Pith. sign in

REVIEW 4 major objections 6 minor 1 cited by

Transformers for Complex Query Answering over Knowledge Hypergraphs

T0 review · 4 major / 6 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read A two-stage transformer called LKHGT answers complex logical queries over knowledge hypergraphs, outperforming prior models on average MRR for EPFO and negation queries on two new benchmark datasets.

desk verdict Useful new datasets and a reasonable model, but the OOD generalization claim in the abstract is contradicted by the paper's own Table 1. read the letter →

arxiv 2504.16537 v1 pith:XGBF5PJN submitted 2025-04-23 cs.CL cs.AI

Hong Ting Tsang , Zihao Wang , Yangqiu Song This is my paper · ORCID
classification cs.CLcs.AI
keywords complexqueryansweringknowledgehypergraphsn-aryrelationslogicalreasoningtransformertypeawarebiasEFO-1queriesMRR
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 graphs usually store facts as three-part statements, but many real facts involve any number of entities acting together; the paper extends complex query answering to knowledge hypergraphs, where a fact is a single hyperedge that treats all entities as equal contributors. It introduces two datasets, JF17k-HCQA and M-FB15k-HCQA, with fourteen query types built from projection, negation, conjunction, and disjunction. The proposed model, the Logical Knowledge Hypergraph Transformer (LKHGT), processes a query as an operator tree in two stages: a Projection Encoder handles each atomic hyperedge, and a Logical Encoder combines the projected variables. The paper reports that LKHGT achieves the best average MRR among compared models on both datasets, and that replacing fuzzy logic with the learned logical encoder raises performance once all query-type input combinations have been seen in training.

What carries the argument

The operator-tree representation plus a two-stage transformer. LKHGT walks the query tree node by node: the Projection Encoder turns each ordered hyperedge (relation tokens, entity tokens, a marked variable token, optional negation) into an answer-set embedding, and the Logical Encoder takes the projected variable embeddings and the logical operator token (intersection or union) to produce the final embedding. Type Aware Bias (TAB) is the inductive-bias component: a learned bias over pairs of the eight token types, added inside softmax attention, that differentiates interactions such as relation-to-variable versus entity-to-negation and lets the transformer aggregate over the hyperedge as a fully connected graph.

What would settle it

On the JF17k-HCQA split that withholds the 3I and 3IN query types from training, LKHGT scores 42.71 and 18.26 MRR while NQE scores 70.40 and 27.80; this is the decisive check for the out-of-distribution generalization claim. A reader could also hold out a different combination of query arities and compare the two models on those types.

Watch

Extended reading notes

Core claim

The central claim is that a transformer can answer existential first-order queries over knowledge hypergraphs by iteratively building the query tree, rather than encoding the whole query at once. Each atomic formula is an ordered hyperedge whose positions carry semantic meaning; the model marks one position as the variable to project. Type Aware Bias (TAB) inserts a learnable bias into the self-attention logits for every ordered pair of token types, which lets the same transformer handle projection, negation, conjunction, and disjunction. The paper argues this architecture is state of the art on the introduced datasets and that the learned logical encoder can replace fuzzy logic for conjunction and disjunction. It also reports a boundary condition: on query types not seen in combination during training, such as three-way conjunction and three-way negated conjunction, LKHGT lags behind NQE until the full training set is provided.

Load-bearing premise

The load-bearing premise is that the logical encoder can handle variable numbers of projected embeddings and generalize to input combinations, such as three-way conjunctions and three-way negated conjunctions, that were not seen during training.

Editorial extensions

If this is right

  • Complex query answering can move from binary relation triples to n-ary facts where every entity plays an equal role, without flattening the fact through star expansion.
  • A transformer-based logical encoder can replace fuzzy-logic operators for conjunction and disjunction, provided the training data covers the input combinations that will appear at test time.
  • Out-of-distribution generalization in this architecture is not automatic: unseen combinations of projected and logical embeddings degrade three-way conjunction and negation queries, and full training restores most of the gap.
  • Processing each atomic projection separately, then combining embeddings, beats encoding the entire query graph in a single transformer pass for complex multi-hop queries.
  • Learning to rank answers for these query types gives a way to reuse standard MRR evaluation on n-ary and negated queries.

Reading between the lines

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

  • If the reported 3I and 3IN gap is a general property of learned logical encoders, robust n-ary CQA will need either combinatorial coverage of query arities or an explicit symbolic fallback for unseen conjunction and negation shapes; the paper's own ablation points this way.
  • The position semantics of ordered hyperedges are likely to matter more as arity grows; one testable extension is to let TAB depend on absolute position within the hyperedge rather than only on token type.
  • The two new datasets could serve as a stress test for whether binary-query methods degrade gracefully as arity increases, since hyperedges can be restricted to arity two to recover ordinary graph queries.
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 introduces Knowledge Hypergraph CQA datasets JF17k-HCQA and M-FB15k-HCQA, and proposes LKHGT, a two-stage transformer model with a Projection Encoder and a Logical Encoder, both augmented with Type Aware Bias (TAB), to answer EFO-1 queries over knowledge hypergraphs. The authors evaluate LKHGT against NQE, LSGT, and a custom HLMPNN baseline on 14 query types, and report that LKHGT achieves the best average MRR for EPFO and negation queries on both datasets. They also argue that LKHGT generalizes to out-of-distribution query types, and support this with an experimental setup that trains on all query types except [3p, 3in, 3i, inp]. The paper includes ablations on positional encoding, fuzzy logic replacement, and logical-encoder input cardinality, as well as a complexity analysis.

Significance. If the results held as stated, the paper would make a useful contribution: it opens a new problem setting (CQA over knowledge hypergraphs), contributes two sampled datasets, and proposes a two-stage transformer architecture with an inductive bias tailored to token-type interactions. The paper also includes a reasonable set of ablations and a complexity analysis. The central positive result, that LKHGT obtains competitive average MRR against adapted baselines, is plausible. However, the more ambitious out-of-distribution generalization claim is contradicted by the paper's own numbers, and the baseline comparison is weakened by the encoder-replacement protocol. With revision of the claims and additional evaluation, the underlying method could still be of interest to the CQA community.

major comments (4)
  1. [Abstract and Section 5.3/5.4, Table 1 and Table 2]
  2. [Section 5.2]
  3. [Tables 1-3]
  4. [Reproducibility (Section 5)]
minor comments (6)
  1. [Section 4.1, Eq. (3)]
  2. [Section 4.1, Eq. (4)]
  3. [Section 5.5, Table 2]
  4. [Section 7, Limitation]
  5. [Appendix A.2]
  6. [Throughout]

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: LKHGT's stated capabilities are backed by held-out evaluations rather than by definitions, fitted inputs, or load-bearing self-citations.

full rationale

The paper's central contribution is a two-stage transformer, LKHGT, evaluated on newly sampled knowledge-hypergraph CQA datasets, and its claims are empirical rather than definitional. The held-out query types are genuinely absent from the training distribution, and the full-training experiments are transparently labeled as such; the poor 3I/3IN results and Section 7's limitation about needing all token-interaction combinations undercut the out-of-distribution claim, but they are empirical consistency issues, not circular derivations. The self-citations to LMPNN, TEGA, and the prior sampling benchmark are used for architectural inspiration and experimental protocol, not as unverified premises that force the conclusion. No equation identifies a predicted quantity with a fitted input by construction, no uniqueness theorem is imported from the authors' prior work, and no known result is merely renamed as a new contribution. The paper therefore has no significant circularity, and any concerns about the OOD claim should be treated as correctness or evaluation-design risks, which are outside the circularity criteria.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on standard transformer machinery, the domain assumption that ordered hyperedges with fixed-arity relations and positional semantics are the right representation for n-ary facts, and the ad hoc architectural assumption that a single logical encoder can process variable cardinalities. No free parameters beyond standard hyperparameters are fit to a derivation; the model is trained end-to-end.

free parameters (3)
  • Model embedding dimension = 400 for LKHGT and NQE, 200 for HLMPNN, 1024 for LSGT
    These dimensions are chosen by the authors to make models converge; they are not derived from theory and affect capacity.
  • Training epochs = 400
    All models are trained for 400 epochs on a single 3090 GPU; no early stopping or tuning is reported.
  • Query sampling counts = Train 60,000 for 1P, 20,000 for others; validation and test 10,000 each
    Dataset sizes are chosen arbitrarily; no analysis of how this choice affects results is provided.
assumptions (4)
  • standard math Transformer self-attention operates as a fully connected graph over input tokens.
    The model treats each hyperedge as a complete graph and applies attention; this is an architectural assumption that token interactions can be captured by attention alone.
  • domain assumption Each relation type has a fixed arity and each position has a fixed semantic meaning.
    Section 3.1; if positions are not semantically aligned, the ordered hyperedge encoding breaks down.
  • domain assumption The query sampling algorithm in Appendix B produces queries representative of real EFO-1 queries over hypergraphs.
    No validation is provided that the sampled query distribution matches real usage patterns.
  • ad hoc to paper The logical encoder can process variable numbers of projected embeddings in a single pass.
    This is the load-bearing assumption for handling n-ary intersections and unions, and the paper's results show it generalizes poorly to unseen cardinalities.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Transformers for Complex Query Answering over Knowledge Hypergraphs." pith.science (2026). https://pith.science/paper/XGBF5PJN

@misc{pith2026250416537,
  author       = {Pith},
  title        = {Pith review of: Transformers for Complex Query Answering over Knowledge Hypergraphs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XGBF5PJN}},
  note         = {Machine review of arXiv:2504.16537}
}
read the original abstract

Complex Query Answering (CQA) has been extensively studied in recent years. In order to model data that is closer to real-world distribution, knowledge graphs with different modalities have been introduced. Triple KGs, as the classic KGs composed of entities and relations of arity 2, have limited representation of real-world facts. Real-world data is more sophisticated. While hyper-relational graphs have been introduced, there are limitations in representing relationships of varying arity that contain entities with equal contributions. To address this gap, we sampled new CQA datasets: JF17k-HCQA and M-FB15k-HCQA. Each dataset contains various query types that include logical operations such as projection, negation, conjunction, and disjunction. In order to answer knowledge hypergraph (KHG) existential first-order queries, we propose a two-stage transformer model, the Logical Knowledge Hypergraph Transformer (LKHGT), which consists of a Projection Encoder for atomic projection and a Logical Encoder for complex logical operations. Both encoders are equipped with Type Aware Bias (TAB) for capturing token interactions. Experimental results on CQA datasets show that LKHGT is a state-of-the-art CQA method over KHG and is able to generalize to out-of-distribution query types.

Figures

Figures reproduced from arXiv: 2504.16537 by the authors.

Figure 1
Figure 1. Example of LKHGT processing query tree of ip type [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Example of LKHGT processing query tree atomic formula. 4.1 Tokens Input and Output Operators type included in operator tree are pro￾jection (p), negation (¬), intersection (∧), union (∨). As defined, each complex query is composed of atomic formula which is made of relaiton (r), subject (s), variable (e). Overally speaking, there are 8 types of token. Relation, existential variable, free variable, entity and negatio… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Enhancing Transformers for Generalizable First-Order Logical Entailment

    cs.CL 2025-01 conditional novelty 6.0 of 10

    Transformers with relative positional encoding beat KGQA baselines, and adding logic-aware attention (TEGA) improves out-of-distribution performance on a new 55-type benchmark.

Reference graph

Works this paper leans on

49 extracted references · 12 canonical work pages · cited by 1 Pith paper

  1. [1]

    Ralph Abboud, İsmail İlkan Ceylan, Thomas Lukasiewicz, and Tommaso Salvatori. 2020. https://doi.org/10.48550/arXiv.2007.06267 BoxE : A Box Embedding Model for Knowledge Base Completion . arXiv preprint. ArXiv:2007.06267 [cs]

    arXiv doi:10.48550/arxiv.2007.06267 2020

  2. [2]

    Sameer Agarwal, Kristin Branson, and Serge Belongie. 2006. https://doi.org/10.1145/1143844.1143847 Higher order learning with graphs . In Proceedings of the 23rd international conference on Machine learning - ICML '06 , pages 17--24, Pittsburgh, Pennsylvania. ACM Press

    arXiv 2006

  3. [3]

    Dimitrios Alivanistos, Max Berrendorf, Michael Cochez, and Mikhail Galkin. 2022. https://doi.org/10.48550/arXiv.2106.08166 Query Embedding on Hyper -relational Knowledge Graphs . arXiv preprint. ArXiv:2106.08166 [cs]

    arXiv doi:10.48550/arxiv.2106.08166 2022

  4. [4]

    Alfonso Amayuelas, Shuai Zhang, Susie Xi Rao, and Ce Zhang. 2022. https://doi.org/10.48550/arXiv.2209.14464 Neural Methods for Logical Reasoning Over Knowledge Graphs . arXiv preprint. ArXiv:2209.14464 [cs]

    work page Pith review arXiv doi:10.48550/arxiv.2209.14464 2022

  5. [5]

    Erik Arakelyan, Daniel Daza, Pasquale Minervini, and Michael Cochez. 2021. https://doi.org/10.48550/arXiv.2011.03459 Complex Query Answering with Neural Link Predictors . arXiv preprint. ArXiv:2011.03459 [cs]

    arXiv doi:10.48550/arxiv.2011.03459 2021

  6. [6]

    Jiaxin Bai, Chen Luo, Zheng Li, Qingyu Yin, and Yangqiu Song. 2024. https://doi.org/10.48550/arXiv.2312.13866 Understanding Inter - Session Intentions via Complex Logical Reasoning . arXiv preprint. ArXiv:2312.13866 [cs]

    work page Pith review arXiv doi:10.48550/arxiv.2312.13866 2024

  7. [7]

    Jiaxin Bai, Tianshi Zheng, and Yangqiu Song. 2023 a . https://doi.org/10.48550/arXiv.2302.13114 Sequential Query Encoding For Complex Query Answering on Knowledge Graphs . arXiv preprint. ArXiv:2302.13114 [cs]

    arXiv doi:10.48550/arxiv.2302.13114 2023

  8. [8]

    Yushi Bai, Xin Lv, Juanzi Li, and Lei Hou. 2023 b . https://doi.org/10.48550/arXiv.2212.09567 Answering Complex Logical Queries on Knowledge Graphs via Query Computation Tree Optimization . arXiv preprint. ArXiv:2212.09567 [cs]

    arXiv doi:10.48550/arxiv.2212.09567 2023

Show all 49 references
  1. [9]

    Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Oksana Yakhnenko. 2013. https://papers.nips.cc/paper_files/paper/2013/hash/1cecc7a77928ca8133fa24680a88d2f9-Abstract.html Translating Embeddings for Modeling Multi -relational Data . In Advances in Neural...

    2013

  2. [10]

    Xuelu Chen, Ziniu Hu, and Yizhou Sun. 2022. https://doi.org/10.48550/arXiv.2108.02390 Fuzzy Logic Based Logical Query Answering on Knowledge Graphs . arXiv preprint. ArXiv:2108.02390 [cs]

    2022 doi arXiv

  3. [11]

    Tim Dettmers, Pasquale Minervini, Pontus Stenetorp, and Sebastian Riedel. 2018. https://doi.org/10.48550/arXiv.1707.01476 Convolutional 2D Knowledge Graph Embeddings . arXiv preprint. ArXiv:1707.01476 [cs]

    2018 doi arXiv

  4. [12]

    Bahare Fatemi, Perouz Taslakian, David Vazquez, and David Poole. 2020. https://doi.org/10.24963/ijcai.2020/303 Knowledge Hypergraphs : Prediction Beyond Binary Relations . In Proceedings of the Twenty - Ninth International Joint Conference on Artificial Intelligence , pages 21...

    2020 doi

  5. [13]

    Bahare Fatemi, Perouz Taslakian, David Vazquez, and David Poole. 2021. https://doi.org/10.48550/arXiv.2102.09557 Knowledge Hypergraph Embedding Meets Relational Algebra . arXiv preprint. ArXiv:2102.09557 [cs]

    2021 doi arXiv

  6. [14]

    Mikhail Galkin, Priyansh Trivedi, Gaurav Maheshwari, Ricardo Usbeck, and Jens Lehmann. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.596 Message Passing for Hyper - Relational Knowledge Graphs . In Proceedings of the 2020 Conference on Empirical Methods in Natural Language...

    2020 doi

  7. [15]

    Saiping Guan, Xiaolong Jin, Jiafeng Guo, Yuanzhuo Wang, and Xueqi Cheng. 2020. https://doi.org/10.18653/v1/2020.acl-main.546 NeuInfer : Knowledge Inference on N -ary Facts . In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics , pages 6141...

    2020 doi

  8. [16]

    Hamilton, Payal Bajaj, Marinka Zitnik, Dan Jurafsky, and Jure Leskovec

    William L. Hamilton, Payal Bajaj, Marinka Zitnik, Dan Jurafsky, and Jure Leskovec. 2019. https://doi.org/10.48550/arXiv.1806.01445 Embedding Logical Queries on Knowledge Graphs . arXiv preprint. ArXiv:1806.01445 [cs]

    2019 doi arXiv

  9. [17]

    Bronstein, and İsmail İlkan Ceylan

    Xingyue Huang, Miguel Romero Orth, Pablo Barceló, Michael M. Bronstein, and İsmail İlkan Ceylan. 2024. https://doi.org/10.48550/arXiv.2402.04062 Link Prediction with Relational Hypergraphs . arXiv preprint. ArXiv:2402.04062 [cs]

    2024 doi arXiv

  10. [18]

    Shaoxiong Ji, Shirui Pan, Erik Cambria, Pekka Marttinen, and Philip S. Yu. 2022. https://doi.org/10.1109/TNNLS.2021.3070843 A Survey on Knowledge Graphs : Representation , Acquisition , and Applications . IEEE Transactions on Neural Networks and Learning Systems, 33(2):494--51...

    2022

  11. [19]

    Junnan Liu, Qianren Mao, Weifeng Jiang, and Jianxin Li. 2024 a . https://doi.org/10.48550/arXiv.2409.12865 KnowFormer : Revisiting Transformers for Knowledge Graph Reasoning . arXiv preprint. ArXiv:2409.12865 [cs]

    2024 doi arXiv

  12. [20]

    Lihui Liu, Boxin Du, Heng Ji, ChengXiang Zhai, and Hanghang Tong. 2021. https://doi.org/10.1145/3447548.3467375 Neural- Answering Logical Queries on Knowledge Graphs . In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining , pages 1087--1097, Vir...

    2021

  13. [21]

    Yu Liu, Shu Yang, Jingtao Ding, Quanming Yao, and Yong Li. 2024 b . https://doi.org/10.48550/arXiv.2411.06191 Generalizing Hyperedge Expansion for Hyper -relational Knowledge Graph Modeling . arXiv preprint. ArXiv:2411.06191 [cs]

    2024 doi arXiv

  14. [22]

    Yu Liu, Quanming Yao, and Yong Li. 2020. https://doi.org/10.48550/arXiv.2007.03988 Generalizing Tensor Decomposition for N -ary Relational Knowledge Bases . arXiv preprint. ArXiv:2007.03988 [cs]

    2020 doi arXiv

  15. [23]

    Haoran Luo, Haihong E, Yuhao Yang, Gengxian Zhou, Yikai Guo, Tianyu Yao, Zichen Tang, Xueyuan Lin, and Kaiyang Wan. 2023. https://doi.org/10.1609/aaai.v37i4.25576 NQE : N -ary Query Embedding for Complex Query Answering over Hyper - Relational Knowledge Graphs . Proceedings of...

    2023 arXiv

  16. [24]

    Maximilian Nickel, Volker Tresp, and Hans-Peter Kriegel. 2011. A three-way model for collective learning on multi-relational data. In Proceedings of the 28th International Conference on International Conference on Machine Learning , ICML '11, pages 809--816, Madison, WI, USA. ...

    2011

  17. [25]

    Hongyu Ren, Mikhail Galkin, Michael Cochez, Zhaocheng Zhu, and Jure Leskovec. 2023. https://doi.org/10.48550/arXiv.2303.14617 Neural Graph Reasoning : Complex Logical Query Answering Meets Graph Databases . arXiv preprint. ArXiv:2303.14617 [cs]

    2023 doi arXiv

  18. [26]

    Hongyu Ren, Weihua Hu, and Jure Leskovec. 2020. https://doi.org/10.48550/arXiv.2002.05969 Query2box: Reasoning over Knowledge Graphs in Vector Space using Box Embeddings . arXiv preprint. ArXiv:2002.05969 [cs]

    2020 doi arXiv

  19. [27]

    Hongyu Ren and Jure Leskovec. 2020. https://papers.nips.cc/paper_files/paper/2020/hash/e43739bba7cdb577e9e3e4e42447f5a5-Abstract.html Beta Embeddings for Multi - Hop Logical Reasoning in Knowledge Graphs . In Advances in Neural Information Processing Systems , volume 33, pages...

    2020

  20. [28]

    Peter Shaw, Jakob Uszkoreit, and Ashish Vaswani. 2018. https://doi.org/10.48550/arXiv.1803.02155 Self- Attention with Relative Position Representations . arXiv preprint. ArXiv:1803.02155 [cs]

    2018 doi arXiv

  21. [29]

    Jianlin Su, Yu Lu, Shengfeng Pan, Ahmed Murtadha, Bo Wen, and Yunfeng Liu. 2023. https://doi.org/10.48550/arXiv.2104.09864 RoFormer : Enhanced Transformer with Rotary Position Embedding . arXiv preprint. ArXiv:2104.09864 [cs]

    2023 doi arXiv

  22. [30]

    Zhiqing Sun, Zhi-Hong Deng, Jian-Yun Nie, and Jian Tang. 2019. https://doi.org/10.48550/arXiv.1902.10197 RotatE : Knowledge Graph Embedding by Relational Rotation in Complex Space . arXiv preprint. ArXiv:1902.10197 [cs]

    2019 doi arXiv

  23. [31]

    Kai Sheng Tai, Richard Socher, and Christopher D. Manning. 2015. https://doi.org/10.48550/arXiv.1503.00075 Improved Semantic Representations From Tree - Structured Long Short - Term Memory Networks . arXiv preprint. ArXiv:1503.00075 [cs]

    2015 doi arXiv

  24. [32]

    Théo Trouillon, Johannes Welbl, Sebastian Riedel, Éric Gaussier, and Guillaume Bouchard. 2016. https://doi.org/10.48550/arXiv.1606.06357 Complex Embeddings for Simple Link Prediction . arXiv preprint. ArXiv:1606.06357 [cs]

    2016 doi arXiv

  25. [33]

    Gomez, Lukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2023. https://doi.org/10.48550/arXiv.1706.03762 Attention Is All You Need . arXiv preprint. ArXiv:1706.03762 [cs]

    2023 doi arXiv

  26. [34]

    Quan Wang, Haifeng Wang, Yajuan Lyu, and Yong Zhu. 2021 a . https://doi.org/10.18653/v1/2021.findings-acl.35 Link Prediction on N -ary Relational Facts : A Graph -based Approach . In Findings of the Association for Computational Linguistics : ACL - IJCNLP 2021 , pages 396--407...

    2021 doi

  27. [35]

    Yu, and Yanfang Ye

    Xiao Wang, Houye Ji, Chuan Shi, Bai Wang, Peng Cui, P. Yu, and Yanfang Ye. 2021 b . https://doi.org/10.48550/arXiv.1903.07293 Heterogeneous Graph Attention Network . arXiv preprint. ArXiv:1903.07293 [cs]

    2021 doi arXiv

  28. [36]

    Wong, and Simon See

    Zihao Wang, Yangqiu Song, Ginny Y. Wong, and Simon See. 2023. https://doi.org/10.48550/arXiv.2301.08859 Logical Message Passing Networks with One -hop Inference on Atomic Formulas . arXiv preprint. ArXiv:2301.08859 [cs]

    2023 doi arXiv

  29. [37]

    Zihao Wang, Hang Yin, and Yangqiu Song. 2021 c . https://doi.org/10.48550/arXiv.2109.08925 Benchmarking the Combinatorial Generalizability of Complex Query Answering on Knowledge Graphs . arXiv preprint. ArXiv:2109.08925 [cs]

    2021 doi arXiv

  30. [38]

    Jianfeng Wen, Jianxin Li, Yongyi Mao, Shini Chen, and Richong Zhang. 2016. https://doi.org/10.48550/arXiv.1604.08642 On the representation and embedding of knowledge bases beyond binary relations . arXiv preprint. ArXiv:1604.08642 [cs]

    2016 doi arXiv

  31. [39]

    Yao Xu, Shizhu He, Cunguang Wang, Li Cai, Kang Liu, and Jun Zhao. 2023. https://doi.org/10.48550/arXiv.2310.11246 Query2Triple : Unified Query Encoding for Answering Diverse Complex Queries over Knowledge Graphs . arXiv preprint. ArXiv:2310.11246 [cs]

    2023 doi

  32. [40]

    Zezhong Xu, Wen Zhang, Peng Ye, Hui Chen, and Huajun Chen. 2022. https://doi.org/10.48550/arXiv.2209.08779 Neural- Symbolic Entangled Framework for Complex Query Answering . arXiv preprint. ArXiv:2209.08779 [cs]

    2022 doi arXiv

  33. [41]

    Naganand Yadati. 2020. https://proceedings.neurips.cc/paper/2020/hash/217eedd1ba8c592db97d0dbe54c7adfc-Abstract.html Neural Message Passing for Multi - Relational Ordered and Recursive Hypergraphs . In Advances in Neural Information Processing Systems , volume 33, pages 3275--...

    2020

  34. [42]

    Naganand Yadati, Madhav Nimishakavi, Prateek Yadav, Vikram Nitin, Anand Louis, and Partha Talukdar. 2019. https://doi.org/10.48550/arXiv.1809.02589 HyperGCN : A New Method of Training Graph Convolutional Networks on Hypergraphs . arXiv preprint. ArXiv:1809.02589 [cs]

    2019 doi arXiv

  35. [43]

    L.A. Zadeh. 1988. https://doi.org/10.1109/2.53 Fuzzy logic . Computer, 21(4):83--93. Conference Name: Computer

    1988 doi

  36. [44]

    Chongzhi Zhang, Zhiping Peng, Junhao Zheng, Linghao Wang, Ruifeng Shi, and Qianli Ma. 2024. https://doi.org/10.48550/arXiv.2406.14880 Pathformer: Recursive Path Query Encoding for Complex Logical Query Answering . arXiv preprint. ArXiv:2406.14880 [cs]

    2024 doi arXiv

  37. [45]

    Richong Zhang, Junpeng Li, Jiajie Mei, and Yongyi Mao. 2018. https://doi.org/10.1145/3178876.3186017 Scalable Instance Reconstruction in Knowledge Bases via Relatedness Affiliated Embedding . In Proceedings of the 2018 World Wide Web Conference on World Wide Web - WWW '18 , pa...

    2018

  38. [46]

    Zhanqiu Zhang, Jie Wang, Jiajun Chen, Shuiwang Ji, and Feng Wu. 2021. https://doi.org/10.48550/arXiv.2110.13715 ConE : Cone Embeddings for Multi - Hop Reasoning over Knowledge Graphs . arXiv preprint. ArXiv:2110.13715 [cs]

    2021 doi arXiv

  39. [47]

    Tianshi Zheng, Jiazheng Wang, Zihao Wang, Jiaxin Bai, Hang Yin, Zheye Deng, Yangqiu Song, and Jianxin Li. 2025. https://doi.org/10.48550/arXiv.2501.00759 Enhancing Transformers for Generalizable First - Order Logical Entailment . arXiv preprint. ArXiv:2501.00759 [cs]

    2025 doi arXiv

  40. [48]

    online" 'onlinestring :=

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint eprinttype howpublished institution journal key month note number organization pages publisher school series title type volume year doi pubmed url lastchecked label extra.label sort.label short.list...

  41. [49]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.