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
22 Own interpretation · 9 Own result · 6 Sources not found · 3 Partly supported · 3 Own method · 3 No basis given · 2 Supported · 1 No full text of the sources · 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)
boostedcorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- 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)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- 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)
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- 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)
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- 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
plaincorrelational- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
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
“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
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
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
“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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
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, 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
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
“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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- 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)
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It 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
boosted- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
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
“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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It 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.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plainconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It 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.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably 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.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ TuckER: Tensor Factorization for Knowledge Graph Completion > Model
“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 )”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 5.2 Relation to Previous Linear Models
“the relation matrices W r ∈ R de×de are the d e × d e slices of the core tensor Z.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ TuckER: Tensor Factorization for Knowledge Graph Completion > Model
“W ∈ R de×de×dr is the core tensor of a Tucker decomposition.”
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 4 Tucker Decomposition for Link Prediction
“w r ∈ R dr the rows of R representing the relation embedding vector and W ∈ R de×dr×de is the core tensor.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ TuckER: Tensor Factorization for Knowledge Graph Completion > 6.3 Link Prediction Results
“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 > Additional information
“| 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
A Survey on Knowledge Graphs: Representation, Acquisition, and Applications
Ji, S., Pan, S., Cambria, E., Marttinen, P., & Yu, P. S. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3070843 · Cited by 2,947
“nt. & Rel. embed. Scoring Function f r(h;t) RotatE [23] h;t 2Cd, r 2Cd khr tk TorusE [30] [h];[t] 2Tn, [r] 2Tn min (x;y)2([h]+[r]) [t] kx yk i SimplE [47] h;t 2Rd, r;r02Rd 1 2 hrt+ tr0t TuckER [51] h;t 2Rd e , r 2R d r W 1 h 2 r 3 t ITransF [36] h;t 2Rd, r 2Rd H r Dh+ r T r Dt ‘ HolEx [40] h;t 2Rd, r 2Rd P l j=0 p…”
BackgroundMethodInfluentialKnowledge Graphs: Opportunities and Challenges
Peng, C., Xia, F., Naseriparsa, M., & Osborne, F. (2023) · Artificial Intelligence Review · doi:10.1007/s10462-023-10465-9 · Cited by 717
“The core idea of tensor factorization-based methods is transforming the triplets in the knowledge graph into a 3D tensor (Balažević et al. 2019).”
MethodKEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation
Wang, X., Gao, T., Zhu, Z., Zhang, Z., Liu, Z., Li, J., & Tang, J. (2021) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00360 · Cited by 676
“This may be because it uses the TuckER (Balazevic et al., 2019) KE model while ERNIE and KEPLER follow TransE (Bordes et al.”
MethodGenerative artificial intelligence: a systematic review and applications
Sengar, S. S., Hasan, A. B., Kumar, S., & Carroll, F. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-20016-1 · Cited by 763
“In this context, Balazevic et al. [10] introduce TuckER, a linear model that employs Tucker decomposition of the binary tensor representation of knowledge graph triples.”
A Comprehensive Survey on Automatic Knowledge Graph Construction
Zhong, L., Wu, J. X., Li, Q., Peng, H., & Wu, X. (2023) · ACM Computing Surveys · doi:10.1145/3618295 · Cited by 494
“Some models, such as RESCAL [246], TuckerER [247], DistMult [248] and NTN [249], consider matching entity pair representations to a latent relational semantic space for predictions with large graphs.”
MethodA Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal
Liang, K., Meng, L., Liu, M., Liu, Y., Tu, W., Wang, S., Zhou, S., Liu, X., Sun, F., & He, K. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3417451 · Cited by 438
“For instance, SimplE [55] enhances the Canonical Polyadic (CP) decompo-sitional for two independent entity embeddings, and Tucker Decomposition is first used by Tucker [56].”
BackgroundMethodInfluentialHeterogeneous Network Representation Learning: A Unified Framework With Survey and Benchmark
Yang, C., Xiao, Y., Zhang, Y., Sun, Y., & Han, J. (2020) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2020.3045924 · Cited by 379
“…expressive power of RESCAL with the efficiency of DistMult; SimplE [122] considers the inverse of relations and calculates the average score of ð u;l;v Þ and ð v;l (cid:6) 1 ;u Þ ; TuckER [123] proposes to use a three-way Tucker tensor decomposition approach to learning node and relation embeddings.”
MethodStructure-Augmented Text Representation Learning for Efficient Knowledge Graph Completion
Wang, B., Shen, T., Long, G., Zhou, T., Wang, Y., & Chang, Y. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3450043 · Cited by 297
“338 - TuckER [1] .”
BackgroundMethodInfluentialA benchmarking study of embedding-based entity alignment for knowledge graphs
Sun, Z., Zhang, Q., Hu, W. P., Wang, C., Chen, M., Akrami, F., & Li, C. (2020) · Proceedings of the VLDB Endowment · doi:10.14778/3407790.3407828 · Cited by 304
“, DistMult [86], ComplEx [76], HolE [54], SimplE [36], RotatE [71] and TuckER [3]; and (iii) deep models, e.”
A Survey on Knowledge Graph Embeddings for Link Prediction
Wang, M., Qiu, L., & Wang, X. (2021) · Symmetry · doi:10.3390/sym13030485 · Cited by 212
“Based on semantic information No additional information RESCAL [64], DistMult [20], Hole [65], ComplEx [21], ANALOGY [66], ComplEx-N3 [67], TuckER [68], TriModel [69], CrossE [70], HolEx [71], MEI [72].”
BackgroundDual Quaternion Knowledge Graph Embeddings
Cao, Z., Xu, Q., Yang, Z., Cao, X., & Huang, Q. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i8.16850 · Cited by 182
Knowledge Graph Embedding: A Survey from the Perspective of Representation Spaces
Cao, J., Fang, J., Meng, Z., & Liang, S. (2024) · ACM Computing Surveys · doi:10.1145/3643806 · Cited by 208
“Specifically, the definition of group can naturally satisfy the basic properties (e.g., inversion, composition) of knowledge graphs , which means the group-based models should have”
BackgroundAli, M., Berrendorf, M., Hoyt, C. T., Vermue, L., Galkin, M., Sharifzadeh, S., Fischer, A., Volker Tresp, C., & Lehmann, J. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2021.3124805 · Cited by 167
“The skyline comprises an intermediate UM, TransE and DistMult models, and a larger TuckER model.”
BackgroundMethodInfluentialTucker decomposition-based temporal knowledge graph completion
Shao, P., Zhang, D., Yang, G., Tao, J., Che, F., & Liu, T. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.107841 · Cited by 116
“Furthermore, [6] proved that ComplEx is equivalent to TuckER on the premise of imposing certain constraints on the core tensor, or ComplEx is a special case of TuckER.”
BackgroundMethodInfluentialMulti-hop Attention Graph Neural Networks
Wang, G., Ying, R., Huang, J., & Leskovec, J. (2021) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/425 · Cited by 195
“Following the training procedure introduced in [Balazevic et al., 2019; Dettmers et al., 2018], we use 1-”
BackgroundMethodInfluentialIMF: Interactive Multimodal Fusion Model for Link Prediction
Li, X., Zhao, X., Xu, J., Zhang, Y., & Xing, C. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583554 · Cited by 117
“• TuckER [1] applies Tucker decomposition to capture the highlevel interactions between entity and relation embeddings.”
MethodRealistic Re-evaluation of Knowledge Graph Completion Methods: An Experimental Study
Akrami, F., Saeef, M. S., Zhang, Q., Hu, W. P., & Li, C. (2020) · ACM SIGMOD International Conference on Management of Data (SIGMOD) · doi:10.1145/3318464.3380599 · Cited by 122
“5 0.94 0.945 – 5007.3 – 47.9 – 0.261 5277 4992.7 48.0 50.4 0.46 0.429 RotatE [30] – 286.2 – 81.1 – 0.584 309 269.7 95.9 96.0 0.949 0.950 – 3374 – 53.0 – 0.306 3340 3359.8 57.1 57.3 0.476 0.476 TuckER [3] – 484.7 – 80.6 – 0.576 – 468.1 95.8 95.8 0.953 0.950 – 6598 – 46.8 – 0.272 – 6584 52.6 50.2 0.470 0.451 AMIE [13] 1…”
BackgroundMethodInfluentialKnowledge graph embedding by relational rotation and complex convolution for link prediction
Le, T., Le, N., & Le, B. H. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.119122 · Cited by 105
Geometry Interaction Knowledge Graph Embeddings
Cao, Z., Xu, Q., Yang, Z., Cao, X., & Huang, Q. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i5.20491 · Cited by 103
From Discrimination to Generation: Knowledge Graph Completion with Generative Transformer
Xie, X. Y., Zhang, N., Li, Z., Deng, S., Chen, H., Xiong, F., Chen, M., & Chen, H. (2022) · International World Wide Web Conference (WWW) · doi:10.1145/3487553.3524238 · Cited by 102
A Review of Knowledge Graph Completion
Zamini, M., Reza, H., & Rabiei, M. (2022) · Information · doi:10.3390/info13080396 · Cited by 108
“Tucker [31] employs a different decomposition model called a Tucker Decomposition to compute a smaller core tensor and a sequence of three matrices where each matrix represents entity embedding and relation embedding separately.”
MethodAutoSF: Searching Scoring Functions for Knowledge Graph Embedding
Zhang, Y., Yao, Q., Dai, W., & Chen, L. (2020) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde48307.2020.00044 · Cited by 97
“TransE [4], TransH [43], and RotatE [35] from TDMs; NTM [34], Neural LP [47], and ConvE [6] from NNMs; TuckER [1], HolE/HolEX [29], [45], Quat [53], DistMult [46], ComplEx [39], Analogy [24] and SimplE [18] from BLMs; and a rule-based method AnyBURL [27].”
MethodGeneralizing Tensor Decomposition for N-ary Relational Knowledge Bases
Liu, Y., Yao, Q., & Li, Y. (2020) · International World Wide Web Conference (WWW) · doi:10.1145/3366423.3380188 · Cited by 86
“issing facts, KBs face a fundamental issue of incompleteness, which drives KB completion researches especially link prediction: predicting whether two entities are related based on known facts in KBs [2]. Extensive studies have been proposed to address this problem, including translational distance models [6, 15, 21,…”
BackgroundMethodInfluentialKnowledge graph quality control: A survey
Wang, X., Chen, L., Ban, T., Usman, M., Guan, Y., Liu, S., Wu, T., & Chen, H. (2021) · Fundamental Research · doi:10.1016/j.fmre.2021.09.003 · Cited by 99
Looking Beyond IoCs: Automatically Extracting Attack Patterns from External CTI
Alam, T., Bhusal, D., Park, Y. H., & Rastogi, N. (2023) · International Symposium on Research in Attacks, Intrusions and Defenses (RAID) · doi:10.1145/3607199.3607208 · Cited by 104
“These are calculated from the ranks of all true (actual) test triples that TuckER returns.”
MethodInfluentialSelf-attention presents low-dimensional knowledge graph embeddings for link prediction
Baghershahi, P., Hosseini, R., & Moradi, H. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.110124 · Cited by 79
Heterogeneous Federated Knowledge Graph Embedding Learning and Unlearning
Zhu, X., Li, G., & Hu, W. P. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583305 · Cited by 84
“(ii) Semantic matching models [1, 31, 37] match latent semantics of entities and relations in their scoring functions.”
BackgroundComplex graph convolutional network for link prediction in knowledge graphs
Zeb, A., Saif, S., Chen, J., Haq, A. U., Gong, Z., & Zhang, D. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.116796 · Cited by 74
Knowledge graph representation learning with simplifying hierarchical feature propagation
Li, Z., Zhang, Q., Zhu, F., Li, D., Zheng, C., & Zhang, Y. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2023.103348 · Cited by 75
A Survey on Temporal Knowledge Graph: Representation Learning and Applications
Cai, L., Mao, X., Zhou, Y., Long, Z., Wu, C., & Lan, M. (2024) · arXiv · doi:10.48550/arxiv.2403.04782 · Cited by 68
Relational Learning Analysis of Social Politics using Knowledge Graph Embedding
Abu-Salih, B., Al-Tawil, M., Aljarah, I., Faris, H., Wongthongtham, P., Chan, K. Y., & Beheshti, A. (2021) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-021-00760-w · Cited by 73
Hyper-node Relational Graph Attention Network for Multi-modal Knowledge Graph Completion
Liang, S., Zhu, A., Zhang, J., & Shao, J. (2022) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3545573 · Cited by 68
“Factorization models [4, 28] formulate the relationship of triplets as three-way tensor decomposition.”
MethodRole-Aware Modeling for N-ary Relational Knowledge Bases
Liu, Y., Yao, Q., & Li, Y. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3449874 · Cited by 67
“Another model GETD [23] generalizes TuckER [5] to the n-ary case with tensor ring decomposition.”
BackgroundMethodInfluentialInductive Relation Prediction by BERT
Zha, H., Chen, Z., & Yan, X. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i5.20537 · Cited by 67
“For TuckER and MINERVA, we assign a minimum score for both positive and negative triples containing unseen entities.”
MethodInfluentialDRGI: Deep Relational Graph Infomax for Knowledge Graph Completion
Liang, S., Shao, J., Zhang, D., Zhang, J., & Cui, B. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3110898 · Cited by 63
“TuckER [9] utilizes three-way Tucker tensor decomposition to learn embedding by output-ting a core tensor W and embedding vectors of entities and relations, and its form of score function is defined as f r ð h;t Þ ¼ W(cid:4) 1 h (cid:4) 2 r (cid:4) 3 t , where W 2 R d e (cid:4) d r (cid:4) d e is…”
MethodTGformer: A Graph Transformer Framework for Knowledge Graph Embedding
Shi, F., Li, D., Wang, X., Li, B., & Wu, X. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3486747 · Cited by 63
“Another interestingperspectiveistoviewknowledgegraphembeddingasathree-waytensordecompositionproblem,whereTuckER[10] employstheTuckertensordecompositiontoinversetheentity-relationinteractionprocess.”
InfluentialTemporal knowledge graph representation learning with local and global evolutions
Zhang, J., Liang, S., Sheng, Y., & Shao, J. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.109234 · Cited by 60
UrbanKG: An Urban Knowledge Graph System
Liu, Y., Ding, J., Fu, Y., & Li, Y. (2023) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3588577 · Cited by 72
“To validate the efectiveness of the Construction layer and the Algorithm layer in the UrbanKG system, we adopt the TuckER model [4] to learn entity embeddings on the UrbanKGs for Beijing.”
BackgroundMethodInfluentialActionable Cyber Threat Intelligence Using Knowledge Graphs and Large Language Models
Fieblinger, R., Alam, T., & Rastogi, N. (2024) · IEEE European Symposium on Security and Privacy Workshops (EuroS&PW) · doi:10.1109/eurospw61312.2024.00018 · Cited by 88
“The evaluation utilized the TuckER [46] model within a transductive setting across two distinct test sets.”
MethodKRACL: Contrastive Learning with Graph Context Modeling for Sparse Knowledge Graph Completion
Tan, Z., Chen, Z., Feng, S., Zhang, Q., Zheng, Q., Li, J., & Luo, M. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583412 · Cited by 62
“We compare our KRACL with state-of-the-art KGE models, including non-neural model TransE [7], DistMult [80], ComplEx [61], RotatE [56], TuckER [4], OTE [58], HAKE [87], and GIE [10]; neural network-based model ConvE [16], HypER [2], RGCN [53], KBAT [42], CompGCN [63], HittER [12], DisenKGAT [73], and CAKE [46].”
MethodZhu, A., Ouyang, D., Liang, S., & Shao, J. (2022) · Knowledge-Based Systems · doi:10.1016/j.knosys.2022.108843 · Cited by 54
Cui, H. N., Peng, T., Xiao, F., Han, J., Han, R., & Liu, L. (2022) · Information Sciences · doi:10.1016/j.ins.2022.11.042 · Cited by 55
DAPath: Distance-aware knowledge graph reasoning based on deep reinforcement learning
Tiwari, P., Zhu, H., & Pandey, H. M. (2020) · Neural Networks · doi:10.1016/j.neunet.2020.11.012 · Cited by 55
“Tensor decomposition models [26], such as DistMult [27], ComplEx [28], Analogy [29], SimplE [30], HolE [31], TuckER [32], and deep learning models, such as ConvE [33], ConvKB [34], ConvR [35], CapsE [36], RSN [37], are also used to represent entities and relations.75 For the path-based method, Lao et al. [38, 39] prop…”
MethodMulDE: Multi-teacher Knowledge Distillation for Low-dimensional Knowledge Graph Embeddings
Wang, K., Liu, Y., Ma, Q., & Sheng, Q. Z. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3449898 · Cited by 54
“[2] propose a linear model based on Tucker decomposition of the binary tensor representation of knowledge graph triples.”
MethodGeoscience Knowledge Graph (GeoKG): Development, construction and challenges
Zhang, X., Huang, Y., Zhang, C., & Ye, P. (2022) · Transactions in GIS · doi:10.1111/tgis.12985 · Cited by 53
“Conventional methods mainly extract basic geoscience knowledge from text data, such as location, time, attribute, event, geo-entity, and so forth (Balažević et al., 2019; Li et al., 2021; Yang et al., 2020).”
MethodWang, J., Zhang, Q., Shi, F., Li, D., Cai, Y., Wang, J., Li, B., Wang, X., Zhang, Z., & Zheng, C. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2023.103350 · Cited by 55
Learning knowledge graph embedding with a dual-attention embedding network
Fang, H., Wang, Y., Tian, Z., & Ye, Y. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.118806 · Cited by 52
Decoupled semantic graph neural network for knowledge graph embedding
Li, Z., Huang, W., Gong, X., Luo, X., Xiao, K., Deng, H., Zhang, M., & Zhang, Y. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128614 · Cited by 51
Convolutional Complex Knowledge Graph Embeddings
Demir, C., & Ngomo, A.-C. N. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-77385-4_24 · Cited by 50
Knowledge graph embedding by relational and entity rotation
Huang, X., Tang, J., Tan, Z., Zeng, W., Wang, J., & Zhao, X. H. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.107310 · Cited by 47
Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning
Chen, X., Boratko, M., Chen, M., Dasgupta, S. S., Li, X. L., & McCallum, A. (2021) · Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2021.naacl-main.68 · Cited by 47
Contrastive Multi-Modal Knowledge Graph Representation Learning
Fang, Q., Zhang, X., Hu, J., Wu, X., & Xu, C. (2022) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2022.3220625 · Cited by 48
“TuckER [26] is a linear model based on the TuckER tensor decomposition of the binary tensor representation of KG triples.”
BackgroundMethodAnytime bottom-up rule learning for large-scale knowledge graph completion
Meilicke, C., Chekol, M. W., Betz, P., Fink, M., & Stuckenschmidt, H. (2023) · The VLDB Journal · doi:10.1007/s00778-023-00800-5 · Cited by 44
“TransE [8] is a seminal model for the translation-based approaches, TuckEr [4] relies on tensor factorizations and HolE [38] combines circular correlations with matrix multiplications.”
MethodZhu, Y., Lin, Q., Lü, H., Shi, K., Liu, D., Chambua, J., Wan, S., & Niu, Z. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3125424 · Cited by 43
“:3) c et al. [36] proposed a tensor decomposition-base approach, namely TuckER .”
MethodKnowledge-infused Contrastive Learning for Urban Imagery-based Socioeconomic Prediction
Liu, Y., Zhang, X., Ding, J., Xi, Y., & Li, Y. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583876 · Cited by 48
“Besides, we use pretrained embeddings from TuckER [3] for initialized embeddings.”
MethodCogKR: Cognitive Graph for Multi-hop Knowledge Reasoning
Du, Z., Zhou, C., Yao, J., Tu, T., Cheng, L., Yang, H., Zhou, J., & Tang, J. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3104310 · Cited by 44
“Following that, more advanced scoring functions have been proposed, such as vector difference [8], [12], vector product [9], [31], convolution [13], [32], and tensor operation [14], [33].”
BackgroundZhang, F., Zhang, Z., Ao, X., Zhuang, F., Xu, Y., & He, Q. (2022) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/3511808.3557233 · Cited by 46
“Tucker [3] is a tucker-decomposition-based method for KGE.”
MethodMultisource hierarchical neural network for knowledge graph embedding
Jiang, D., Wang, R., Xue, L., & Yang, J. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121446 · Cited by 41
HJE: Joint Convolutional Representation Learning for Knowledge Hypergraph Completion
Li, Z., Wang, C., Wang, X., Chen, Z., & Li, J. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3365727 · Cited by 42
“RotatE[36]treatstherelation asarotationoperationwhenprojectingheadandtailentities intocomplexspace,whileTuckER[37]andAutoSF[38]in- troducesTuckERdecompositionandautomatedrelationma-trixintoknowledgegraphrepresentationlearning,respectively.”
BackgroundNeural-Symbolic Methods for Knowledge Graph Reasoning: A Survey
Cheng, K., Ahmed, N. K., Rossi, R. A., Willke, T. L., & Sun, Y. (2024) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3686806 · Cited by 50
“TuckER [4], on the other hand, employs Tucker decomposition to factorize the third-order tensor comprising KG triples for learning entity and relationship representations.”
Inductive Learning on Commonsense Knowledge Graph Completion
Wang, B., Wang, G., Huang, J., You, J., Leskovec, J., & Kuo, C. ‐. J. (2021) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn52387.2021.9534355 · Cited by 45
Cybersecurity knowledge graphs construction and quality assessment
Li, H., Shi, Z., Pan, C., Zhao, D., & Sun, N. (2023) · Complex & Intelligent Systems · doi:10.1007/s40747-023-01205-1 · Cited by 50
“…feed-forward to generate the final output representation H output for each triple of the knowledge graphs: where During the decoding phase, based on the simplicity and efficiency of the Tucker model [37], we construct the knowledge graph quality assessment model on the basis of the Tucker model.”
Knowledge Transfer via Pre-training for Recommendation: A Review and Prospect
Zeng, Z., Xiao, C., Yao, Y., Xie, R., Liu, Z., Lin, F., Lin, L., & Sun, M. (2021) · Frontiers in Big Data · doi:10.3389/fdata.2021.602071 · Cited by 38
“In order to exploit KGs, one line of KG-based methods seeks to encode the KG into low-dimensional pre-trained embeddings with the knowledge graph embedding (KGE) methods (Bordes et al., 2013; Wang et al., 2014; Lin et al., 2015; Ji et al., 2015; Trouillon et al., 2016; Shi and Weninger, 2017; Balazevic et al., 2019).”
MethodAn improving reasoning network for complex question answering over temporal knowledge graphs
Jiao, S., Zhu, Z., Wu, W., Zuo, Z., Qi, J., Wang, W.-L., Zhang, G., & Liu, P. (2022) · Applied Intelligence · doi:10.1007/s10489-022-03913-6 · Cited by 41
Yang, R., Zhu, J., Man, J., Liu, H., Li, F., & Zhou, Y. J. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102868 · Cited by 42
Searching to Sparsify Tensor Decomposition for N-ary Relational Data
Di, S., Yao, Q., & Chen, L. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3449853 · Cited by 37
“But existing tensor decomposition models [3, 25] can only learn embeddings from facts with a specific arity n, while the N-ary relational data usually contains facts with different arities [30, 39].”
BackgroundMethodInfluentialXie, Z., Zhu, R., Liu, J., Zhou, G., & Huang, J. X. (2022) · Information Processing & Management · doi:10.1016/j.ipm.2022.103076 · Cited by 36
Khan, A. (2023) · ACM SIGMOD Record · doi:10.1145/3615952.3615956 · Cited by 41
Iterative rule-guided reasoning over sparse knowledge graphs with deep reinforcement learning
Xia, Y., Lan, M., Luo, J., Chen, X., & Zhou, G. (2022) · Information Processing & Management · doi:10.1016/j.ipm.2022.103040 · Cited by 34
From data to insights: the application and challenges of knowledge graphs in intelligent audit
Zhong, H., Yang, D., Shi, S., Wei, L., & Wang, Y. (2024) · Journal of Cloud Computing Advances Systems and Applications · doi:10.1186/s13677-024-00674-0 · Cited by 53
“Tucker [63] utilizes a different decomposition model called Tucker decomposition to compute a smaller core tensor and a sequence composed of three matrices, each representing embed-dings for entities and relations.”
MethodHierarchical graph attention network for temporal knowledge graph reasoning
Shao, P., He, J., Li, G., Zhang, D., & Tao, J. (2023) · Neurocomputing · doi:10.1016/j.neucom.2023.126390 · Cited by 34
LAFA: Multimodal Knowledge Graph Completion with Link Aware Fusion and Aggregation
Shang, B., Zhao, Y., Liu, J., & Wang, D. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i8.28744 · Cited by 41
Node Co-occurrence based Graph Neural Networks for Knowledge Graph Link Prediction
Nguyen, D. Q., Tong, V. V., Phung, D., & Nguyen, D. Q. (2022) · ACM International Conference on Web Search and Data Mining (WSDM) · doi:10.1145/3488560.3502183 · Cited by 33
“, 2018), and TuckER (Balažević et al., 2019) are taken from (Safavi and Koutra, 2020).”
BackgroundMethodAutomatic detection of COVID-19 vaccine misinformation with graph link prediction
Weinzierl, M., & Harabagiu, S. M. (2021) · Journal of Biomedical Informatics · doi:10.1016/j.jbi.2021.103955 · Cited by 40
TARGAT: A Time-Aware Relational Graph Attention Model for Temporal Knowledge Graph Embedding
Xie, Z., Zhu, R., Liu, J., Zhou, G., & Huang, J. X. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3282101 · Cited by 34
“Bilinear models [23], [24], [34], [35], [36], [61], [62] leverage bilinear score functions to represent linear transformations on the embedding of entities.”
BackgroundRoAN: A relation-oriented attention network for temporal knowledge graph completion
Bai, L., Ma, X., Meng, X., Ren, X., & Ke, Y. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106308 · Cited by 32
Lu, H., Hu, H., & Lin, X. S. (2022) · Neurocomputing · doi:10.1016/j.neucom.2021.12.079 · Cited by 48
“In addition to these methods, we also compare our model with more recently proposed KGE models (denoted as Recent), such as RotatE (Sun et al., 2019), QuatE (Jia, 2019), D4-STE (Xu and Li, 2019), TuckER (Balažević et al., 2019), Rotate3D (Gao et al., 2020) and HAKE (Zhang et al., 2020).”
MethodFully automated diagnosis of thyroid nodule ultrasound using brain-inspired inference
Li, G., Huang, Q., Liu, C., Wang, G., Guo, L., Liu, R., & Liu, L. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127497 · Cited by 32
Position-Aware Relational Transformer for Knowledge Graph Embedding
Li, G., Sun, Z., Hu, W. P., Cheng, G., & Qu, Y. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3262937 · Cited by 33
“TuckER [1] adopts another tensor decomposition called Tucker decomposition, which is beneficial for embedding learning across relations.”
BackgroundA survey on visual transfer learning using knowledge graphs
Monka, S., Halilaj, L., & Rettinger, A. (2022) · Semantic Web · doi:10.3233/sw-212959 · Cited by 36
“Therefore, recently standard KGE-Methods have been adopted to directly operate on hyper-relational graphs (m-TransH [33], HypE, HSimple [32], RAE [34], GETD[35], TuckER [36], NaLP[37], HINGE[31], StarE [38]) and hypergraphs (HEBE [39], HGE [40], Hyper2vec [41], HNN [42], HCN [43], DHNE [44], HHNE [45], Hyper-SAGNN [46…”
MethodCui, H. N., Peng, T., Han, R., Zhu, B., Bi, H., & Liu, L. (2023) · Information Processing & Management · doi:10.1016/j.ipm.2023.103283 · Cited by 31
Link Prediction with Attention Applied on Multiple Knowledge Graph Embedding Models
Gregucci, C., Nayyeri, M., Hernández, D., & Staab, S. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583358 · Cited by 38
Modeling Heterogeneous Hierarchies with Relation-specific Hyperbolic Cones
Bai, Y., Ying, R., Ren, H., & Leskovec, J. (2021) · arXiv · doi:10.48550/arxiv.2110.14923 · Cited by 31
“Recent works in this area propose the use of a variety of embedding geometries such as hyperbolic embeddings, box embeddings, and cone embeddings [10, 11, 12] to model partial ordering property of hierarchical relations, but two important challenges remain: (1) Existing works that consider hierarchical relations [13]…”
BackgroundMethodInfluentialOpen-world knowledge graph completion with multiple interaction attention
Niu, L., Fu, C., Yang, Q., Li, Z., Chen, Z., Liu, Q., & Zheng, K. (2020) · World Wide Web · doi:10.1007/s11280-020-00847-2 · Cited by 30
Bilinear Scoring Function Search for Knowledge Graph Learning
Zhang, Y., Yao, Q., & Kwok, J. T.-Y. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3157321 · Cited by 31
“AutoBLM and AutoBLM+ are compared with the following popular human-designed KG embedding models(3): (i) TDM, including TransH [11], RotatE [13] and PairE [61]; (ii) NNM, includingConvE [21], RSN [22] andCompGCN [24]; (iii) BLM, including TuckER [35], Quat [19], DistMult [7], ComplEx [14], HolE [15], Analogy [16] Simpl…”
BackgroundMethodInfluentialHui, S., Wang, H., Li, T., Yang, X., Wang, X., Feng, J., Zhu, L., Deng, C., Hui, P., Jin, D., et al. (2023) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3580305.3599853 · Cited by 31
“Based on the constructed KG, we learn the embedding of each base station via TuckER [4].”
MethodBalanced Unfolding Induced Tensor Nuclear Norms for High-Order Tensor Completion
Qiu, Y., Zhou, G., Wang, A., et al. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3373384 · Cited by 29
“…from incomplete tensors, and has been widely used in a variety of research fields, e.g., multiiew clustering [6], [7], [8], knowledge graph completion [9], [10], [11], traffic data com2162-237X pletion [12], [13], [14], hyperspectral image restoration [15], [16], [17], [18], [19], and so on.”
Knowledge Graph Completion with Counterfactual Augmentation
Chang, H., Cai, J., & Li, J. (2023) · International World Wide Web Conference (WWW) · doi:10.1145/3543507.3583401 · Cited by 29
“Embeddings TransE [5] DistMult [48] ComplEx [42] RotatE [37] HAKE [51] LowFER [2] TuckER [4] ConvE [12] 0.”
BackgroundEIGAT: Incorporating global information in local attention for knowledge representation learning
Zhao, Y., Feng, H., Zhou, H., Yang, Y., Chen, X., Xie, R., Zhuang, F., & Li, Q. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.107909 · Cited by 27
Multi-Hop Reasoning With Relation Based Node Quality Evaluation for Sparse Medical Knowledge Graph
Zhang, T., Cheng, J., Miao, L., Chen, H., Li, Q., He, Q., Lyu, J., & Ma, L. (2024) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2024.3452748 · Cited by 33
Knowledge graph embedding with the special orthogonal group in quaternion space for link prediction
Le, T., Tran, H., & Le, B. H. (2023) · Knowledge-Based Systems · doi:10.1016/j.knosys.2023.110400 · Cited by 28
Construction and Applications of Billion-Scale Pre-Trained Multimodal Business Knowledge Graph
Deng, S., Wang, C., Li, Z., Zhang, N., Dai, Z., Chen, H., Xiong, F., Yan, M., Chen, Q., Chen, M., et al. (2023) · 2023 IEEE 39th International Conference on Data Engineering (ICDE) · doi:10.1109/icde55515.2023.00229 · Cited by 18
“We notice that TuckER can archive higher Hits@K and MRR scores due to the powerful representation ability of tucker decomposition.”
BackgroundMethodInfluentialReasoning Through Memorization: Nearest Neighbor Knowledge Graph Embeddings
Wang, P., Xie, X. Y., Wang, X., & Zhang, N. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-44693-1_9 · Cited by 33
Hierarchical Neighbor Propagation With Bidirectional Graph Attention Network for Relation Prediction
Xie, Z., Zhu, R., Liu, J., Zhou, G., & Huang, J. X. (2021) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2021.3079812 · Cited by 25
“TuckER [39] is then proposed to learn embeddings using threeway Tucker tensor decomposition, which includes a core tensor and the vectors of entities and relations.”
MethodKnowledge embedding via hyperbolic skipped graph convolutional networks
Yao, S., Pi, D., & Chen, J. (2022) · Neurocomputing · doi:10.1016/j.neucom.2022.01.037 · Cited by 24
Duality-Induced Regularizer for Semantic Matching Knowledge Graph Embeddings
Wang, J., Zhang, Z., Shi, Z., Cai, J., Ji, S., & Wu, F. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3161804 · Cited by 26
“For example, ConT [25] extends TuckER [26] by additionally learning timestamp embeddings.”
BackgroundA Novel Tensor Learning Model for Joint Relational Triplet Extraction
Wang, Z., Nie, H., Zheng, W., Wang, Y., & Li, X. (2023) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2023.3265851 · Cited by 27
“For example, MUTAN [59] utilizes Tucker decomposition to parametrize bilinear interactions between representations of image and text for the visual question answering task; TuckER [60] employs Tucker decomposition to model correlations between entities and relations in the link prediction task.”
MethodAn attention mechanism and residual network based knowledge graph-enhanced recommender system
Li, W., Zhong, H., Zhou, J., Chang, C., Lin, R., & Tang, Y. Z. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112042 · Cited by 25
Large Language Model Enhanced Knowledge Representation Learning: A Survey
Wang, X., Chen, Z., Wang, H., Leong Hou, U., Li, Z., & Guo, W. (2025) · Data Science and Engineering · doi:10.1007/s41019-025-00285-y · Cited by 37
Learning graph attention-aware knowledge graph embedding
Li, C., Peng, X., Niu, Y., Zhang, S., Peng, H. M., Zhou, C. D., & Li, J. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.01.139 · Cited by 24
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
