Pith. sign in

REVIEW 4 major objections 6 minor 35 references

Efficient Relational Context Perception for Knowledge Graph Completion

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that replacing a Transformer context encoder with a recurrent TRP encoder and scoring triples with a Tucker decomposition core tensor yields knowledge graph completion results that match or exceed several…

desk verdict A useful but thinly evidenced RWKV-for-KGC engineering paper: the combination is new and the ablations are consistent, but the headline efficiency claim is unverifiable as presented and the baseline comparisons are not apples-to-apples. read the letter →

arxiv 2501.00397 v1 pith:CUKVXQG2 submitted 2024-12-31 cs.LG cs.AIcs.CL

Wenkai Tu , Guojia Wan , Zhengchun Shang , Bo Du This is my paper · ORCID
classification cs.LGcs.AIcs.CL
keywords knowledgegraphcompletionlinkpredictiontripleclassificationrecurrentencoderReceptancePerceptionTuckerdecompositioncontextualembeddingsparameterefficiency
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Knowledge graph completion models typically assign one static embedding per entity and relation, or pay the high cost of a Transformer encoder to make those embeddings context-dependent. This paper proposes Triple Receptance Perception (TRP), a recurrent encoder that dynamically models sequential context in triples, and pairs it with a Tucker decomposition decoder for triple scoring. The authors report that this combination matches or surpasses several state-of-the-art models on link prediction (FB15k, YAGO3-10, UMLS) and triple classification (FB13, FB15k), while needing fewer parameters and less training time than a Transformer-based context encoder. If these results hold under a common evaluation protocol, the work shows that recurrence, not attention, can carry the context modelling needed for competitive knowledge graph completion.

What carries the argument

The load-bearing object is the TRP block, a residual recurrent block whose time-mixing sub-block computes a causal, exponentially-decayed attention-style weighted sum (key–value with a learnable decay vector) and whose channel-mixing sub-block applies a squared ReLU; both mix the current input with its predecessor through interpolation factors. This recurrence keeps inference linear in sequence length and is what the paper credits for the parameter and time savings over a Transformer encoder. The decoder is a Tucker decomposition core tensor, scoring a triple as the n-mode product of the encoded head and relation representations with the tail embedding, which the ablation identifies as outperforming MLP, TransE, DistMult, and ComplEx scoring functions on FB15k.

What would settle it

Retrain the main baselines and the proposed model from one codebase with identical reciprocal augmentation, 1-N scoring, filtered ranking, and hyperparameter search on FB15k and YAGO3-10; if the baselines meet or exceed the reported MRR of 0.85 and 0.57 respectively, the paper's claim of consistent outperformance is not supported.

Watch

Extended reading notes

Core claim

The central discovery, as the paper states it, is that a recurrent context encoder (TRP) combined with a tensor-decomposition decoder offers a better accuracy-efficiency trade-off than both static embedding models and Transformer-based context models for knowledge graph completion. On FB15k the model reaches MRR 0.85, tying the Transformer-based baseline while reporting lower Hits@1/10; on YAGO3-10 it reports the best MRR of 0.57 and Hits@10 of 70.0; on UMLS the best MRR of 0.95 and Hits@10 of 99.9; and on FB13 triple classification accuracy of 88.6%. The authors take these numbers as evidence that context-dependent embeddings produced by a linear-cost recurrent mechanism, together with relational decoding by a learned core tensor $\phi(h,r,t)=W_c\times_1 \tilde e_h \times_2 \tilde e_r \times_3 e_t$, give more expressive representations than static embeddings or attention-based contextualization.

Load-bearing premise

The paper's claim of consistent outperformance depends on the assumption that the baseline results cited from other papers were obtained under the same evaluation rules as its own experiments.

Editorial extensions

If this is right

  • On YAGO3-10 and UMLS, TRP plus tensor decoding reports the highest MRR and Hits@10 among the listed baselines, suggesting recurrence-based context encoding scales to large entity sets.
  • On FB15k, the model matches the Transformer baseline on MRR (0.85) with lower Hits@1/10, so the main claimed advantage there is parameter and training efficiency rather than raw accuracy.
  • Because the encoder is recurrent, inference cost grows linearly rather than quadratically with context length, and the multi-hop training-time experiment reports slower growth than a Transformer.
  • The ablation shows each component contributes: removing the Tucker decoder drops FB15k MRR from 0.85 to 0.82, and removing the TRP encoder drops it to 0.79, leaving a plain tensor-factorization model.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same TRP block could encode longer paths or neighborhoods without re-training the whole model, since its recurrence accepts variable-length input; the paper's multi-hop experiment measures training time but not the resulting link-prediction accuracy on long paths.
  • If parameter efficiency transfers, TRP could be used as a drop-in replacement for Transformer encoders in other triple-based tasks, such as fact checking or relation extraction, though that is beyond the paper's experiments.
  • The comparison against the Transformer-based context encoder on FB15k suggests that on dense graphs the practical win may be cost, not accuracy; on sparse and large graphs like YAGO3-10 the win may be both, but this distinction is an inference from the reported tables.
  • A re-evaluation with all baselines trained under identical hyperparameter search and 1-N filtered settings could narrow or change the reported margins; that is a testable consequence of the way the baseline numbers are currently assembled.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes Triple Receptance Perception (TRP), a recurrent RWKV-style encoder with time-mixing and channel-mixing blocks, coupled with a Tucker decomposition decoder, for knowledge graph completion. It uses reciprocal augmentation and 1-N scoring during training and reports link prediction results on UMLS, FB15k, and YAGO3-10 plus triple classification on FB13 and FB15k. The paper claims consistent improvements over several state-of-the-art models, higher parameter efficiency and lower training time than Transformer encoders, and presents ablations, sensitivity analyses, and embedding visualizations.

Significance. If substantiated by reproducible experiments, the proposed architecture is a plausible efficient alternative to Transformer-based context encoders for KGC, with linear-time encoding and a compact Tucker decoder. The paper uses standard public datasets, selects hyperparameters on a dev set, and includes ablation and sensitivity analyses. However, the central claims are currently supported only by copied baseline numbers, unresolved figure references, and single runs without variance; the claimed efficiency advantage is not visible in the manuscript. The contribution is incremental relative to RWKV and TuckER, but the combination is reasonable and worth publishing if the comparisons are made apples-to-apples.

major comments (4)
  1. [Section 4.2.1, Table 2] The claim that the method 'consistently outperforms' prior models is not established by Table 2. Most baseline numbers are taken from the original papers and the footnote says '-' means results could not be accessed from original code or paper; no baseline is retrained under the same protocol. TRP is trained with reciprocal augmentation and 1-N scoring (Section 3.3), which doubles the training triples and changes the filtered candidate set, so the small margins over literature numbers may reflect evaluation-protocol differences rather than the TRP encoder. On FB15k, TRP ties CoKE on MRR (0.85) and is worse on H@1 (81.2 vs. 82.6) and H@10 (90.3 vs. 90.6), directly contradicting the 'consistently outperforming' wording in the text; the authors should rerun the baselines under an identical protocol or restrict their claims to matched comparisons.
  2. [Section 4.2.5] The efficiency and parameter-count comparisons are load-bearing for the paper's title and abstract, but the section refers twice to missing figures ('Figure ??'), and no training-time or parameter-count numbers appear elsewhere. The claim that TRP is more parameter-efficient and faster than a Transformer cannot be checked without the actual plots, axis labels, the multi-hop data-generation procedure, and preferably multiple runs or confidence intervals. These should be provided before the efficiency claim is asserted.
  3. [Section 4.1.4, Tables 2 and 3] The final hyperparameters for each dataset are not reported; Section 4.1.4 gives only search ranges for embedding size, TRP blocks, dropout, and learning rate, and the paper reports a single run per setting with no standard deviations or significance tests. Since the headline margins are small (e.g., 0.9 points on FB13 and -0.3 points on FB15k in Table 3, and tied or worse FB15k numbers in Table 2), the absence of variance estimates leaves the main empirical claim unquantified; the authors should provide the selected hyperparameters, number of seeds, and error bars.
  4. [Section 4.2.2, Table 3] The triple-classification evaluation is underspecified. The negative triples for FB15k are said to be constructed after [19], but the construction details are not given, and the decision rule is internally inconsistent: it says a triple is classified as false when its dissimilarity score is below delta_r, yet the model produces a similarity logit from the Tucker decoder and no dissimilarity mapping is defined. The reported accuracy numbers in Table 3 are therefore not interpretable without a precise decision rule and negative-sampling description.
minor comments (6)
  1. [Section 3.1.3, Eq. (12)] The output equation uses W_{v'}, but the surrounding text lists Wr', Wk', and Wr' as the separate weighting matrices; one of these should be Wv'.
  2. [Section 4.1.1 and Table 1] The text states that YAGO3-10 has 46 relations, while Table 1 lists 37; please reconcile these numbers.
  3. [Section 4.2.4] The sensitivity text says 2, 4, 6, 8, and 10 layers were tested, but the left panel of Figure 3 shows only 2, 4, 6, and 8 layers on the x-axis; the text and figure should be aligned.
  4. [Section 3.1.2] The notation t in {h,r} is confusing because t is also used for the tail entity and for time-step indices elsewhere; renaming the sequence positions would improve readability.
  5. [Section 5] The conclusion says the results 'prove' that the integration allows more expressive representations; for single-run benchmark comparisons without error bars, 'support' or 'indicate' would be more appropriate.
  6. [Section 4.2.6] The t-SNE figures are presented as evidence of semantic separation, but visual cluster separation of embeddings is a qualitative result; the text should avoid claiming that the model 'has successfully captured semantic relationships' solely from this visualization.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: empirical evaluation on held-out benchmark triples; design components cited from external work; weaknesses are baseline comparability, not derivation circularity.

full rationale

The paper's central claims are empirical: the TRP encoder plus Tucker decoder obtains MRR and Hits@k on held-out test triples from FB15k, YAGO3-10, UMLS, and FB13, plus triple-classification accuracy. Nothing in the method section defines the evaluation metric in terms of a fitted parameter or a self-citation. The TRP equations (1)-(14) are standard RWKV-style time and channel mixing adopted from external references [30,31], and the decoder (15) is Tucker decomposition; the target predictions are softmax scores over all entities trained with cross-entropy (16)-(17) and evaluated by filtered ranking. Hyperparameters are chosen on the dev set by MRR (Section 4.1.4), which is legitimate model selection, not fitting the test set. The citations that are structurally load-bearing, namely RWKV, Attention-Free Transformer, and Tucker decomposition, are external algorithm definitions, and the paper invokes no uniqueness theorem or prior result by the same authors to force its design. The claim of consistent outperformance is weakened by baseline protocol differences and small margins (e.g., on FB15k MRR is tied with CoKE at 0.85 while H@1 and H@10 are lower), but that is a comparability and correctness concern, not circularity: the reported numbers are not constructed from the baselines' outputs. Missing figure references in Section 4.2.5 also undermine the parameter-efficiency evidence without making the derivation circular. Overall, the paper is self-contained against external benchmarks, so no circularity is found.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The ledger is dominated by unstated experimental settings rather than by invented entities. The method has no parameter-free derivation; all scores come from trained weights and per-dataset hyperparameter choices. The most consequential assumption for the reported comparison is that baseline numbers from other papers are directly comparable.

free parameters (5)
  • embedding dimension per dataset = not reported (searched over 64, 96, 128, 192, 256)
    Selected by dev MRR per Section 4.1.4; the final configuration is not given, so the reported numbers cannot be rerun exactly.
  • number of TRP blocks = not reported (searched over 2, 4, 6, 8)
    Final per-dataset layer count is omitted; Figure 3 suggests layer count has little effect.
  • dropout rate = not reported (searched over 0.2, 0.3, 0.4, 0.5)
    Chosen per dataset by dev MRR; the final value is not listed.
  • learning rate = not reported (range 0.0005 to 0.01)
    Per-dataset learning rate is chosen, but the final value is not listed.
  • relation-specific thresholds for triple classification = not reported
    The threshold delta_r is fit per relation on validation data in Section 4.2.2; the fitted values are absent.
assumptions (5)
  • standard math The time-mixing recurrence in Eqs. (3)-(8) is a valid, correctly normalized attention-free pooling over the two input positions.
    The method assumes the RWKV/AFT recurrence computes the claimed weighted sums and that initial states a0=b0=0 are handled consistently.
  • domain assumption Representing a triple by the ordered sequence (head, relation) is sufficient context for link prediction, without graph neighborhoods or textual context.
    The method's dynamic context is only the two-token sequence; if the ordering and interaction of h and r are not informative, the encoder contributes little.
  • domain assumption A single shared Tucker core Wc and per-entity tail embeddings can score all candidate tails after head and relation encoding.
    The decoder design assumes Tucker decomposition captures the needed relational interactions; benchmark results depend on this modeling choice.
  • domain assumption Baseline numbers transcribed from prior papers are comparable to the authors' experimental protocol.
    No baselines are retrained; the Table 2 footnote admits missing results, so comparability is assumed rather than demonstrated.
  • domain assumption t-SNE separation of entity categories indicates that the model has captured semantic structure.
    The visualization section treats clustering as evidence of effectiveness, although t-SNE clusters can arise from geometric artifacts.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Efficient Relational Context Perception for Knowledge Graph Completion." pith.science (2026). https://pith.science/paper/CUKVXQG2

@misc{pith2026250100397,
  author       = {Pith},
  title        = {Pith review of: Efficient Relational Context Perception for Knowledge Graph Completion},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CUKVXQG2}},
  note         = {Machine review of arXiv:2501.00397}
}
read the original abstract

Knowledge Graphs (KGs) provide a structured representation of knowledge but often suffer from challenges of incompleteness. To address this, link prediction or knowledge graph completion (KGC) aims to infer missing new facts based on existing facts in KGs. Previous knowledge graph embedding models are limited in their ability to capture expressive features, especially when compared to deeper, multi-layer models. These approaches also assign a single static embedding to each entity and relation, disregarding the fact that entities and relations can exhibit different behaviors in varying graph contexts. Due to complex context over a fact triple of a KG, existing methods have to leverage complex non-linear context encoder, like transformer, to project entity and relation into low dimensional representations, resulting in high computation cost. To overcome these limitations, we propose Triple Receptance Perception (TRP) architecture to model sequential information, enabling the learning of dynamic context of entities and relations. Then we use tensor decomposition to calculate triple scores, providing robust relational decoding capabilities. This integration allows for more expressive representations. Experiments on benchmark datasets such as YAGO3-10, UMLS, FB15k, and FB13 in link prediction and triple classification tasks demonstrate that our method performs better than several state-of-the-art models, proving the effectiveness of the integration.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

35 extracted references · 32 canonical work pages

  1. [19]

    ADV ANCES IN NEURAL INFORMA- TION PROCESSING SYSTEMS 26

    Socher R, Chen D, Manning CD, et al (2013) Reasoning with neural tensor networks for knowledge base completion. ADV ANCES IN NEURAL INFORMA- TION PROCESSING SYSTEMS 26

    work page 2013

  2. [1]

    PROCEEDINGS OF THE IEEE 104(1):11–33

    Nickel M, Murphy K, Tresp V, et al (2015) A review of relational machine learning for knowledge graphs. PROCEEDINGS OF THE IEEE 104(1):11–33

    work page 2015

  3. [2]

    In: PROCEED- INGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, pp 1211–1220

    Lukovnikov D, Fischer A, Lehmann J, et al (2017) Neural network-based question answering over knowledge graphs on word and character level. In: PROCEED- INGS OF THE 26TH INTERNATIONAL CONFERENCE ON WORLD WIDE WEB, pp 1211–1220

    work page 2017

  4. [3]

    In: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, pp 353–362

    Zhang F, Yuan NJ, Lian D, et al (2016) Collaborative knowledge base embedding for recommender systems. In: PROCEEDINGS OF THE 22ND ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, pp 353–362

    work page 2016

  5. [4]

    In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE

    Wang Z, Zhang J, Feng J, et al (2014) Knowledge graph embedding by translat- ing on hyperplanes. In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE

    work page 2014

  6. [5]

    ADV ANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 26

    Bordes A, Usunier N, Garcia-Duran A, et al (2013) Translating embeddings for modeling multi-relational data. ADV ANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 26

    work page 2013

  7. [6]

    Sun Z, Deng Z, Nie J, et al (2019) Rotate: Knowledge graph embedding by relational rotation in complex space. CoRR 16

    work page 2019

  8. [7]

    In: ICML, pp 809–816

    Nickel M, Tresp V, Kriegel H (2011) A three-way model for collective learning on multi-relational data. In: ICML, pp 809–816

    work page 2011

Show all 35 references
  1. [8]

    In: PROCEEDINGS OF THE INTER- NATIONAL CONFERENCE ON LEARNING REPRESENTATIONS (ICLR) 2015

    Yang B, Yih SWt, He X, et al (2015) Embedding entities and relations for learning and inference in knowledge bases. In: PROCEEDINGS OF THE INTER- NATIONAL CONFERENCE ON LEARNING REPRESENTATIONS (ICLR) 2015

    2015

  2. [9]

    In: INTERNATIONAL CONFERENCE ON MACHINE LEARNING, pp 2071–2080

    Trouillon T, Welbl J, Riedel S, et al (2016) Complex embeddings for sim- ple link prediction. In: INTERNATIONAL CONFERENCE ON MACHINE LEARNING, pp 2071–2080

    2016

  3. [10]

    In: PROCEEDINGS OF CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING, pp 5185– 5194

    Balaˇ zevi´ c I, Allen C, Hospedales T (2019) Tucker: Tensor factorization for knowledge graph completion. In: PROCEEDINGS OF CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING, pp 5185– 5194

    2019

  4. [11]

    IEEE ACCESS 8:192435–192456

    Chen Z, Wang Y, Zhao B, et al (2020) Knowledge graph completion: A review. IEEE ACCESS 8:192435–192456

    2020

  5. [12]

    In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE

    Dettmers T, Minervini P, Stenetorp P, et al (2018) Convolutional 2d knowledge graph embeddings. In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE

    2018

  6. [13]

    Wang Q, Huang P, Wang H, et al (2019) Coke: Contextualized knowledge graph embedding. CoRR

    2019

  7. [14]

    In: ADV ANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, pp 5998–6008

    Vaswani A (2017) Attention is all you need. In: ADV ANCES IN NEURAL INFORMATION PROCESSING SYSTEMS, pp 5998–6008

    2017

  8. [15]

    In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE

    Lin Y, Liu Z, Sun M, et al (2015) Learning entity and relation embed- dings for knowledge graph completion. In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE

    2015

  9. [16]

    Ji G, He S, Xu L, et al (2015) Knowledge graph embedding via dynamic mapping matrix. In: PROCEEDINGS OF THE 53RD ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS AND THE 7TH INTERNATIONAL JOINT CONFERENCE ON NATURAL LANGUAGE PROCESSING (VOLUME 1: LONG PAPERS), p...

    2015

  10. [17]

    In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, pp 3065–3072

    Zhang Z, Cai J, Zhang Y, et al (2020) Learning hierarchy-aware knowledge graph embeddings for link prediction. In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE, pp 3065–3072

    2020

  11. [18]

    In: INTERNATIONAL CONFERENCE ON MACHINE LEARNING, PMLR, pp 13209–13224 17

    Li R, Zhao J, Li C, et al (2022) House: Knowledge graph embedding with householder parameterization. In: INTERNATIONAL CONFERENCE ON MACHINE LEARNING, PMLR, pp 13209–13224 17

    2022

  12. [20]

    CoRR abs/1712.02121

    Dai Quoc Nguyen TDN, Nguyen DQ, Phung D (2017) A novel embedding model for knowledge base completion based on convolutional neural network. CoRR abs/1712.02121

    2017 arXiv

  13. [21]

    In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLI- GENCE, pp 3009–3016

    Vashishth S, Sanyal S, Nitin V, et al (2020) Interacte: Improving convolution- based knowledge graph embeddings by increasing feature interactions. In: PROCEEDINGS OF THE AAAI CONFERENCE ON ARTIFICIAL INTELLI- GENCE, pp 3009–3016

    2020

  14. [22]

    In: INTERNATIONAL CONFERENCE ON EXTENDED SEMANTIC WEB CONFERENCE, pp 593–607

    Schlichtkrull M, Kipf TN, Bloem P, et al (2018) Modeling relational data with graph convolutional networks. In: INTERNATIONAL CONFERENCE ON EXTENDED SEMANTIC WEB CONFERENCE, pp 593–607

    2018

  15. [23]

    ARXIV PREPRINT ARXIV:191103082

    Vashishth S, Sanyal S, Nitin V, et al (2019) Composition-based multi-relational graph convolutional networks. ARXIV PREPRINT ARXIV:191103082

    2019

  16. [24]

    ARXIV PREPRINT ARXIV:190903193

    Yao L, Mao C, Luo Y (2019) Kg-bert: Bert for knowledge graph completion. ARXIV PREPRINT ARXIV:190903193

    2019

  17. [25]

    In: PROCEEDINGS OF THE 2021 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING, pp 10395–10407

    Chen S, Liu X, Gao J, et al (2021) Hitter: Hierarchical transformers for knowl- edge graph embeddings. In: PROCEEDINGS OF THE 2021 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING, pp 10395–10407

    2021

  18. [26]

    ADV ANCES IN NEURAL INFORMATION PRO- CESSING SYSTEMS 30

    Yang F, Yang Z, Cohen WW (2017) Differentiable learning of logical rules for knowledge base reasoning. ADV ANCES IN NEURAL INFORMATION PRO- CESSING SYSTEMS 30

    2017

  19. [27]

    ADV ANCES IN NEURAL INFORMA- TION PROCESSING SYSTEMS 32

    Sadeghian A, Armandpour M, Ding P, et al (2019) Drum: End-to-end differen- tiable rule mining on knowledge graphs. ADV ANCES IN NEURAL INFORMA- TION PROCESSING SYSTEMS 32

    2019

  20. [28]

    IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING

    Pan S, Luo L, Wang Y, et al (2024) Unifying large language models and knowledge graphs: A roadmap. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING

    2024

  21. [29]

    CoRR abs/2309.01538

    Luo L, Ju J, Xiong B, et al (2023) Chatrule: Mining logical rules with large language models for knowledge graph reasoning. CoRR abs/2309.01538

    2023 arXiv

  22. [30]

    ARXIV PREPRINT ARXIV:230513048 18

    Peng B, Alcaide E, Anthony Q, et al (2023) Rwkv: Reinventing rnns for the transformer era. ARXIV PREPRINT ARXIV:230513048 18

    2023

  23. [31]

    CoRR abs/2105.14103

    Zhai S, Talbott W, Srivastava N, et al (2021) An attention free transformer. CoRR abs/2105.14103

    2021 arXiv

  24. [32]

    COM- PARATIVE AND FUNCTIONAL GENOMICS 4(1):80–84

    McCray AT (2003) An upper-level ontology for the biomedical domain. COM- PARATIVE AND FUNCTIONAL GENOMICS 4(1):80–84

    2003

  25. [33]

    In: BIENNIAL CONFERENCE ON INNOV ATIVE DATA SYSTEMS RESEARCH, CIDR

    Mahdisoltani F, Biega J, Suchanek FM (2015) Yago3: A knowledge base from mul- tilingual wikipedias. In: BIENNIAL CONFERENCE ON INNOV ATIVE DATA SYSTEMS RESEARCH, CIDR

    2015

  26. [34]

    In: PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, pp 179–189

    Cheng K, Liu J, Wang W, et al (2022) Rlogic: Recursive logical rule learning from knowledge graphs. In: PROCEEDINGS OF THE 28TH ACM SIGKDD CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING, pp 179–189

    2022

  27. [35]

    JOURNAL OF MACHINE LEARNING RESEARCH 9(11) 19

    Van der Maaten L, Hinton G (2008) Visualizing data using t-sne. JOURNAL OF MACHINE LEARNING RESEARCH 9(11) 19

    2008

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.