Pith. sign in

REVIEW 4 major objections 5 minor 37 references

A Method for Multi-Hop Question Answering on Persian Knowledge Graph

T0 review · 4 major / 5 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 a decompose-then-execute pipeline for Persian multi-hop questions raises F1 to 75.55% and accuracy to 74.81% on the PeCoQ benchmark, beating the only prior Persian KGQA system by 12.57 and 12.06 points.

desk verdict Useful first Persian question-decomposition dataset, but the 12.57 F1 headline needs a clean test-split disclosure before I'd trust it. read the letter →

arxiv 2501.16350 v1 pith:5GK7F7CL submitted 2025-01-18 cs.IR cs.AIcs.CL

Arash Ghafouri , Mahdi Firouzmandi , Hasan Naderi This is my paper · ORCID
classification cs.IRcs.AIcs.CL
keywords knowledgegraphquestionansweringmulti-hopPersiandecompositionSPARQLgenerationFarsBasePeCoQdatasetnamedentitylinking
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

This paper tries to establish that Persian multi-hop questions over a knowledge graph can be answered by decomposition: split the question into a sequence of small semantic units, link the entities in each unit to the FarsBase knowledge graph, generate one SPARQL query per unit, and execute the queries in order. To do this the authors built MRDCPQ, a 5,600-question Persian dataset of complex questions paired with their step-by-step semantic decompositions, and fine-tuned multilingual and Persian transformer models for each stage. On the PeCoQ test set the complete system scores 75.55% F1 and 74.81% accuracy, beating the only prior Persian KGQA system by 12.57 and 12.06 points respectively. A reader should care because this supplies the first public Persian dataset for question decomposition and a complete recipe for a language that previously had a single comparable KGQA system.

What carries the argument

The load-bearing object is the Meaning Representation of Decomposed Complex Persian Question (MRDCPQ), a numbered list of short semantic units in which a complex question is rewritten as a sequence of entity and relation steps with back-references such as '#1' and '#2'. The paper's architecture has four components: a multilingual text-to-text decomposer fine-tuned on the 5,600-question MRDCPQ dataset, a Persian transformer named-entity recognizer and zero-shot entity linker that maps entities to FarsBase, a multilingual text-to-text unit-to-SPARQL generator, and an executor that chains the generated queries so the object of one hop becomes the subject of the next. The MRDCPQ representation is what makes sequential execution possible: it turns multi-hop reasoning into a fixed executable order rather than a single monolithic query.

What would settle it

Replace the predicted decompositions on the PeCoQ test set with the gold decompositions from MRDCPQ and rerun the rest of the pipeline unchanged. If the F1 score stays near 75.55%, decomposition errors are not the binding constraint; if it rises substantially, the claimed end-to-end gain depends on the 77.61%-accurate decomposer and would shrink for questions the decomposer gets wrong.

Watch

Extended reading notes

Core claim

Stated on the paper's own terms, the discovery is that the bottleneck in Persian KGQA is the transformation of multi-hop questions into a semantically equivalent SPARQL query, and that this transformation becomes tractable when the question is first rewritten as a chain of meaning units. Each unit is either an entity or a relation, and later units refer back to earlier ones with numbered markers, so the chain is executable: the answer of one unit feeds the next. Trained on the MRDCPQ dataset, the decomposition model reaches 77.61% task-decomposition accuracy, the entity recognizer reaches 99.16% accuracy, and the unit-to-SPARQL generator reaches 82.35% F1; end-to-end, the system outperforms the only prior Persian KGQA approach on precision (84.36% vs 71.24%), recall (68.41% vs 56.45%), F1 (75.55% vs 62.98%), and accuracy (74.81% vs 62.75%).

Load-bearing premise

The load-bearing premise is that the model's decomposition of a complex Persian question is semantically equivalent to the original question, so executing the decomposed steps in sequence returns the same answer as one correct SPARQL query; the paper's own 77.61% decomposition accuracy shows this equivalence fails on a nontrivial fraction of questions.

Editorial extensions

If this is right

  • Any Persian KGQA system can treat question decomposition as a separate trainable component, because MRDCPQ provides the first benchmark for it.
  • Because the units are human-readable, the system's reasoning path is inspectable: a wrong answer can be traced to the exact hop that produced it.
  • The per-component results suggest the NER and entity-linking stages are near-saturated for this benchmark, so future gains on PeCoQ will likely come from decomposition and SPARQL generation.
  • The same decompose-then-execute pipeline could be applied to other low-resource languages with a multilingual transformer, provided a knowledge graph and an entity-linking resource exist.

Reading between the lines

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

  • The 77.61% decomposition accuracy is probably the effective ceiling on end-to-end accuracy: if a step drops or reorders a relation, every downstream hop inherits the error, so improving decomposition is the highest-leverage intervention even though the paper does not report this ablation.
  • Ablating the system with oracle decompositions from MRDCPQ would separate the contribution of the decomposer from the contributions of entity linking and SPARQL generation; the paper does not report this experiment.
  • The 12.57-point F1 gap is measured against a single prior system; LLM-based KGQA methods, which the paper reviews for English, have not yet been evaluated on Persian PeCoQ and would be a natural next comparison.
  • The MRDCPQ format mirrors the BREAK-style question-understanding benchmark, so the decomposition dataset could support other Persian tasks such as semantic parsing and explainable QA, not only KGQA.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper proposes a four-component pipeline for answering multi-hop complex Persian questions over the FarsBase knowledge graph: (1) decomposition of the question into BREAK-style semantic segments using a fine-tuned mT5 model, (2) named entity recognition with ParsBERT and zero-shot entity linking to FarsBase, (3) generation of one SPARQL query per segment using a fine-tuned mT5 model, and (4) sequential execution of these queries to compose the final answer. The authors introduce MRDCPQ, a dataset of 5,600 Persian multi-hop questions with semantic decompositions, derived from the PeCoQ dataset. Component-level results include task decomposition accuracy of 77.61%, NER accuracy of 99.16%, and SPARQL generation F1 of 82.35%. The main empirical claim, reported in Section 4.4, is that the full pipeline outperforms the only prior Persian KGQA system of Etezadi et al. on the PeCoQ test set by 12.57% F1 and 12.06% accuracy.

Significance. If the empirical claims are validated, this is a useful contribution to Persian KGQA: it provides the first Persian dataset for semantic question decomposition, demonstrates that a decomposition-based pipeline can exceed the previous state of the art on PeCoQ, and reports per-component results that are informative for future work. The strengths are the creation of MRDCPQ, the use of a standard BREAK-style decomposition formalism, and the explicit evaluation of decomposition, NER, and SPARQL generation as separate stages. However, the significance is currently limited by evaluation weaknesses and unresolved data-provenance questions that could affect the validity of the headline improvement.

major comments (4)
  1. [§3.1.1, §4.4, Table 4] The central claim of a 12.57% F1 improvement on the PeCoQ test set is not supported without a clear statement that the PeCoQ test examples are disjoint from the MRDCPQ training split. MRDCPQ is built 'based on the PeCoQ dataset' and split 80/10/10 for training, testing, and validation, while Table 4 reports results 'on the test dataset of PeCoQ.' The paper never specifies whether the PeCoQ test questions used in Table 4 are the same as, overlapping with, or disjoint from the MRDCPQ training set. It also does not state whether Table 4 is computed on all PeCoQ test questions or only on the 5,600 'correctly decomposed' questions. This is load-bearing for the headline result, and the inconsistency that 'approximately 15,000 multi-hop complex questions were extracted from PeCoQ' despite PeCoQ being described as containing 10,000 questions makes the curation step underspecified.
  2. [§4.4, Table 4] The end-to-end evaluation uses a single baseline, reports no variance or significance tests, and gives no details about the number of test questions or the exact evaluation protocol. With a 12.57 F1-point difference over one comparator, the result could be meaningful, but the absence of standard deviations over multiple fine-tuning seeds, confidence intervals, or a significance test makes it impossible to assess whether the gap is robust. The paper should report results over multiple runs with statistical testing, and should also report end-to-end accuracy conditioned on whether the decomposition was correct, since the decomposition TDA is only 77.61% and error propagation is a stated risk of the sequential execution architecture.
  3. [§3.1, §3.4, §4.1] The method assumes that the semantic decomposition produced in Section 3.1 is exactly semantically equivalent to the original complex question, so that sequential execution of the per-step SPARQL queries yields the same answer as a single correct query. This assumption is never verified. The TDA metric is described qualitatively, with no formula or exact matching definition, and the paper does not report how decomposition failures affect downstream entity linking, SPARQL generation, and final answer accuracy. Since 77.61% TDA means a nontrivial fraction of test decompositions are incorrect, the paper needs an error analysis tracing those failures to end-to-end answers, and ideally a comparison against a non-decomposition baseline on the same test set.
  4. [§3.2, §4.4] The entity linking component is a critical part of the pipeline, yet no quantitative evaluation of entity linking accuracy is provided. Section 3.2 claims that the proposed NER method outperforms existing Persian tools, and Figure 6 describes a zero-shot linking process, but Table 3 only reports NER accuracy and the final system evaluation does not isolate linking errors. Linking errors will propagate directly into SPARQL generation and final answers, so the paper should report entity linking accuracy on the MRDCPQ test set and, if possible, an ablation that shows the contribution of the linking step to the end-to-end F1.
minor comments (5)
  1. [§3.1.1] The dataset description states that 'approximately 15,000 multi-hop complex questions were extracted from PeCoQ,' but the same paper describes PeCoQ as containing 10,000 questions. This inconsistency should be resolved and the exact filtering rules for the 5,600 selected questions should be stated.
  2. [§3, References] Reference [25] is cited as FarsBase in Section 3.1.1, but [25] is the Freebase paper; FarsBase is correctly cited as [31] and [32] elsewhere. The incorrect citation should be fixed.
  3. [§3.2, Figure 6] The text in Section 3.2 describes entity linking using cosine similarity, while Figure 6 mentions a cross-encoder and a ranking step. These are two different procedures; the paper should describe one consistent linking method and explain what is actually implemented.
  4. [§4.4, Table 4] Table 4 lacks information about the number of test questions, whether the same questions were used for both methods, and whether the reported numbers are averages over runs. Adding this information would make the comparison interpretable.
  5. [Throughout] The English language needs another round of editing: examples include 'Figures 8 demonstrates' in Section 3.4, and several Persian-to-English translations such as 'جهانن' and 'کد منطقهای' appear inconsistent with the English glosses. Typos and grammatical errors do not affect the technical content but should be corrected.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular derivation: the reported F1 and accuracy gains are the outcome of held-out evaluation on PeCoQ and are not constructed from the training labels.

full rationale

The paper's claimed derivation chain is empirical, not definitional. The decomposer (mT5, §3.1.2), the NER/linker (ParsBERT plus zero-shot dense retrieval, §3.2), and the SPARQL generator (mT5, §3.3) are all fine-tuned on MRDCPQ and then measured with standard metrics on the PeCoQ test dataset (§4.1–§4.4, Table 4). None of the reported numbers, including the headline 12.57% F1 and 12.06% accuracy improvements, is obtained by substituting a fitted parameter back into the same equation that defined it; the comparison against Etezadi et al. is an external evaluation on a separate system's dataset. There is no self-citation chain that supplies the core claim, no uniqueness theorem imported from the authors' prior work, and no ansatz concealed in a citation. The main legitimate concern is data provenance rather than circularity: MRDCPQ is built 'based on the PeCoQ dataset' (§3.1.1), and the paper does not state whether the PeCoQ test questions used in Table 4 are disjoint from the 80% MRDCPQ training split. That omission is a potential leakage/external-validity risk, but the paper provides no quoted mapping that would let one exhibit a specific reduction of the prediction to the training input, so it does not meet the standard for a circularity finding under the stated rules. Accordingly, the appropriate circularity score is 0.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central claim rests on assumptions about dataset annotation quality, semantic equivalence of decomposition, knowledge graph coverage, baseline comparability, and representativeness of the curated subset. None of these are machine-checked or released, and the selection filter is an explicit threat to the reported gains.

free parameters (2)
  • Trained weights of mT5 and ParsBERT models = Not reported or released
    All three modules are fine-tuned on MRDCPQ-derived data; their weights are fit to the training split and directly determine end-to-end F1 and accuracy. No model checkpoints are provided.
  • Entity linking candidate count k = 5
    Section 3.2 states 'we first extract five candidates from the FarsBase knowledge graph using string similarity.' No sensitivity analysis for k is reported.
assumptions (5)
  • domain assumption PeCoQ questions, SPARQL queries, named entities, and relations are correctly annotated.
    PeCoQ is the sole source for training and evaluation; errors in it propagate through every component (Sections 3.1.1 and 3.3).
  • ad hoc to paper A BREAK-style semantic decomposition preserves the answer semantics of the original complex question.
    The method assumes that sequential execution of per-segment SPARQL queries recovers the original answer (Section 3.4, Figure 8); no proof or error analysis for semantic equivalence is provided.
  • domain assumption FarsBase has sufficient coverage for all entities and relations used in PeCoQ questions.
    Entity linking and SPARQL execution rely on FarsBase; as the only Persian KG used, its completeness is assumed (Section 3.2).
  • domain assumption Etezadi et al.'s reported metrics are computed on the same PeCoQ test split with the same evaluation normalization.
    Section 4.4 directly compares absolute numbers without re-running the baseline or describing the test split used by Etezadi et al.
  • ad hoc to paper The curated 5,600-question subset is representative of the full PeCoQ distribution.
    Section 3.1.1 keeps only questions that annotators could decompose correctly; the effect of this filter on difficulty is not analyzed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Method for Multi-Hop Question Answering on Persian Knowledge Graph." pith.science (2026). https://pith.science/paper/5GK7F7CL

@misc{pith2026250116350,
  author       = {Pith},
  title        = {Pith review of: A Method for Multi-Hop Question Answering on Persian Knowledge Graph},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5GK7F7CL}},
  note         = {Machine review of arXiv:2501.16350}
}
read the original abstract

Question answering systems are the latest evolution in information retrieval technology, designed to accept complex queries in natural language and provide accurate answers using both unstructured and structured knowledge sources. Knowledge Graph Question Answering (KGQA) systems fulfill users' information needs by utilizing structured data, representing a vast number of facts as a graph. However, despite significant advancements, major challenges persist in answering multi-hop complex questions, particularly in Persian. One of the main challenges is the accurate understanding and transformation of these multi-hop complex questions into semantically equivalent SPARQL queries, which allows for precise answer retrieval from knowledge graphs. In this study, to address this issue, a dataset of 5,600 Persian multi-hop complex questions was developed, along with their decomposed forms based on the semantic representation of the questions. Following this, Persian language models were trained using this dataset, and an architecture was proposed for answering complex questions using a Persian knowledge graph. Finally, the proposed method was evaluated against similar systems on the PeCoQ dataset. The results demonstrated the superiority of our approach, with an improvement of 12.57% in F1-score and 12.06% in accuracy compared to the best comparable method.

Figures

Figures reproduced from arXiv: 2501.16350 by the authors.

Figure 3
Figure 3. This image illustrates the abstract structure of a knowledge graph, which consists of three components: the subject, predicate, and object, forming a Resource Description Framework (RDF) triple. In the diagram, Ali Daei is shown as the subject, connected to Ardabil as the object through the predicate "place of birth". Additionally, Ardabil, which served as the object in the previous triple, now becomes the subject i… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 18 canonical work pages

  1. [1]

    Faceted search over RDF -based knowledge graphs,

    M. Arenas, B. Cuenca Grau, E. Kharlamov, Š. Marciuška, and D. Zheleznyakov, “Faceted search over RDF -based knowledge graphs,” J. Web Semant., vol. 37–38, pp. 55–74, Mar. 2016, doi: 10.1016/J.WEBSEM.2015.12.002

    work page doi:10.1016/j.websem.2015.12.002 2016

  2. [2]

    Hic -KGQA: Improving multi -hop question answering over knowledge graph via hypergraph and inference chain,

    J. Wang, W. Li, F. Liu, B. Sheng, W. Liu, and Q. Jin, “Hic -KGQA: Improving multi -hop question answering over knowledge graph via hypergraph and inference chain,” Knowledge-Based Syst. , vol. 277, p. 110810, 2023, doi: https://doi.org/10.1016/j.knosys.2023.110810

    work page arXiv 2023

  3. [7]

    Semantic parsing on freebase from question -answer pairs,

    J. Berant, A. Chou, R. Frostig, and P. Liang, “Semantic parsing on freebase from question -answer pairs,” in EMNLP 2013 - 2013 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference , Association for Computational Linguistics, 2013, pp. 1533–1544. [Online]. Available: http://nlp.stanford.edu/

    work page 2013

  4. [8]

    More accurate question answering on freebase,

    H. Bast and E. Haussmann, “More accurate question answering on freebase,” Int. Conf. Inf. Knowl. Manag. Proc. , vol. 19- 23-Oct-2015, pp. 1431–1440, Oct. 2015, doi: 10.1145/2806416.2806472

    work page arXiv 2015

  5. [9]

    A Survey on Complex Knowledge Base Question Answering: Methods, Challenges and Solutions,

    Y. Lan, G. He, J. Jiang, J. Jiang, W. X. Zhao, and J.-R. Wen, “A Survey on Complex Knowledge Base Question Answering: Methods, Challenges and Solutions,” in Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, California: International Joint Conferences on Artificial Intelligence Organization, Aug. 2021, pp. 4483 –4491. ...

    work page doi:10.24963/ijcai.2021/611 2021

  6. [10]

    Case-based Reasoning for Natural Language Queries over Knowledge Bases

    R. Das et al. , “Case-based Reasoning for Natural Language Queries over Knowledge Bases.” 2021. [Online]. Available: https://arxiv.org/abs/2104.08762

    arXiv 2021

  7. [11]

    {RNG} -{KBQA}: Generation Augmented Iterative Ranking for Knowledge Base Question Answering,

    X. Ye, S. Yavuz, K. Hashimoto, Y. Zhou, and C. Xiong, “{RNG} -{KBQA}: Generation Augmented Iterative Ranking for Knowledge Base Question Answering,” in Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , S. Muresan, P. Nakov, and A. Villavicencio, Eds., Dublin, Ireland: Association for Computat...

    doi:10.18653/v1/2022.acl-long.417 2022

  8. [12]

    {TIARA}: Multi -grained Retrieval for Robust Question Answering over Large Knowledge Base,

    Y. Shu et al. , “{TIARA}: Multi -grained Retrieval for Robust Question Answering over Large Knowledge Base,” in Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing , Y. Goldberg, Z. Kozareva, and Y. Zhang, Eds., Abu Dhabi, United Arab Emirates: Association for Computational Linguistics, Dec. 2022, pp. 8108 –

    work page 2022

Show all 37 references
  1. [13]

    DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Bases

    D. Yu et al. , “DecAF: Joint Decoding of Answers and Logical Forms for Question Answering over Knowledge Bases.”

  2. [14]

    Program Transfer for Answering Complex Questions over Knowledge Bases

    S. Cao et al. , “Program Transfer for Answering Complex Questions over Knowledge Bases.” 2022. [Online]. Available: https://arxiv.org/abs/2110.05743

    2022 arXiv

  3. [15]

    A Survey on Complex Question Answering over Knowledge Base: Recent Advances and Challenges,

    B. Fu, Y. Qiu, C. Tang, Y. Li, H. Yu, and J. Sun, “A Survey on Complex Question Answering over Knowledge Base: Recent Advances and Challenges,” no. 1, pp. 1–19, 2020, [Online]. Available: http://arxiv.org/abs/2007.13069

    2020 arXiv

  4. [16]

    Enhancing Complex Question Answering over Knowledge Graphs through Evidence Pattern Retrieval

    W. Ding, J. Li, L. Luo, and Y. Qu, “Enhancing Complex Question Answering over Knowledge Graphs through Evidence Pattern Retrieval.” 2024. [Online]. Available: https://arxiv.org/abs/2402.02175

    2024 arXiv

  5. [17]

    Improving Multi -hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings,

    A. Saxena, A. Tripathi, and P. Talukdar, “Improving Multi -hop Question Answering over Knowledge Graphs using Knowledge Base Embeddings,” Proc. 58th Annu. Meet. Assoc. Comput. Linguist. , pp. 4498 –4507, 2020, doi: 10.18653/v1/2020.acl-main.412

    2020 doi

  6. [18]

    Improving Multi -hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals,

    G. He, Y. Lan, J. Jiang, W. X. Zhao, and J. R. Wen, “Improving Multi -hop Knowledge Base Question Answering by Learning Intermediate Supervision Signals,” in WSDM 2021 - Proceedings of the 14th ACM International Conference on Web Search and Data Mining, 2021, pp. 553–561. doi:...

    2021

  7. [19]

    Subgraph Retrieval Enhanced Model for Multi -hop Knowledge Base Question Answering,

    J. Zhang et al., “Subgraph Retrieval Enhanced Model for Multi -hop Knowledge Base Question Answering,” in Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) , Association for Computational Linguistics, 2022. doi: 10....

    2022 doi

  8. [20]

    UniKGQA: Unified Retrieval and Reasoning for Solving Multi -hop Question Answering Over Knowledge Graph

    J. Jiang, K. Zhou, W. X. Zhao, and J. -R. Wen, “UniKGQA: Unified Retrieval and Reasoning for Solving Multi -hop Question Answering Over Knowledge Graph.” 2023. [Online]. Available: https://arxiv.org/abs/2212.00959

    2023 arXiv

  9. [21]

    Multi -hop Question Answering over Knowledge Graphs using Large Language Models

    A. Chakraborty, “Multi -hop Question Answering over Knowledge Graphs using Large Language Models.” 2024. [Online]. Available: https://arxiv.org/abs/2404.19234

    2024 arXiv

  10. [22]

    Knowledge Graph -augmented Language Models for Complex Question Answering,

    P. Sen, S. Mavadia, and A. Saffari, “Knowledge Graph -augmented Language Models for Complex Question Answering,” in Proceedings of the 1st Workshop on Natural Language Reasoning and Structured Explanations (NLRSE) , B. Dalvi Mishra, G. Durrett, P. Jansen, D. Neves Ribeiro, and...

    2023 doi

  11. [23]

    A Knowledge -based Approach for Answering Complex Questions in Persian

    R. Etezadi and M. Shamsfard, “A Knowledge -based Approach for Answering Complex Questions in Persian.” 2021. [Online]. Available: https://arxiv.org/abs/2107.02040

    2021 arXiv

  12. [24]

    Wikidata: A Free Collaborative Knowledgebase,

    D. Vrandečić and M. Krötzsch, “Wikidata: A Free Collaborative Knowledgebase,” Commun. ACM, vol. 57, no. 10, pp. 78 – 85, Sep. 2014, doi: 10.1145/2629489

    2014 doi

  13. [25]

    Freebase: A collaboratively created graph da tabase for structuring human knowledge,

    K. Bollacker, C. Evans, P. Paritosh, T. Sturge, and J. Taylor, “Freebase: A collaboratively created graph da tabase for structuring human knowledge,” Proc. ACM SIGMOD Int. Conf. Manag. Data , pp. 1247 –1249, 2008, doi: 10.1145/1376616.1376746

    2008

  14. [26]

    Yago: A core of semantic knowledge,

    F. M. Suchanek, G. Kasneci, and G. Weikum, “Yago: A core of semantic knowledge,” 16th Int. World Wide Web Conf. WWW2007, pp. 697–706, 2007, doi: 10.1145/1242572.1242667

    2007

  15. [27]

    DBpedia: A nucleus for a Web of open data,

    S. Auer, C. Bizer, G. Kobilarov, J. Lehmann, R. Cyganiak, and Z. Ives, “DBpedia: A nucleus for a Web of open data,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics) , vol. 4825 LNCS, pp. 722 –735, 2007, doi: 10.1007/978-3-540-7...

    2007 doi

  16. [28]

    Large -scale Simple Question Answering with Memory Networks,

    A. Bordes, N. Usunier, S. Chopra, and J. Weston, “Large -scale Simple Question Answering with Memory Networks,” Jun. 2015, [Online]. Available: http://arxiv.org/abs/1506.02075

    2015 arXiv

  17. [29]

    Building event-centric knowledge graphs from news,

    M. Rospocher et al., “Building event-centric knowledge graphs from news,” J. Web Semant., vol. 37–38, pp. 132–151, Mar. 2016, doi: 10.1016/j.websem.2015.12.004

    2016 doi

  18. [30]

    Towards a definition of knowledge graphs,

    L. Ehrlinger and W. Wöß, “Towards a definition of knowledge graphs,” CEUR Workshop Proceedings, vol. 1695. pp. 1--4,- -1--4, 2016

    2016

  19. [31]

    FarsBase: A cross -domain farsi knowledge graph,

    M. B. Sajadi, B. M. Bidgoli, and A. Hadian, “FarsBase: A cross -domain farsi knowledge graph,” CEUR Workshop Proc. , vol. 2198, pp. 4–7, 2018

    2018

  20. [32]

    FarsBase: The Persian knowledge graph,

    M. Asgari-Bidhendi, A. Hadian, and B. Minaei -Bidgoli, “FarsBase: The Persian knowledge graph,” Semant. Web, vol. 10, no. 6, pp. 1169–1196, 2019, doi: 10.3233/SW-190369

    2019 doi

  21. [33]

    A survey of RDF storage approaches,

    D. C. Faye, O. Curé, and G. Blin, “A survey of RDF storage approaches,” Rev. Africaine la Rech. en Inform. Mathématiques Appliquées, vol. Volume 15, Sep. 2012, doi: 10.46298/ARIMA.1956

    2012 doi

  22. [34]

    PeCoQ: A Dataset for Persian Complex Question Answering over Knowledge Graph,

    R. Etezadi and M. Shamsfard, “PeCoQ: A Dataset for Persian Complex Question Answering over Knowledge Graph,” in 2020 11th International Conference on Information and Knowledge Technology (IKT) , IEEE, Dec. 2020, pp. 102 –106. doi: 10.1109/IKT51791.2020.9345610

    2020

  23. [35]

    Break it down: A question understanding benchmark,

    T. Wolfson et al., “Break it down: A question understanding benchmark,” Trans. Assoc. Comput. Linguist., vol. 8, pp. 183– 198, 2020, doi: 10.1162/TACL_A_00309/1923705/TACL_A_00309.PDF

    2020 doi

  24. [36]

    Break it down: A question understanding benchmark,

    T. Wolfson et al., “Break it down: A question understanding benchmark,” Trans. Assoc. Comput. Linguist., vol. 8, pp. 183– 198, Jan. 2020, doi: 10.1162/tacl_a_00309

    2020 doi

  25. [37]

    mT5: A massively multilingual pre -trained text -to-text transformer

    L. Xue et al. , “mT5: A massively multilingual pre -trained text -to-text transformer.” arXiv, 2020. doi: 10.48550/ARXIV.2010.11934

    2020 doi arXiv

  26. [38]

    ParsBERT: Transformer -based model for Persian language understanding,

    M. Farahani, M. Gharachorloo, M. Farahani, and M. Manthouri, “ParsBERT: Transformer -based model for Persian language understanding,” arXiv. arXiv, May 2020. [Online]. Available: http://arxiv.org/abs/2005.12515

    2020 arXiv

  27. [39]

    Scalable Zero -shot Entity Linking with Dense Entity Retrieval

    L. Wu, F. Petroni, M. Josifoski, S. Riedel, and L. Zettlemoyer, “Scalable Zero -shot Entity Linking with Dense Entity Retrieval.” 2020. [Online]. Available: https://arxiv.org/abs/1911.03814 Authors’ Profiles Arash Ghafouri: He is a PhD candidate in Computer Engineering at the ...

    2020 arXiv

  28. [2023]

    Available: https://arxiv.org/abs/2210.00063

    [Online]. Available: https://arxiv.org/abs/2210.00063

    arXiv

  29. [8121]

    doi: 10.18653/v1/2022.emnlp-main.555

    2022 doi

Pith tools

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