parameter sharing between relations; Section 2.2 (p. 2) instead describes ComplEx’s parameter sharing as occurring between the subject and object modes. (LLM, conf=0.84)
Strength:plain· not yet calibratedCausal:direct· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passes- Partly supported
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“TuckER outperforms previous state-of-the-art models across standard link prediction datasets, acting as a strong baseline for more elaborate models.”
- Also stated in: § 1 Introduction · plain
“proposing TuckER, a new linear model for link prediction on knowledge graphs, that is simple, expressive and achieves state-of-theart results across all standard datasets”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > Abstract
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes- The authors state this as an assumption: Subject and object entity embedding matrices are equivalent, i.e. no distinction is made between the embeddings of an entity depending on whether it appears as a subject or as an object.
- Also stated in: § 1 Introduction · plain
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This is supported by the fact that the margin by which TuckER outperforms other linear models is notably increased on datasets with a large number of relations.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesThe paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“Even with fewer parameters than ComplEx and SimplE at d e = 200 and d r = 30 on WN18RR (∼9.4 vs ∼16.4 million), TuckER consistently obtains better results than any of those models.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We can see that the difference between the MRRs of ComplEx, SimplE and TuckER is approximately constant for embedding sizes 100 and 200.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.4 Influence of Parameter Sharing
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes? Load-bearing: unclear✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, running their model with equivalent parameter number to TuckER shows comparable performance, supporting our belief that the two models both attain the benefits of multi-task learning, although by different means.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesA result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It is important to note that at embedding dimensionalities 20, 50 and 100, TuckER has fewer parameters than Com-plEx and SimplE (e.g.ComplEx and SimplE have ∼3 million and TuckER has ∼2.5 million parameters for embedding dimensionality 100).”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.4 Influence of Parameter Sharing
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passes
- Probably not a substantial claim. 9
- It may state two separate claims as one. 9
Every source the paper cites for this claim was checked and supports it.
In the paper
“As shown in (Trouillon et al., 2017), ComplEx is fully expressive with the embedding dimensionality bound d e = d r = n e • n r .”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.1 Full Expressiveness and Embedding Dimensionality
What the sources say
- Supportedcites Trouillon et al., 2017 source
Theorem 4 in §3.2 (p. 14) shows that ComplEx can represent arbitrary families of relation matrices using a shared entity embedding matrix and diagonal relation matrices of dimension nm, where n is the number of entities and m the number of relations. It states that the decomposition rank is at most nm, matching the claimed dimensionality bound for entity and relation embeddings. (LLM, conf=0.92)
Strength:boosted· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
- Supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“However, for lower embedding sizes, the dif-ference between MRRs increases by 0.7% for embedding size 50 and by 4.2% for embedding size 20 for ComplEx and by 3% for embedding size 50 and by 9.9% for embedding size 20 for Sim-plE.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.4 Influence of Parameter Sharing
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Results achieved by TuckER are not only better than those of other linear models, such as DistMult, ComplEx and SimplE, but also better than the results of many more complex deep neural network and reinforcement learning architectures, e.g.R-GCN, MINERVA, ConvE and HypER, demonstrating the expressive power of linear models and supporting our claim that simple linear models should serve as a”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
Load-bearingStrength:boosted· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Overall, TuckER outperforms previous state-of-the-art models on all metrics across all datasets (apart from hits@10 on WN18 where a non-linear model, R-GCN, does better).”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For example, improvement on FB15k is +14% over ComplEx and +8% over SimplE on the toughest hits@1 metric.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“It is trivial to show that both RESCAL and DistMult belong to the family of bilinear models.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.2 Relation to Previous Linear Models
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“It is interesting to note that the TuckER interpretation of the DistMult scoring function, given that matrices A and C are identical, can alternatively be interpreted as a special case of CP decomposition (Hitchcock, 1927), since Tucker decomposition with a superdiagonal core tensor is equivalent to CP decomposition.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.2 Relation to Previous Linear Models
What the sources say
- Not checked — no full text of the sourcecites Hitchcock, 1927 source
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“We show that TuckER is a fully expressive model, derive sufficient bounds on its embedding dimensionalities and demonstrate that several previously introduced linear models can be viewed as special cases of TuckER.”
- Also stated in: § 1 Introduction · plain
“Finally, we show that several previous stateof-the-art linear models, RESCAL (Nickel et al., 2011), DistMult (Yang et al., 2015), ComplEx (Trouillon et al., 2016) and SimplE (Kazemi and Poole, 2018), are special cases of TuckER.”
- Also stated in: § 1 Introduction · plain
“showing how TuckER subsumes several previously proposed tensor factorization approaches to link prediction.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > Abstract
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
- The authors state this as an assumption: Subject and object entity embedding matrices are equivalent, i.e. no distinction is made between the embeddings of an entity depending on whether it appears as a subject or as an object.
- Also stated in: § 1 Introduction · plain
In the paper
“Similarly to Com-plEx, Kazemi and Poole (2018) show that SimplE is fully expressive with entity and relation embeddings of size d e = d r = min(n e • n r , γ + 1), where γ represents the number of true facts.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.1 Full Expressiveness and Embedding Dimensionality
What the sources say
- Not matchedcites Kazemi and Poole (2018)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
- Not matched
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“So far, there have been two possible ways in which linear link prediction models introduce asymmetry into factorization of the binary tensor of triples: • distinct (although possibly related) embeddings for subject and object entities and a diagonal matrix (or equivalently a vector) for each relation, as is the case with models such as ComplEx and SimplE; or • equivalent subject and object entity”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.3 Representing Asymmetric Relations
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim. It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“We use evaluation metrics standard across the link prediction literature: mean reciprocal rank (MRR) and hits@k, k ∈ {1, 3, 10}.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.2 Implementation and Experiments
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“TuckER overcomes this by representing relations as vectors w r , which makes the parameter number grow linearly with the number of relations, while still keeping the desirable property of allowing relations to be asymmetric”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.3 Representing Asymmetric Relations
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
- Rests on an assumption: The full rank matrix approach for relations often leads to overfitting, especially for relations with a small number of training triples.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“To test this, we re-implement ComplEx and SimplE with reciprocal relations, 1-N scoring, batch normalization and dropout for fair comparison, perform random search to choose best hyper-parameters”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.4 Influence of Parameter Sharing
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“TuckER achieves state-of-the-art results on standard link prediction datasets, in part due to its ability to perform multi-task learning across relations.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 7 Conclusion
Tests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesIt may state two separate claims as one.
In the paper
“We use the filtered setting (Bordes et al., 2013), i.e. all known true triples are removed from the candidate set except for the current test triple.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.2 Implementation and Experiments
What the sources say
- Not matchedcites Bordes et al., 2013
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Load-bearingStrength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✓ Load-bearing: passes✓ Based: passesProbably not a substantial claim.
- Not matched
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“We choose all hyper-parameters by random search based on validation set performance.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.2 Implementation and Experiments
Strength:plain· not yet calibratedTests✓ Asserted: passes✓ Checkable: passes✗ Load-bearing: fails✓ Based: passesProbably not a substantial claim. The paper’s conclusions may not depend on it.
The paper relies on this without support.
Knowledge graphs typically contain only a small subset of all possible facts.
“However, they typically contain only a small subset of all possible facts.”
“However, far from all available information is currently stored in existing knowledge graphs”
The paper relies on this without support.
Manually adding new information to knowledge graphs is costly.
“and manually adding new information is costly”
The authors state this as an assumption.
Subject and object entity embedding matrices are equivalent, i.e. no distinction is made between the embeddings of an entity depending on whether it appears as a subject or as an object.
“Subject and object entity embedding matrices are assumed equivalent, i.e. we make no distinction between the embeddings of an entity depending on whether it appears as a subject or as an object in a particular triple.”
The paper relies on this without support.
The full rank matrix approach for relations often leads to overfitting, especially for relations with a small number of training triples.
“which often leads to overfitting, especially for relations with a small number of training triples.”
The authors state this as an assumption.
TuckER should need a lower number of parameters for obtaining good results than ComplEx or SimplE.
“The ability of knowledge sharing through the core tensor suggests that TuckER should need a lower number of parameters for obtaining good results than ComplEx or SimplE.”
“which supports our initial assumption.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
RESCAL parameter count squares the wrong dimension
MinorTable 1 gives RESCAL a term proportional to the square of the relation embedding dimension, d_r, but the paper describes a d_e × d_e matrix for each relation. That description implies a term proportional to the square of d_e instead.
“RESCAL (Nickel et al., 2011) e s W r e o W r ∈ R de 2 O(n e d e + n r d 2 r )”
“the relation matrices W r ∈ R de×de are the d e × d e slices of the core tensor Z.”
The core tensor has conflicting dimension orders
MinorTable 1 gives the TuckER core tensor dimensions as entity × entity × relation, whereas the model definition gives entity × relation × entity. These orders differ when entity and relation embedding dimensions differ, as they do for WN18 and WN18RR.
“W ∈ R de×de×dr is the core tensor of a Tucker decomposition.”
“w r ∈ R dr the rows of R representing the relation embedding vector and W ∈ R de×dr×de is the core tensor.”
TuckER does not outperform HypER on WN18 hits@3
MinorThe results table reports the same WN18 hits@3 value, .955, for TuckER and HypER. This is a tie, although the text says TuckER outperforms previous models on every metric except WN18 hits@10.
“Overall, TuckER outperforms previous state-of-the-art models on all metrics across all datasets (apart from hits@10 on WN18 where a non-linear model, R-GCN, does better).”
“| HypER (Balažević et al., 2019) | no | .951 | 958 | .955 | .947 | .790 | .885 | .829 | .734 | | SimplE (Kazemi and Poole, 2018) | yes | .942 | .947 | .944 | .939 | .727 | .838 | .773 | .660 | | TuckER (ours) | yes | .953 | .958 | .955 | .949 | .795 | .892 | .833 | .741 |”
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 6
- Antoine Bordes, Nicolas Usunier, Alberto Garcia- Duran, Jason Weston, and Oksana Yakhnenko. 2013. Translating Embeddings for Modeling Multi- relational Data. In Advances in Neural Information Processing Systems.
- Rajarshi Das, Shehzaad Dhuliawala, Manzil Zaheer, Luke Vilnis, Ishan Durugkar, Akshay Krishna- murthy, Alex Smola, and Andrew McCallum. 2018. Go for a Walk and Arrive at the Answer: Reason- ing over Paths in Knowledge Bases Using Rein- forcement Learning. In International Conference on Learning Representations.
- Seyed Mehran Kazemi and David Poole. 2018. Sim- plE Embedding for Link Prediction in Knowledge Graphs. In Advances in Neural Information Pro- cessing Systems.
- Maximilian Nickel, Volker Tresp, and Hans-Peter Kriegel. 2011. A Three-Way Model for Collective Learning on Multi-Relational Data. In International Conference on Machine Learning.
- Adam Paszke, Sam Gross, Soumith Chintala, Gre- gory Chanan, Edward Yang, Zachary DeVito, Zem- ing Lin, Alban Desmaison, Luca Antiga, and Adam Lerer. 2017. Automatic Differentiation in PyTorch. In NIPS-W.
- Bishan Yang, Wen-tau Yih, Xiaodong He, Jianfeng Gao, and Li Deng. 2015. Embedding Entities and Relations for Learning and Inference in Knowledge Bases. In International Conference on Learning Representations.
Cited by
Sorted by citations
Items 201–300 of 699
A divide and conquer framework for Knowledge Editing
Han, X., Li, R., Li, X., & Pan, J. Z. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110826 · Cited by 12
KnowSite: Leveraging Urban Knowledge Graph for Site Selection
Liu, Y., Ding, J., & Li, Y. (2023) · ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems (SIGSPATIAL GIS) · doi:10.1145/3589132.3625640 · Cited by 10
“As for KRL to learn embeddings of entities and relations, though complete structures like GNN have been introduced [28, 35], tensor decomposition models still achieve the best performance [14], such as DistMult [46] and TuckER [2].”
MethodRevisiting Probabilistic Latent Semantic Analysis: Extensions, Challenges and Insights
Figuera, P., & Bringas, P. G. (2024) · Technologies · doi:10.3390/technologies12010005 · Cited by 13
Relation-based multi-type aware knowledge graph embedding
Xue, Y., Jin, J., Song, A., Zhang, Y., Liu, Y., & Wang, K. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.05.021 · Cited by 10
LHP: Logical hypergraph link prediction
Yang, Y., Li, X., Guan, Y., Wang, H., Kong, C., & Jiang, J. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.119842 · Cited by 11
Overview of indoor scene recognition and representation methods based on multimodal knowledge graphs
Li, J., Si, G., Tian, P., An, Z., & Zhou, F. (2023) · Applied Intelligence · doi:10.1007/s10489-023-05235-7 · Cited by 11
Deep hyperbolic convolutional model for knowledge graph embedding
Lu, M., Li, Y., Zhang, J., Ren, H., & Zhang, X. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112183 · Cited by 10
Balloccu, G., Boratto, L., Fenu, G., Marras, M., & Soccol, A. (2024) · ACM Conference on Recommender Systems (RecSys) · doi:10.1145/3640457.3691703 · Cited by 10
“General-purposemethodsinclude RESCAL [20], DistMult [37], HolE [19], ComplEx [31], Analogy [17], TransE [6], TransH [34], TransR [16], TransD [15], TorusE [11], RotatE [28], ConvE [10],and TuckER [2].”
Liu, W., Ren, Z., & Chen, L. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113063 · Cited by 9
KERMIT: Knowledge graph completion of enhanced relation modeling with inverse transformation
Li, H., Yu, B., Wei, Y., Wang, K., Xu, R. Y. D., & Wang, B. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.113500 · Cited by 13
Path-Based Knowledge Graph Link Prediction Method With Graph Context
Lu, J., Chen, J., Zhu, H., Xiao, Z., Xiao, G., & Wang, Q. (2026) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2026.3669923 · Cited by 12
“However, the majority of these embedding methods, e.g., TransE [3], KGLG [4], TransL [5], DualE [6], DualDE [7], RKGE [8], RotatE [9], TuckER [10], and convKB [11], disregard paths and only consider direct relations among entities.”
Specific Time Embedding for Temporal Knowledge Graph Completion
Ni, R., Ma, Z., Yu, K., & Xu, X. (2020) · IEEE International Conference on Cognitive Informatics & Cognitive Computing (ICCI*CC) · doi:10.1109/iccicc50026.2020.9450214 · Cited by 9
“…representation from multi-source spatio-temporal data and recognizes the network structure and semantic relationships about multi-source spatio-temporal data [9] ConT is the best performing model of [10], which is an extension of Tucker [11], use time embedding to replace shared vector in Tucker.”
MethodWeinzierl, M., Hopfer, S., & Harabagiu, S. M. (2022) · Journal of Behavioral Medicine · doi:10.1007/s10865-022-00328-z · Cited by 9
ContE: contextualized knowledge graph embedding for circular relations
Ma, T., Li, M., Lv, S., Zhu, F., Huang, L., & Hu, S. (2022) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-022-00851-2 · Cited by 9
“The models combined with the ContE include TransE+ContE, TransH+ContE, DistMult+ContE, ComplEx+ContE, RotatE+ContE, TuckER +ContE and ConvE+ContE.”
MethodInfluentialKnowledge Graph Embedding by Normalizing Flows
Xiao, C., He, X., & Cao, Y. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i4.25600 · Cited by 9
“…(Sun et al. 2018), bilinear models, including DistMult (Yang et al. 2014), ComplEx (Toutanova et al. 2015), QuatE (Zhang et al. 2019), TuckER (Balaˇzevi´c, Allen, and Hospedales 2019a), neural networks models, including ConvE (Dettmers et al. 2018), HypER (Bala ˇ zevi ´ c, Allen, and…”
MethodInfluentialReasoning beyond Triples: Recent Advances in Knowledge Graph Embeddings
Xiong, B., Nayyeri, M., Daza, D., & Cochez, M. (2023) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3583780.3615294 · Cited by 10
“For example, HsimplE [17], m-CP [17], and GETD [32] generalize SimplE [33], Canonical Polyadic (CP) decomposition [34], and TuckER [35], respectively.”
BackgroundShao, P., Wen, Y., & Tao, J. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106146 · Cited by 9
TSA-Net: a temporal knowledge graph completion method with temporal-structural adaptation
Xie, R., Ruan, K., Huang, B., Yu, W., Xiao, J., & Huang, J. (2024) · Applied Intelligence · doi:10.1007/s10489-024-05734-1 · Cited by 9
“TuckERT [27] builds a TuckERT decomposition method inspired by Tucker decomposition of order 4 tensor.”
Zhou, K., Zhang, S., Wang, Y., et al. (2022) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2021.103973 · Cited by 9
Taxonomy enrichment with text and graph vector representations
Nikishina, I., Tikhomirov, M. M., Logacheva, V., Nazarov, Y. P., Panchenko, A., & Loukachevitch, N. (2022) · Semantic Web · doi:10.3233/sw-212955 · Cited by 11
“Another group of approaches makes use of tensor decomposition approaches: Tucker decomposition [21] and Canonical Polyadic decomposition (CANDECOMP/PARAFAC) [22, 23].”
MethodDeng, Y., Bai, W., Jiang, Y., et al. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.109906 · Cited by 9
Liu, W., Bansal, D., Daruna, A., & Chernova, S. (2023) · Autonomous Robots · doi:10.1007/s10514-023-10099-4 · Cited by 9
“• TuckER is a recent state of the art knowledge graph embedding model [6].”
BackgroundA Semantic Enhanced Knowledge Graph Embedding Model With AIGC Designed for Healthcare Prediction
Yang, Q., He, M., Li, Z., He, T., & Jeon, S. (2024) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2024.3398207 · Cited by 10
“In the feature extraction layer, the range of Batch Size Setting is [32,64,128], the number of filters is [25,32,50,64], and the number of Invalid Triplets Setting is [40,80,160].”
Hamming Distance Encoding Multihop Relation Knowledge Graph Completion
Chen, P., Wang, Y., Yu, Q., Fan, Y. H., & Feng, R. (2020) · IEEE Access · doi:10.1109/access.2020.3004448 · Cited by 9
“The one is computing the score and ranking for the best ordered entities as candidates similar to TuckER [27] by Eq.”
BackgroundLearning hierarchy-aware complex knowledge graph embeddings for link prediction
Zhang, J., Shen, B., & Zhang, Y. (2024) · Neural Computing and Applications · doi:10.1007/s00521-024-09775-y · Cited by 8
“Tucker [25] factorizes a tensor into a core tensor multiplied by a matrix along each mode, with the core tensor controlling the level of interaction between entities and relations and the matrix containing the embeddings of entities and relations.”
Detect-Then-Resolve: Enhancing Knowledge Graph Conflict Resolution with Large Language Model
Peng, H., Zhang, P., Tang, J., Xu, H., & Zeng, W. (2024) · Mathematics · doi:10.3390/math12152318 · Cited by 9
Multi-level Shared Knowledge Guided Learning for Knowledge Graph Completion
Shan, Y., Zhou, J., Peng, J., Zhou, X., Yin, J., & Wang, X. (2024) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00686 · Cited by 9
“Notably, we observe considerable performance variations across various KGC subtasks, even when applied to the same dataset.”
BackgroundYang, J., Yang, S., Gao, Y., et al. (2024) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3664647.3681696 · Cited by 8
“Additional models likeMKGformer[32],OTKGE[49],MKBE[22],QEB[33],CMGNN[23], TBKGC[13],RSME[20],TransAE[48],VBKGC[42],MULTIFORM[36] andAdaMF-MAT[43]alsoutilizediversemodalitiesbutoftenstrug- gletopreservetheuniquepropertiesofeachmodalitywhenmerg-ingthemintoasingleembeddingspace.”
Towards Structure-Aware Model for Multi-Modal Knowledge Graph Completion
Li, L., Jin, Z. G., Zhang, Y., Jin, D., Dou, C., He, Y., Zhang, X., & Zhao, H. (2025) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2025.3639983 · Cited by 9
“Thus, the visual tag Tokens visual ( e ) of each entity e is obtained: where n represents the number of visual modality tokens.”
Design Patterns for Large Language Model Based Neuro-Symbolic Systems
de Boer, M., Smit, Q. T. S., van Bekkum, M. A., Meyer-Vitali, A., & Schmid, T. (2025) · Neurosymbolic Artificial Intelligence · doi:10.1177/29498732251377499 · Cited by 13
“By integrating the Knowledge Attention and Recontextualisation layers (Balaževi´c et al., 2019), graph entity embeddings are used that are processed through an attention mechanism to enhance entity span embeddings.”
Beyond facts – a survey and conceptualisation of claims in online discourse analysis
Boland, K., Fafalios, P., Tchechmedjiev, A., Dietze, S., & Todorov, K. (2022) · Semantic Web · doi:10.3233/sw-212838 · Cited by 10
“Populating and building KBs and thus providing structured knowledge on the Web has been of central interest in the NLP, web, data mining and the semantic web communities over the past decades, focusing on a variety of tasks such as named entity recognition, entity linking, relation extraction or word sense disambiguat…”
BackgroundInfluentialMessage Function Search for Knowledge Graph Embedding
Di, S., & Chen, L. (2023) · ACM Web Conference (WWW) · doi:10.1145/3543507.3583546 · Cited by 8
“To manipulate web-scale KG/NRD/HKG, various promising embedding models [35, 39, 53] have been proposed to encode sets of relations R and entities E into d-dimensional vector space R ∈ R |R |×d , E ∈ R |E |×d , such as geometric models [7, 43, 58], neural network models [19, 40, 49], bilinear models [25, 47, 61], and m…”
MethodInfluentialWeighted Knowledge Graph Embedding
Zhang, Z., Guan, Z., Zhang, F., Zhuang, F., An, Z., Wang, F., & Xu, Y. (2023) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3539618.3591784 · Cited by 8
“• Tensor decomposition methods, including DistMult [46], Complex [39], TuckER [1] and MEI [23].”
MethodFederated Knowledge Graph Completion via Latent Embedding Sharing and Tensor Factorization
Wang, M., Zeng, D., Xu, Z., Guo, R., & Zhao, X. (2023) · Proceedings · doi:10.1109/icdm58522.2023.00175 · Cited by 9
“Tensor contraction [18], [19] means that two tensors are contracted into one tensor along their associated pairs of indices.”
BackgroundMethodInfluentialLi, Z., Wang, X., Zhao, J., et al. (2025) · doi:10.1145/3696410.3714549 · Cited by 8
Fantastic Knowledge Graph Embeddings and How to Find the Right Space for Them
Nayyeri, M., Xu, C., Vahdati, S., Vassilyeva, N., Sallinger, E., Yazdi, H. S., & Lehmann, J. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-62419-4_25 · Cited by 8
“MR MMR Hits@1 Hits@3 Hits@10 MR MMR Hits@1 Hits@3 Hits@10 [19]; results for TuckER are from [1]; and results from TransComplEx are quoted from [16].”
BackgroundLe, T., Huynh, N., & Le, B. H. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-88081-1_3 · Cited by 8
GFCNet: Utilizing graph feature collection networks for coronavirus knowledge graph embeddings
Xie, Z., Zhu, R., Liu, J., Zhou, G., Huang, J. X., & Cui, X. (2022) · Information Sciences · doi:10.1016/j.ins.2022.07.031 · Cited by 8
“Bilinear models, also called semantic matching models, use matrix decomposition to learn knowledge graph embeddings, including Rescal [36], DistMult [58], ComplEx [43], SimplE [25], TuckER [5] and MARINE [13].”
InfluentialCP Tensor Factorization for Knowledge Graph Completion
Luo, Y., Yang, C., Li, B., Zhao, X., & Zhang, H. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-10983-6_19 · Cited by 7
“Typical models include RESCAL [27], DisMult [28], ComplEx [29], SimplE [30], and TuckER [15], all of which apply different decomposition methods for this third-order binary tensor to solve the KG completion problem.”
BackgroundMethodDifferentiable learning of rules with constants in knowledge graph
Xu, Z., Ye, P., Li, J., Chen, H., & Zhang, W. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110686 · Cited by 8
SAppKG: Mobile App Recommendation Using Knowledge Graph and Side Information-A Secure Framework
Dave, D., Sharma, A., Abdulhamid, S. M., et al. (2023) · IEEE Access · doi:10.1109/access.2023.3296466 · Cited by 8
“• TuckER [36]: TuckER is a generalized version of the linear models Rescal, DistMult, ComplEx, and SimplE.”
BackgroundEntity-Relation Guided Random Walk for Link Prediction in Knowledge Graphs
Li, W., Zhong, H., Lin, R., Chang, C., Pan, Z., & Tang, Y. Z. (2024) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2024.3382263 · Cited by 7
“…head and tail entity embedding exhibit circular correlation, contributing to its ability to obtain comprehensive representations of compositional vector space in KGs. TuckER [42] learns embedding for entities and relations by generating a core tensor alongside the corresponding embedding vectors.”
BackgroundExpressiveness Analysis and Enhancing Framework for Geometric Knowledge Graph Embedding Models
Song, T., Yin, L., Liu, Y., Liao, L., Luo, J., & Xu, Z. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3486915 · Cited by 7
Chen, B., Li, H., Zhao, D., Yang, Y., & Pan, C. (2024) · Complex & Intelligent Systems · doi:10.1007/s40747-024-01661-3 · Cited by 9
“The TuckER model [27] proposed by Ivana et al. is a typical example based on binary tensor decomposition.”
Luo, X., Cheng, S., Wang, L., et al. (2024) · International Journal of Geographical Information Systems · doi:10.1080/13658816.2024.2434665 · Cited by 7
Spatio-Temporal Knowledge Driven Diffusion Model for Mobile Traffic Generation
Chai, H., Qi, X., & Li, Y. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3527966 · Cited by 10
“Based on this idea, we adopt a contrastive learning-based graph representation method, TuckER [38], to extract BS embeddings from the UKG.”
C 2 RS: Multimodal Knowledge Graph Completion With Cross-Modal Consistency and Relation Semantics
Shu, Y., Li, W., Wang, J., Zhang, Y., Guan, J., & Zhou, S. (2025) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2025.3548621 · Cited by 9
ReranKGC: A cooperative retrieve-and-rerank framework for multi-modal knowledge graph completion
Gao, M., Xie, Y., Chen, W., Zhang, F., Ding, F., Wang, T., Yao, J., Zheng, J., & Wong, K. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107467 · Cited by 9
Two-View Graph Neural Networks for Knowledge Graph Completion
Tong, V. V., Nguyen, D. Q., Phung, D., & Nguyen, D. Q. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-33455-9_16 · Cited by 8
TemporalFC: A Temporal Fact Checking Approach over Knowledge Graphs
Qudus, U., Röder, M., Kirrane, S., & Ngomo, A.-C. N. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-47240-4_25 · Cited by 8
Spatiotemporal knowledge graph completion via diachronic and transregional word embedding
Xu, X., Jia, W. P., Li, Y., et al. (2024) · Information Sciences · doi:10.1016/j.ins.2024.120477 · Cited by 7
Filling the gap: Enhancing borehole imaging with a tensor neural network
Liu, D., Wang, Q., You, N., Sacchi, M. D., & Chen, W. (2024) · Geophysics · doi:10.1190/geo2024-0242.1 · Cited by 7
Embedding Knowledge Graphs Attentive to Positional and Centrality Qualities
Sadeghi, A., Collarana, D., Graux, D., & Lehmann, J. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-86520-7_34 · Cited by 9
“Our baselines include RotatE [20], TuckER [1], ComplEx-N3 [11], QuatE [36], MDE [17] and the recent graph neural network C omp GCN [25].”
MethodKnowledge graph embedding with entity attributes using hypergraph neural networks
Xu, Y.-W., Zhang, H., Cheng, K. D., Liao, X.-L., Zhang, Z., & Li, Y. (2022) · Intelligent Data Analysis · doi:10.3233/ida-216007 · Cited by 7
GCKG: Novel Gated Convolutional embedding model for Knowledge Graphs
Yao, S., Pi, D., Chen, J., & Xu, Y. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.118142 · Cited by 7
Regularized online tensor factorization for sparse knowledge graph embeddings
Zulaika, U., Almeida, A., & López–de–Ipiña, D. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-07796-z · Cited by 7
“These factorization algorithms [2, 25, 44] produce distributed representations of the entities and relations within the Knowledge Graph, also known as Knowledge Graph Embeddings.”
MethodTCKGE: Transformers with contrastive learning for knowledge graph embedding
Zhang, X., Fang, Q., Hu, J., Qian, S., & Xu, C. (2022) · International Journal of Multimedia Information Retrieval · doi:10.1007/s13735-022-00256-3 · Cited by 7
STaR: Knowledge Graph Embedding by Scaling, Translation and Rotation
Li, J., & Yang, Y. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-23504-7_3 · Cited by 8
“MuRE (Balazevic, Allen, and Hospedales 2019a) incorporates rotation with scaling while RotE (Chami et al. 2020) combines rotation and translation.”
BackgroundAn efficient joint framework for interacting knowledge graph and item recommendation
Du, H., Tang, Y., & Cheng, Z. (2022) · Knowledge and Information Systems · doi:10.1007/s10115-022-01808-z · Cited by 7
Knowledge-Based Techniques for Document Fraud Detection: A Comprehensive Study
Tornés, B. M., Boroş, E., Doucet, A., et al. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-24337-0_2 · Cited by 7
Relation-attention semantic-correlative knowledge graph embedding for inductive link prediction
Li, X., Ning, B., Li, G., & Wang, J. (2023) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-023-01865-y · Cited by 7
Hu, K., Zhu, X., Liu, H., Qu, Y., Wang, F. L., & Hao, T. (2023) · IEEE Transactions on Consumer Electronics · doi:10.1109/tce.2023.3302297 · Cited by 7
“Differed from the translational distance-based models, semantic matching models capture interactions between entities by tensor factorization methods [11], [12], [13], [14].”
MethodA structure distinguishable graph attention network for knowledge base completion
Zhou, X., Hui, B., Zhang, L., & Ji, K. (2021) · Neural Computing and Applications · doi:10.1007/s00521-021-06221-1 · Cited by 7
“TuckER [9] decomposes the three-dimensional tensor representing KG into a core tensor and three matrices.”
BackgroundMethodOne-Shot Relation Learning for Knowledge Graphs via Neighborhood Aggregation and Paths Encoding
Sun, J., Zhou, Y. P., & Zong, C. (2021) · ACM Transactions on Asian and Low-Resource Language Information Processing · doi:10.1145/3484729 · Cited by 7
TBicomR: Event Prediction in Temporal Knowledge Graphs with Bicomplex Rotation
Nguyen, N.-T., Tran, C., & Le, T. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112711 · Cited by 6
FTPComplEx: A flexible time perspective approach to temporal knowledge graph completion
Nguyen, N.-T., Ngo, T., Hoàng, N. V., et al. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109717 · Cited by 6
SMAUG: Semantic-Enhanced Mobile App Usage Data Generation With LLM
Huang, Z., Li, T., & Li, Y. (2025) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2025.3648444 · Cited by 8
“To encode this graph, we apply TuckER [24], an advanced tensor factorization method for knowledge graph embedding.”
Transductive Learning for BI-RADS Knowledge Graph based on Knowledge Tensor Factorization
Xi, J., Miao, Z., & Huang, Q. (2021) · International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) · doi:10.1109/cisp-bmei53629.2021.9624217 · Cited by 6
HARPA: hierarchical attention with relation paths for knowledge graph embedding adversarial learning
Zhang, N., Wang, J., & He, J. (2022) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-022-00888-3 · Cited by 6
“TuckER(Balazevicetal.2019) decomposesatensorintoasetofvectorsandacoretensor.”
Kim, G., Kim, S., Kim, K. K., Park, S., Jung, H., & Park, H. (2023) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3580305.3599338 · Cited by 7
“The asymmetric nature of the datasets’ relations is well captured by models such as TransE, ConvE, and TuckER, whereas symmetric models, such as DistMult, lag behind.”
BackgroundInfluentialKnowledge graph completion model based on hyperbolic hierarchical attention network
Luo, J.-H., & Song, C. (2024) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-024-02126-2 · Cited by 6
Do Similar Entities Have Similar Embeddings?
Hubert, N., Paulheim, H., Brun, A., & Monticolo, D. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-60626-7_1 · Cited by 10
Learning Dynamic Knowledge Graph Embedding in Evolving Service Ecosystems via Meta-Learning
Sun, H., Liu, J., Wang, C., Sui, D., Tu, Z., & Xu, X. (2024) · IEEE International Conference on Web Services (ICWS) · doi:10.1109/icws62655.2024.00080 · Cited by 6
“Additionally, RESCAL [17], DistMult [9], ComplEx [10], ANALOGY [18], TuckER [19], and LowFER [20] utilize tensor factorization for modeling knowledge graph completion, with their scoring functions typically comprising inner products between real or complex vectors.”
MethodN-ary relation prediction based on knowledge graphs with important entity detection
Wang, P., Chen, J., Su, L., et al. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.119755 · Cited by 6
A Cybersecurity Knowledge Graph Completion Method for Penetration Testing
Wang, P., Liu, J., Zhong, X., & Zhou, S. (2023) · Electronics · doi:10.3390/electronics12081837 · Cited by 8
Knowledge Graph Completion with Triple Structure and Text Representation
Liu, S., Qin, Y., Xu, M., & Kolmanič, S. (2023) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-023-00271-0 · Cited by 6
“TuckER [17] embeds the entities and relations in a three-dimensional tensor and decomposes it into a product of multiple low-rank matrices.”
BackgroundMLSFF: Multi-level structural features fusion for multi-modal knowledge graph completion
Zhai, H., Lv, X., Hou, Z., Tong, X., & Bu, F. (2023) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2023630 · Cited by 6
Clifford Embeddings – A Generalized Approach for Embedding in Normed Algebras
Demir, C., & Ngomo, A.-C. N. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-43418-1_34 · Cited by 6
Kang, T., Li, S., Tang, J., Li, D., Wang, P., & Wang, T. (2024) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-023-00998-6 · Cited by 6
“For the former, we select translation models TransE (Bordes et al. 2013) & RotatE (Sun et al. 2018), tensor-decomposi-tion models Tucker (Balazevic et al. 2019), DistMult (Yang et al. 2015) and Deep-learning based ConvE (Dettmers et al. 2017).”
Hubert, N., Monnin, P., Brun, A., et al. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-60626-7_2 · Cited by 6
“In this work, several KGEMs are considered: TransE [3], TransH [30], and DistMult [33] using L PHL ; ComplEx [27] and SimplE [16] using L PLL ; ConvE [8], TuckER [2], and RGCN [25]”
MethodZhang, J., Feng, F., Han, T., Duan, F., Sun, Z., Caiafa, C. F., & Solé‐Casals, J. (2021) · Science China Technological Sciences · doi:10.1007/s11431-020-1876-3 · Cited by 6
SpaceE: Knowledge Graph Embedding by Relational Linear Transformation in the Entity Space
Yu, J., Cai, Y., Sun, M., & Li, P. (2022) · ACM Conference on Hypertext and Social Media (HT) · doi:10.1145/3511095.3531284 · Cited by 7
“…et al., 2011), TransH (Wang et al., 2014), RotatE (Sun et al., 2019), BoxE (Abboud et al., 2020)), bi-linear semantic matching methods (Dist-Mult (Yang et al., 2015), ComplEx (Trouillon et al., 2016), TuckER (Balazevic et al., 2019)), and a deep learning method ConvE (Dettmers et al., 2018).”
MethodInfluentialGuo, J., Yu, J. X., Zhao, M., Yu, M., Yu, R., Xu, L., Pan, Y., & Li, X. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112268 · Cited by 5
A data-centric framework of improving graph neural networks for knowledge graph embedding
Cao, Y., Lin, X., Wu, Y., Shi, F., Shang, Y., Tan, Q., Zhou, C., & Zhang, P. (2024) · World Wide Web · doi:10.1007/s11280-024-01320-0 · Cited by 7
“• Semantic Matching Models : DistMult [7], ComplEx [8], TuckER [63], ConvE [9], InteractE [21], and ProcrustEs [64].”
Zero-Shot Link Prediction in Knowledge Graphs with Large Language Models
Li, M., Ling, C., Zhang, R., & Zhao, L. (2024) · Proceedings · doi:10.1109/icdm59182.2024.00088 · Cited by 13
“This is because they learn entities and relation embeddings using the topological structure of the KG. Zero-shot Link Prediction.”
InfluentialStructure-Aware Transformer for hyper-relational knowledge graph completion
Wang, J., Chen, H., & Zhang, W. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.126992 · Cited by 7
Linear self-attention with multi-relational graph for knowledge graph completion
Liu, W., Qiang, B., Chen, R., Xie, Y., Chen, L., & Chen, Z. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06592-1 · Cited by 7
“We compare our LTRGN with state-of-the-art baselines: TransE [27], DistMult [30], ComplEx [55], RotatE [31], and TuckER [56]; CNN-based models: ConvE [52], Inter-actE [57]; GCN-based KGC models: R-GCN [35], which is the first GCN-based KGC model utilizing GCN to handle the impact of different edge…”
Lei, R., Zhou, Y., Wang, Z., & Zhang, X. (2025) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2025.3578483 · Cited by 5
“2) Semantic matching models (e.g., DistMult [65], TuckER [66], and RESCAL [67]), use bilinear or tensor factorization approaches to model interactions between entities and relations.”
SeAttE: An Embedding Model Based on Separating Attribute Space for Knowledge Graph Completion
Liang, Z., Yang, J., Liu, H., Huang, K., Qu, L., Cui, L., & Li, X. (2022) · Electronics · doi:10.3390/electronics11071058 · Cited by 5
Knowledgebra: An Algebraic Learning Framework for Knowledge Graph
Yang, T., Wang, Y., Sha, L., Engelbrecht, J. R., & Hong, P. (2022) · Machine Learning and Knowledge Extraction · doi:10.3390/make4020019 · Cited by 6
A Knowledge-Enhanced Framework for Imitative Transportation Trajectory Generation
Zhu, Q., Chen, Y., Wang, H., Zeng, Z., & Liu, H. (2022) · Proceedings · doi:10.1109/icdm54844.2022.00093 · Cited by 6
“N scoring training procedure, the loss function of TuckER is: LKG = ‖KGi − (W ×1 es ×2 wr ET ) ‖; (5) where E ∈ Rne×de represents entity embedding matrix that is equal to source and object entities, ne and de denote the number of entities and their dimensions.”
BackgroundMethodInfluentialA contrastive knowledge graph embedding model with hierarchical attention and dynamic completion
Shang, B., Zhao, Y., Liu, J., Liu, Y., & Wang, C. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08514-z · Cited by 5
“JoBi ComplEx [21] and TuckER [26] are further improved models based on ComplEx.”
MethodYu, M., Jiang, T., Yu, J. X., Zhao, M., Guo, J., Yang, M., Yu, R., & Li, X. (2023) · Applied Intelligence · doi:10.1007/s10489-023-04723-0 · Cited by 5
“Semantic matching approaches such as DistMult [22], ComplEx [23], and TuckER [24] compute the plausibility of a triple via a similarity-based scoring function over the representations of relations and entities.”
MethodPeng, H., Zeng, W., Tang, J., Wang, M., Huang, H., & Zhao, X. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102729 · Cited by 6
Numerical Literals in Link Prediction: A Critical Examination of Models and Datasets
Blum, M., Ell, B., Ill, H., & Cimiano, P. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-77844-5_2 · Cited by 6
Accurate Coupled Tensor Factorization with Knowledge Graph
Lee, S.-J., Park, Y.-C., & Kang, U. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10825614 · Cited by 5
HyperspherE: An Embedding Method for Knowledge Graph Completion Based on Hypersphere
Dong, Y., Guo, X., Ji, X., Liu, K., & Tang, Z. G. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-82136-4_42 · Cited by 5
A Real-time Inference Method of Graph Attention Network Based on Knowledge Graph for Lung Cancer
Zhang, M. Y., & Du, R. Z. (2021) · doi:10.1145/3458380.3458436 · Cited by 5
“Multi-head attention first introduced by literature [9] is used to stabilize the learning process and encapsulate more information about the neighbor.”
BackgroundFast Hubness-Reduced Nearest Neighbor Search for Entity Alignment in Knowledge Graphs
Obraczka, D., & Rahm, E. (2022) · SN Computer Science · doi:10.1007/s42979-022-01417-1 · Cited by 8
Discover Important Paths in the Knowledge Graph Based on Dynamic Relation Confidence
Yu, S., Wu, Y., Gan, R., Zhou, J., Zheng, Z., & Xuan, Q. (2022) · Communications in computer and information science · doi:10.1007/978-981-19-7532-5_22 · Cited by 5
Metadata
- Authors
- Balazevic, I., Allen, C., & Hospedales, T.
- Affiliations
- University of Edinburgh (GB)
- Venue
- Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
- Published
- 2019
- Pages
- 5184-5193
- OpenAlex
- W2914592219
- Cited by
- 696
- References
- 0
- Citations by year
- 2026: 81 · 2025: 125 · 2024: 165 · 2023: 137 · 2022: 109 · 2021: 55 · 2020: 21 · 2019: 3
- FWCI
- 17.97
- Type
- conference paper
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- arXiv (Cornell University) (submittedVersion)
- Edinburgh Research Explorer (University of Edinburgh) (submittedVersion, cc-by)
- arXiv (Cornell University)
- Funders
- University of Edinburgh
- Engineering and Physical Sciences Research Council (EP/L016427/1)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Advanced Graph Neural Networks
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Mathematics › Computational Mathematics › Tensor decomposition and applications
- Keywords
- knowledge graph completion, tensor decomposition, Tucker, knowledge graphs
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-23, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-23 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
