pith. sign in

arxiv: 2411.14072 · v1 · submitted 2024-11-21 · 💻 cs.CL · cs.PL

The Master-Slave Encoder Model for Improving Patent Text Summarization: A New Approach to Combining Specifications and Claims

Shu Zhou , Xin Wang , Zhengda Zhou , Haohan Yi , Xuhui Zheng , Hao Wan This is my paper

Pith reviewed 2026-05-23 17:26 UTC · model grok-4.3

classification 💻 cs.CL cs.PL
keywords patent text summarizationmaster-slave encoderpointer networkrepetition suppressiontext generationnatural language processingROUGE evaluation
0
0 comments X

The pith

The master-slave encoder model improves patent text summarization by combining specifications and claims as paired inputs.

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

The paper presents the MSEA model to generate patent abstracts that overcome limitations of models using only specifications. It introduces a master-slave encoder to jointly process specifications and claims, a pointer network to handle new technical terms, and an enhanced repetition suppression mechanism to cut redundancy. These components are applied to the high-professionalism and uniqueness of patent language. On a public dataset the model records small ROUGE gains over the prior Improved Multi-Head Attention Mechanism baseline. Readers interested in technical document summarization would see value in an architecture that explicitly links two distinct patent sections rather than treating the text as a single stream.

Core claim

The MSEA model designs a master-slave encoder that takes patent instructions and claims together as input to explore their characteristics and details, augments the encoder output with a pointer network to emphasize new technical terms, re-weights remembered and forgotten sequence parts for stronger input correlation, and applies an enhanced repetition suppression mechanism to produce accurate, non-redundant abstracts.

What carries the argument

Master-slave encoder architecture that processes specifications and claims as complementary inputs to capture their interrelations before decoding.

If this is right

  • Summaries handle out-of-vocabulary technical terms more reliably through the pointer network.
  • Generated abstracts exhibit lower information redundancy due to the repetition suppression mechanism.
  • The model produces higher ROUGE-1, ROUGE-2, and ROUGE-L scores than the Improved Multi-Head Attention Mechanism baseline.
  • Patent-specific accuracy and uniqueness are better preserved by jointly encoding specifications and claims.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same encoder pairing might be tested on other paired legal-technical texts such as regulatory filings.
  • A larger-scale experiment across multiple patent offices could show whether the gains hold when terminology distributions shift.
  • Replacing the pointer component with a modern retrieval module could be checked to see if the master-slave structure still adds value.

Load-bearing premise

That the master-slave encoder combined with pointer network and repetition suppression will effectively mitigate OOV terms, information redundancy, and generation quality issues in patent texts without additional data or validation.

What would settle it

Evaluating MSEA against IMHAM on a fresh patent corpus containing many novel technical terms and measuring whether ROUGE scores stay higher and whether the generated abstracts contain fewer factual inaccuracies or repetitions.

Figures

Figures reproduced from arXiv: 2411.14072 by Haohan Yi, Hao Wan, Shu Zhou, Xin Wang, Xuhui Zheng, Zhengda Zhou.

Figure 1
Figure 1. Figure 1: The overall architecture of the model MSEA. MSEA has a master encoder, [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
read the original abstract

In order to solve the problem of insufficient generation quality caused by traditional patent text abstract generation models only originating from patent specifications, the problem of new terminology OOV caused by rapid patent updates, and the problem of information redundancy caused by insufficient consideration of the high professionalism, accuracy, and uniqueness of patent texts, we proposes a patent text abstract generation model (MSEA) based on a master-slave encoder architecture; Firstly, the MSEA model designs a master-slave encoder, which combines the instructions in the patent text with the claims as input, and fully explores the characteristics and details between the two through the master-slave encoder; Then, the model enhances the consideration of new technical terms in the input sequence based on the pointer network, and further enhances the correlation with the input text by re weighing the "remembered" and "for-gotten" parts of the input sequence from the encoder; Finally, an enhanced repetition suppression mechanism for patent text was introduced to ensure accurate and non redundant abstracts generated. On a publicly available patent text dataset, compared to the state-of-the-art model, Improved Multi-Head Attention Mechanism (IMHAM), the MSEA model achieves an improvement of 0.006, 0.005, and 0.005 in Rouge-1, Rouge-2, and Rouge-L scores, respectively. MSEA leverages the characteristics of patent texts to effectively enhance the quality of patent text generation, demonstrating its advancement and effectiveness in the experiments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 3 minor

Summary. The manuscript proposes the Master-Slave Encoder Architecture (MSEA) model for patent text summarization. It combines a master-slave encoder that takes patent specifications and claims as input, augments it with a pointer network for OOV terms, and adds an enhanced repetition suppression mechanism. On a publicly available patent dataset, MSEA is reported to improve ROUGE-1/2/L by 0.006/0.005/0.005 over the IMHAM baseline.

Significance. If the small reported gains prove robust and attributable to the proposed architecture, the work could supply a domain-adapted encoder design for handling the distinctive properties of patent text (OOV terminology, redundancy, and precision requirements). The absence of ablations and statistical validation, however, leaves the practical advance modest at best.

major comments (3)
  1. [Abstract] Abstract: the central claim of +0.006/+0.005/+0.005 ROUGE gains is presented with no error bars, no standard deviations across runs, no p-values, and no dataset size or number of test instances, so it is impossible to determine whether the deltas exceed run-to-run variance.
  2. [Abstract] Abstract: no ablation experiments are described that isolate the master-slave encoder from the pointer network and repetition-suppression components (both already common in summarization models), leaving open whether the architecture itself drives the reported improvement.
  3. [Abstract] Abstract: the dataset is characterized only as 'publicly available' with no information on size, train/test split, preprocessing steps, or confirmation that the IMHAM baseline was re-implemented under identical conditions, any of which could nullify the tiny absolute gains.
minor comments (3)
  1. [Abstract] Abstract contains grammatical error: 'we proposes' should read 'we propose'.
  2. [Abstract] Abstract contains hyphenation typo: 'for-gotten' should be 'forgotten'.
  3. [Abstract] The description of the 'enhanced repetition suppression mechanism' is too high-level; a concrete algorithmic difference from standard coverage or repetition penalties would improve clarity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and commit to revisions that provide the requested details and experiments.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim of +0.006/+0.005/+0.005 ROUGE gains is presented with no error bars, no standard deviations across runs, no p-values, and no dataset size or number of test instances, so it is impossible to determine whether the deltas exceed run-to-run variance.

    Authors: We agree that the abstract should include more statistical context. In revision we will add the test set size, number of instances, and either standard deviations from repeated runs (if we can obtain them) or an explicit statement that results are from single runs. The small deltas are reported exactly as obtained; we do not claim statistical significance beyond the raw scores. revision: yes

  2. Referee: [Abstract] Abstract: no ablation experiments are described that isolate the master-slave encoder from the pointer network and repetition-suppression components (both already common in summarization models), leaving open whether the architecture itself drives the reported improvement.

    Authors: The manuscript presents the integrated MSEA model. We acknowledge the absence of component-wise ablations. In the revision we will add ablation experiments that remove or replace the master-slave encoder while keeping the pointer network and repetition suppression fixed, to isolate its contribution. revision: yes

  3. Referee: [Abstract] Abstract: the dataset is characterized only as 'publicly available' with no information on size, train/test split, preprocessing steps, or confirmation that the IMHAM baseline was re-implemented under identical conditions, any of which could nullify the tiny absolute gains.

    Authors: We will name the specific public patent dataset, report its size, train/test split ratios, and preprocessing pipeline. We confirm that the IMHAM baseline was re-implemented by the authors on the identical data splits and preprocessing; this detail will be stated explicitly in the revision. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical test-set ROUGE deltas on public dataset are independent of architecture description

full rationale

The paper's central claim is an empirical result: on a held-out portion of a publicly available patent dataset, the proposed MSEA model records ROUGE-1/2/L gains of 0.006/0.005/0.005 over the IMHAM baseline. This is a standard train-then-evaluate protocol; the reported numbers are not obtained by fitting a parameter to the evaluation metric itself, nor do any equations or self-citations reduce the architecture description to the performance numbers by construction. No self-definitional loops, fitted-input-as-prediction, or load-bearing self-citations appear in the abstract or the described derivation chain. The result therefore remains falsifiable against external re-runs and does not collapse to its own inputs.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

Based solely on abstract, the model relies on standard neural network training assumptions and domain-specific design choices not independently verified from external benchmarks.

free parameters (1)
  • encoder architecture parameters
    Design choices for master and slave encoders and their interaction fitted during training on patent data.
axioms (2)
  • domain assumption Combining patent specifications and claims via master-slave encoder improves summary quality over specification-only models
    Core premise invoked to justify the architecture.
  • domain assumption Pointer network effectively handles new technical terms in patent text
    Assumed to solve OOV problem.

pith-pipeline@v0.9.0 · 5814 in / 1259 out tokens · 54619 ms · 2026-05-23T17:26:32.061713+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

149 extracted references · 149 canonical work pages · 2 internal anchors

  1. [1]

    presidential election (2016)

    Following Hillary Clinton and Donald Trump: A dissection of their tweets in the 2016 U.S. presidential election (2016)

    work page 2016

  2. [2]

    Cogent Economics & Finance 11(1), 2186039 (Dec 2023)

    Abdulahi, S.M., Yitayaw, M.K., Feyisa, H.L., Mamo, W.B.: Factor affecting tech- nical efficiency of the banking sector: Evidence from Ethiopia. Cogent Economics & Finance 11(1), 2186039 (Dec 2023). https://doi.org/10.1080/23322039. 2023.2186039

    work page doi:10.1080/23322039 2023

  3. [3]

    Applied Sciences13(15), 8815 (Jul 2023)

    Al-Saif, H.F., Al-Dossari, H.Z.: Exploring the Role of Emotions in Arabic Rumor Detection in Social Media. Applied Sciences13(15), 8815 (Jul 2023). https: //doi.org/10.3390/app13158815

    work page doi:10.3390/app13158815 2023

  4. [4]

    Multimedia Tools and Applications82(5), 7017–7038 (Feb 2023).https://doi.org/10.1007/s11042-022-13595-4

    Ali, G., Malik, M.S.I.: Rumour identification on Twitter as a function of novel textual and language-context features. Multimedia Tools and Applications82(5), 7017–7038 (Feb 2023).https://doi.org/10.1007/s11042-022-13595-4

    work page doi:10.1007/s11042-022-13595-4 2023

  5. [5]

    International Journal of Business Perfor- mance Management24(3-4), 286–303 (2023).https://doi.org/10.1504/IJBPM

    Armoti, A.A., Oswal, N., Jawabri, A.: The relationship between the recruitment strategy and the competitive advantage. International Journal of Business Perfor- mance Management24(3-4), 286–303 (2023).https://doi.org/10.1504/IJBPM. 2023.132313

    work page doi:10.1504/ijbpm 2023

  6. [6]

    Computing104(5), 1155–1171 (May 2022)

    Bai, N., Meng, F., Rui, X., Wang, Z.: Rumor detection based on a Source-Replies conversation Tree Convolutional Neural Net. Computing104(5), 1155–1171 (May 2022). https://doi.org/10.1007/s00607-021-01034-5

    work page doi:10.1007/s00607-021-01034-5 2022

  7. [7]

    https://doi.org/10.7759/cureus.35037

    Barton, M., Hamza, M., Guevel, B.: Racial Equity in Healthcare Machine Learn- ing:IllustratingBiasinModelsWithMinimalBiasMitigation.Cureus(Feb2023). https://doi.org/10.7759/cureus.35037

    work page doi:10.7759/cureus.35037

  8. [8]

    Natural Language Engineering 29(5), 1275–1304 (Sep 2023).https://doi.org/10.1017/S1351324922000195

    Baykara, B., Güngör, T.: Turkish abstractive text summarizationusingpretrained sequence-to-sequence models. Natural Language Engineering 29(5), 1275–1304 (Sep 2023).https://doi.org/10.1017/S1351324922000195

    work page doi:10.1017/s1351324922000195 2023

  9. [9]

    MIS Quarterly37(2), 471–482 (Feb 2013)

    Bharadwaj, A., El Sawy, O.A., University of Southern California, Pavlou, P.A., Temple University, Venkatraman, N., Boston University: Digital Business Strat- egy: Toward a Next Generation of Insights. MIS Quarterly37(2), 471–482 (Feb 2013). https://doi.org/10.25300/MISQ/2013/37:2.3

    work page doi:10.25300/misq/2013/37:2.3 2013

  10. [10]

    Bian, T., Xiao, X., Xu, T., Zhao, P., Huang, W., Rong, Y., Huang, J.: Rumor Detection on Social Media with Bi-Directional Graph Convolutional Networks (Jan 2020)

    work page 2020

  11. [11]

    Information13(1), 25 (Jan 2022)

    Bing, C., Wu, Y., Dong, F., Xu, S., Liu, X., Sun, S.: Dual Co-Attention-Based Multi-Feature Fusion Method for Rumor Detection. Information13(1), 25 (Jan 2022). https://doi.org/10.3390/info13010025

    work page doi:10.3390/info13010025 2022

  12. [12]

    Multimedia Tools and Ap- plications 79(41-42), 31467–31496 (Nov 2020)

    Bordoloi, M., Chatterjee, P.C., Biswas, S.K., Purkayastha, B.: Keyword ex- traction using supervised cumulative TextRank. Multimedia Tools and Ap- plications 79(41-42), 31467–31496 (Nov 2020). https://doi.org/10.1007/ s11042-020-09335-1

    work page 2020

  13. [13]

    Engineering Applications of Artificial Intelligence123, 106137 (Aug 2023).https://doi.org/10.1016/j.engappai.2023.106137 16 S

    Boreshban, Y., Mirbostani, S.M., Ghassem-Sani, G., Mirroshandel, S.A., Amiri- parian, S.: Improving question answering performance using knowledge distilla- tion and active learning. Engineering Applications of Artificial Intelligence123, 106137 (Aug 2023).https://doi.org/10.1016/j.engappai.2023.106137 16 S. Zhou et al

    work page doi:10.1016/j.engappai.2023.106137 2023

  14. [14]

    In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

    Bosselut, A., Rashkin, H., Sap, M., Malaviya, C., Celikyilmaz, A., Choi, Y.: COMET: Commonsense Transformers for Automatic Knowledge Graph Con- struction. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. pp. 4762–4779. Association for Computational Lin- guistics, Florence, Italy (2019).https://doi.org/10.18653/v...

    work page doi:10.18653/v1/p19-1470 2019

  15. [15]

    Annals of Operations Research311(2), 611–639 (Apr 2022).https://doi.org/ 10.1007/s10479-020-03817-z

    Boubaker, S., Do, D.T., Hammami, H., Ly, K.C.: The role of bank affiliation in bank efficiency: A fuzzy multi-objective data envelopment analysis approach. Annals of Operations Research311(2), 611–639 (Apr 2022).https://doi.org/ 10.1007/s10479-020-03817-z

    work page doi:10.1007/s10479-020-03817-z 2022

  16. [16]

    Computer Speech & Lan- guage 63, 101080 (Sep 2020).https://doi.org/10.1016/j.csl.2020.101080

    Boussaha, B.E.A., Hernandez, N., Jacquin, C., Morin, E.: End-to-end response se- lection based on multi-level context response matching. Computer Speech & Lan- guage 63, 101080 (Sep 2020).https://doi.org/10.1016/j.csl.2020.101080

    work page doi:10.1016/j.csl.2020.101080 2020

  17. [17]

    Cadene, R., Dancette, C.: RUBi: Reducing Unimodal Biases for Visual Question Answering

    work page

  18. [18]

    IEEE Intelligent Systems37(1), 25–37 (Jan 2022).https://doi.org/10.1109/MIS.2022.3150944

    Cao, L.: A New Age of AI: Features and Futures. IEEE Intelligent Systems37(1), 25–37 (Jan 2022).https://doi.org/10.1109/MIS.2022.3150944

    work page doi:10.1109/mis.2022.3150944 2022

  19. [19]

    Security and Communication Networks2022, 6785828 (Apr 2022).https://doi.org/10.1155/2022/6785828

    Chen, J., Wei, N., Yang, H.: Immune Algorithm to Suppress Rumor Propagation Based on Influence Maximization. Security and Communication Networks2022, 6785828 (Apr 2022).https://doi.org/10.1155/2022/6785828

    work page doi:10.1155/2022/6785828 2022

  20. [20]

    Pattern Recognition Letters 105, 226–233 (Apr 2018).https://doi.org/10.1016/j.patrec.2017.10.014

    Chen, W., Zhang, Y., Yeo, C.K., Lau, C.T., Lee, B.S.: Unsupervised rumor detec- tion based on users’ behaviors using neural networks. Pattern Recognition Letters 105, 226–233 (Apr 2018).https://doi.org/10.1016/j.patrec.2017.10.014

    work page doi:10.1016/j.patrec.2017.10.014 2018

  21. [21]

    EXPERT SYSTEMS WITH APPLICATIONS141, 112948 (Mar 2020)

    Chen, X., Jia, S., Xiang, Y.: A review: Knowledge reasoning over knowledge graph. EXPERT SYSTEMS WITH APPLICATIONS141, 112948 (Mar 2020). https://doi.org/10.1016/j.eswa.2019.112948

    work page doi:10.1016/j.eswa.2019.112948 2020

  22. [22]

    Knowledge-Based Systems240, 108085 (Mar 2022).https://doi.org/10.1016/j.knosys.2021.108085

    Chen, X., Zhou, F., Trajcevski, G., Bonsangue, M.: Multi-view learning with dis- tinguishable feature fusion for rumor detection. Knowledge-Based Systems240, 108085 (Mar 2022).https://doi.org/10.1016/j.knosys.2021.108085

    work page doi:10.1016/j.knosys.2021.108085 2022

  23. [23]

    arXiv preprint arXiv:2402.05391 (2024)

    Chen, Z., Zhang, Y., Fang, Y., Geng, Y., Guo, L., Chen, X., Li, Q., Zhang, W., Chen, J., Zhu, Y., et al.: Knowledge graphs meet multi-modal learning: A comprehensive survey. arXiv preprint arXiv:2402.05391 (2024)

    work page arXiv 2024

  24. [24]

    In: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

    Choi,E.,Hewlett,D.,Uszkoreit,J.,Polosukhin,I.,Lacoste,A.,Berant,J.:Coarse- to-Fine Question Answering for Long Documents. In: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). pp. 209–220. Association for Computational Linguistics, Vancou- ver, Canada (2017).https://doi.org/10.18653/v1/P17-1020

    work page doi:10.18653/v1/p17-1020 2017

  25. [26]

    Proceedings of the Association for Information Science and Tech- nology 60(1), 538–542 (Oct 2023).https://doi.org/10.1002/pra2.812

    Choi, K.: Computational Thematic Analysis of Poetry via Bimodal Large Lan- guage Models. Proceedings of the Association for Information Science and Tech- nology 60(1), 538–542 (Oct 2023).https://doi.org/10.1002/pra2.812

    work page doi:10.1002/pra2.812 2023

  26. [27]

    In: ICLR (2020),https:// openreview.net/pdf?id=r1xMH1BtvB

    Clark, K., Luong, M.T., Le, Q.V., Manning, C.D.: ELECTRA: Pre-training text encoders as discriminators rather than generators. In: ICLR (2020),https:// openreview.net/pdf?id=r1xMH1BtvB

    work page 2020

  27. [28]

    Applied Sciences 11(17), 7945 (Aug 2021).https://doi.org/10.3390/app11177945

    Dai, Y., Fu, Y., Yang, L.: A Multiple-Choice Machine Reading Comprehension Model with Multi-Granularity Semantic Reasoning. Applied Sciences 11(17), 7945 (Aug 2021).https://doi.org/10.3390/app11177945

    work page doi:10.3390/app11177945 2021

  28. [29]

    IEEE/ACM Transactions on audio, speech, and language processing27(4), 704–718 (2019) MSEA 17

    Deng, D., Jing, L., Yu, J., Sun, S., Ng, M.K.: Sentiment lexicon construction with hierarchical supervision topic model. IEEE/ACM Transactions on audio, speech, and language processing27(4), 704–718 (2019) MSEA 17

    work page 2019

  29. [30]

    In: Proceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial Intelligence

    Deng, Z., Zou, D., Jiang, D.: Multi-turn response selection with multi-level gran- ularity representations and 3D convolutional neural network. In: Proceedings of the 2020 4th High Performance Computing and Cluster Technologies Conference & 2020 3rd International Conference on Big Data and Artificial Intelligence. pp. 18–23. ACM, Qingdao China (Jul 2020)....

    work page doi:10.1145/3409501 2020

  30. [31]

    Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: Bert: Pre-training of deep bidirectionaltransformersforlanguageunderstanding.In:Proceedingsofthe2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). pp. 4171–4186 (2019)

    work page 2019

  31. [32]

    Devlin, J., Chang, M.W., Lee, K., Toutanova, K.: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. pp. 4171–4186 (2019). https://doi.org/10.18653/v1/N19-1423

    work page doi:10.18653/v1/n19-1423 2019

  32. [33]

    Ditkaew,K.:StrategicManagementAccountingonCompetitiveAdvantage.Inter- national Journal of Asian Business and Information Management10(2), 232–244 (2022)

    work page 2022

  33. [34]

    Journal of Su- percomputing 77(10), 12050–12074 (Oct 2021)

    Dong, X., Lian, Y., Chi, Y., Tang, X., Liu, Y.: A two-step rumor detec- tion model based on the supernetwork theory about Weibo. Journal of Su- percomputing 77(10), 12050–12074 (Oct 2021). https://doi.org/10.1007/ s11227-021-03748-x

    work page 2021

  34. [35]

    In: 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC)

    Dou, X., Li, M., Zhao, J., Gao, S.: A Text Classification Model Based on Vir- tual Adversarial Training and Bilateral Contrastive Learning. In: 2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC). pp. 1978–

    work page 2023

  35. [36]

    IEEE, Honolulu, Oahu, HI, USA (Oct 2023).https://doi.org/10.1109/ SMC53992.2023.10394238

    work page arXiv 2023

  36. [37]

    Draude, C., Klumbyte, G., Lücking, P., Treusch, P.: Situated algorithms: A so- ciotechnicalsystemicapproachtobias.OnlineInformationReview 44(2),325–342 (Nov 2019).https://doi.org/10.1108/OIR-10-2018-0332

    work page doi:10.1108/oir-10-2018-0332 2019

  37. [38]

    https://doi.org/10.20944/ preprints202103.0049.v1

    Garrido-Muñoz, I., Montejo-Ráez, A., Martínez-Santiago, F., Ureña-López, L.A.: A Survey on Bias in Deep NLP (Mar 2021). https://doi.org/10.20944/ preprints202103.0049.v1

    work page arXiv 2021

  38. [39]

    In: Li, X

    Ge, X., Zhang, M., Wei, B., Liu, Y.: A Rumor Detection Method Based on Graph Convolutional Network. In: Li, X. (ed.) Advances in Intelligent Automation and Soft Computing, vol. 80, pp. 423–429. Springer International Publishing, Cham (2022). https://doi.org/10.1007/978-3-030-81007-8_47

    work page doi:10.1007/978-3-030-81007-8_47 2022

  39. [40]

    In: Proceedings of the 2019 Conference on Empirical Methods in Natural Lan- guageProcessingandthe9thInternationalJointConferenceonNaturalLanguage Processing (EMNLP-IJCNLP)

    Ghosal, D., Majumder, N., Poria, S., Chhaya, N., Gelbukh, A.: DialogueGCN: A Graph Convolutional Neural Network for Emotion Recognition in Conversation. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Lan- guageProcessingandthe9thInternationalJointConferenceonNaturalLanguage Processing (EMNLP-IJCNLP). pp. 154–164. Association for Co...

    work page doi:10.18653/v1/d19-1015 2019

  40. [41]

    Simple rules

    Goby, V.P., Karimova, G.Z.: “Simple rules” as an approach to corporate selection of CSR strategies. International Journal of Organizational Analysis30(2), 197– 206 (Jan 2022).https://doi.org/10.1108/IJOA-07-2020-2320

    work page doi:10.1108/ijoa-07-2020-2320 2022

  41. [42]

    Artificial Intelligence 86(2), 269–357 (1996)

    Grosz, B.J., Kraus, S.: Collaborative plans for complex group action. Artificial Intelligence 86(2), 269–357 (1996)

    work page 1996

  42. [43]

    https://doi.org/10.1109/TASLP.2019.2955290 18 S

    Gu, J.C., Ling, Z.H., Liu, Q.: Utterance-to-Utterance Interactive Matching Net- workforMulti-TurnResponseSelectioninRetrieval-BasedChatbots.IEEE/ACM Transactions on Audio, Speech, and Language Processing28, 369–379 (2020). https://doi.org/10.1109/TASLP.2019.2955290 18 S. Zhou et al

    work page doi:10.1109/taslp.2019.2955290 2020

  43. [44]

    ACM Transactions on Asian and Low-Resource Language Information Processing21(1), 1–16 (Jan 2022).https://doi.org/10

    Gumaei, A., Al-Rakhami, M.S., Hassan, M.M., De Albuquerque, V.H.C., Cama- cho, D.: An Effective Approach for Rumor Detection of Arabic Tweets Using eX- treme Gradient Boosting Method. ACM Transactions on Asian and Low-Resource Language Information Processing21(1), 1–16 (Jan 2022).https://doi.org/10. 1145/3461697

    work page 2022

  44. [45]

    Guo, Y., Nie, L., Cheng, Z., Tian, Q., Zhang, M.: Loss re-scaling VQA: Revisiting the LanguagePrior Problem from a Class-imbalance View (Dec 2021)

    work page 2021

  45. [46]

    Data Analysis and Knowledge Discovery7(6), 61–72 (2023)

    Guoliang, S., Shu, Z., Yunfeng, W., Chunjiang, S., Liang, L.: Generating patent text abstracts based on improved multi-head attention mechanism. Data Analysis and Knowledge Discovery7(6), 61–72 (2023)

    work page 2023

  46. [47]

    In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

    Hu, M., Peng, Y., Huang, Z., Li, D.: Retrieve, Read, Rerank: Towards End- to-End Multi-Document Reading Comprehension. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. pp. 2285–2295. Association for Computational Linguistics, Florence, Italy (2019).https://doi. org/10.18653/v1/P19-1221

    work page doi:10.18653/v1/p19-1221 2019

  47. [48]

    In: Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

    Jia, R., Liang, P.: Adversarial examples for evaluating reading comprehension systems. In: Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. pp. 2021–2031 (2017)

    work page 2017

  48. [49]

    Proceedings of the AAAI Conference on Artificial Intelligence34(07), 11181–11188 (Apr 2020).https://doi.org/10

    Jing,C.,Wu,Y.,Zhang,X.,Jia,Y.,Wu,Q.:OvercomingLanguagePriorsinVQA via Decomposed Linguistic Representations. Proceedings of the AAAI Conference on Artificial Intelligence34(07), 11181–11188 (Apr 2020).https://doi.org/10. 1609/aaai.v34i07.6776

    work page 2020

  49. [50]

    Joshi, M., Chen, D., Liu, Y., Weld, D.S., Zettlemoyer, L., Levy, O.: SpanBERT: Improving Pre-training by Representing and Predicting Spans (Jan 2020)

    work page 2020

  50. [51]

    In: Proceedings of the 10th European Conference on Artificial Intelligence (ECAI)

    Kautz, H.A., Selman, B.: Planning as satisfiability. In: Proceedings of the 10th European Conference on Artificial Intelligence (ECAI). pp. 359–363 (1992)

    work page 1992

  51. [52]

    Proceedings of the As- sociation for Information Science and Technology 60(1), 203–212 (Oct 2023)

    Khoury, C., Owen-Smith, A., Joh, U., Duan, Y., Hemsley, J.: Multi-Modal Crisis Discourse and Collective Sensemaking onTikTok. Proceedings of the As- sociation for Information Science and Technology 60(1), 203–212 (Oct 2023). https://doi.org/10.1002/pra2.781

    work page doi:10.1002/pra2.781 2023

  52. [53]

    In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demonstrations

    Kudo, T., Richardson, J.: Sentencepiece: A simple and language independent subword tokenizer and detokenizer for neural text processing. In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. pp. 66–71 (2018)

    work page 2018

  53. [54]

    Data Analysis and Knowledge Discovery5(7), 59–69 (2021)

    Le, Z., Jidong, L., Xueqiang, L., Zhuo, C., Lei, W., Xindong, Y.: Rlcpar: A rewrit- ing model for chinese patent abstracts based on reinforcement learning. Data Analysis and Knowledge Discovery5(7), 59–69 (2021)

    work page 2021

  54. [55]

    Rehabilitation Counseling Bul- letin 64(3), 172–180 (Apr 2021).https://doi.org/10.1177/0034355220910238

    Levine, A., Park, J., Kuo, H.J.: Understanding Disability Biases in Undergraduate Rehabilitation Students: An Exploratory Study. Rehabilitation Counseling Bul- letin 64(3), 172–180 (Apr 2021).https://doi.org/10.1177/0034355220910238

    work page doi:10.1177/0034355220910238 2021

  55. [56]

    Li, J., Liu, M., Kan, M.Y., Zheng, Z., Wang, Z., Lei, W., Liu, T., Qin, B.: Mol- weni: A Challenge Multiparty Dialogue-based Machine Reading Comprehension Dataset with Discourse Structure

    work page

  56. [57]

    Li, J., Liu, M., Zheng, Z., Zhang, H., Qin, B., Kan, M.Y., Liu, T.: DADgraph: A Discourse-aware Dialogue Graph Neural Network for Multiparty Dialogue Ma- chine Reading Comprehension (Apr 2021)

    work page 2021

  57. [58]

    In: 2019 International Joint Conference on Neural Networks (IJCNN)

    Li, J., Zhang, C., Chen, X., Cao, Y., Liao, P., Zhang, P.: Abstractive Text Sum- marization with Multi-Head Attention. In: 2019 International Joint Conference on Neural Networks (IJCNN). pp. 1–8. IEEE, Budapest, Hungary (Jul 2019). https://doi.org/10.1109/IJCNN.2019.8851885 MSEA 19

    work page doi:10.1109/ijcnn.2019.8851885 2019

  58. [59]

    ACM Transactions on Information Systems39(4), 1–25 (Oct 2021)

    Li, J., Liu, C., Tao, C., Chan, Z., Zhao, D., Zhang, M., Yan, R.: Dialogue His- tory Matters! Personalized Response Selection in Multi-Turn Retrieval-Based Chatbots. ACM Transactions on Information Systems39(4), 1–25 (Oct 2021). https://doi.org/10.1145/3453183

    work page doi:10.1145/3453183 2021

  59. [60]

    Information Processing & Management58(1), 102415 (Jan 2021)

    Li, L., Li, C., Ji, D.: Deep context modeling for multi-turn response selection in dialogue systems. Information Processing & Management58(1), 102415 (Jan 2021). https://doi.org/10.1016/j.ipm.2020.102415

    work page doi:10.1016/j.ipm.2020.102415 2021

  60. [61]

    In: Findings of the Association for Computational Linguistics: EMNLP 2021

    Li, Y., Zhao, H.: Self- and Pseudo-self-supervised Prediction of Speaker and Key- utterance for Multi-party Dialogue Reading Comprehension. In: Findings of the Association for Computational Linguistics: EMNLP 2021. pp. 2053–2063. Asso- ciation for Computational Linguistics, Punta Cana, Dominican Republic (2021). https://doi.org/10.18653/v1/2021.findings-emnlp.176

    work page doi:10.18653/v1/2021.findings-emnlp.176 2021

  61. [62]

    In: Proceedings of the 2022 Con- ference on Empirical Methods in Natural Language Processing

    Li, Y., Zhao, H., Zhang, Z.: Back to the Future: Bidirectional Information Decou- pling Network for Multi-turn Dialogue Modeling. In: Proceedings of the 2022 Con- ference on Empirical Methods in Natural Language Processing. pp. 2761–2774. Association for Computational Linguistics, Abu Dhabi, United Arab Emirates (2022). https://doi.org/10.18653/v1/2022.em...

    work page doi:10.18653/v1/2022.emnlp-main.177 2022

  62. [63]

    Cybersecurity6(1), 49 (Sep 2023).https://doi.org/10

    Liew, S.R.C., Law, N.F.: Use of subword tokenization for domain generation al- gorithm classification. Cybersecurity6(1), 49 (Sep 2023).https://doi.org/10. 1186/s42400-023-00183-8

    work page 2023

  63. [64]

    In: Proceedings of the 2018 Con- ference on Empirical Methods in Natural Language Processing

    Lin, J., Su, Q., Yang, P., Ma, S., Sun, X.: Semantic-Unit-Based Dilated Con- volution for Multi-Label Text Classification. In: Proceedings of the 2018 Con- ference on Empirical Methods in Natural Language Processing. pp. 4554–4564. Association for Computational Linguistics, Brussels, Belgium (2018). https: //doi.org/10.18653/v1/D18-1485

    work page doi:10.18653/v1/d18-1485 2018

  64. [65]

    Proceedings of the AAAI Conference on Artificial Intelligence35(15), 13406–13414 (May 2021)

    Liu, L., Zhang, Z., Zhao, H., Zhou, X., Zhou, X.: Filling the Gap of Utterance- aware and Speaker-aware Representation for Multi-turn Dialogue. Proceedings of the AAAI Conference on Artificial Intelligence35(15), 13406–13414 (May 2021). https://doi.org/10.1609/aaai.v35i15.17582

    work page doi:10.1609/aaai.v35i15.17582 2021

  65. [66]

    In: Proceedings of the 28th International Conference on Computational Linguistics

    Liu, M., Yang, E., Xiong, D., Zhang, Y., Meng, Y., Hu, C., Xu, J., Chen, Y.: A learning-exploring method to generate diverse paraphrases with multi-objective deep reinforcement learning. In: Proceedings of the 28th International Conference on Computational Linguistics. pp. 2310–2321 (2020)

    work page 2020

  66. [67]

    Journal of Intelligent & Robotic Systems 109(2), 39 (Oct 2023)

    Lu, Y., Guo, C., Dou, Y., Dai, X., Wang, F.Y.: Could ChatGPT Imagine: Content Control for Artistic Painting Generation Via Large Language Models. Journal of Intelligent & Robotic Systems 109(2), 39 (Oct 2023). https://doi.org/10. 1007/s10846-023-01956-6

    work page 2023

  67. [68]

    In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

    Ma, J., Gao, W., Wong, K.F.: Rumor Detection on Twitter with Tree-structured Recursive Neural Networks. In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). pp. 1980–

    work page 1980

  68. [69]

    https://doi.org/10.18653/v1/P18-1184

    Association for Computational Linguistics, Melbourne, Australia (2018). https://doi.org/10.18653/v1/P18-1184

    work page doi:10.18653/v1/p18-1184 2018

  69. [71]

    Zhou et al

    Manzini, T., Lim, Y.C., Tsvetkov, Y., Black, A.W.: Black is to Criminal as Cau- casian is to Police: Detecting and Removing Multiclass Bias in Word Embeddings (Jul 2019) 20 S. Zhou et al

    work page 2019

  70. [72]

    Marodon, R.: Can Development Banks Step Up to the Challenge of Sustainable Development? Review of Political Economy34(2), 268–285 (Apr 2022).https: //doi.org/10.1080/09538259.2021.1977542

    work page doi:10.1080/09538259.2021.1977542 2022

  71. [73]

    In: Proceedings of the 2004 conference on empirical methods in natural language processing

    Mihalcea, R., Tarau, P.: Textrank: Bringing order into text. In: Proceedings of the 2004 conference on empirical methods in natural language processing. pp. 404–411 (2004)

    work page 2004

  72. [74]

    In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Vol- ume 1: Long Papers)

    Mihaylov, T., Frank, A.: Knowledgeable Reader: Enhancing Cloze-Style Read- ing Comprehension with External Commonsense Knowledge. In: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Vol- ume 1: Long Papers). pp. 821–832. Association for Computational Linguistics, Melbourne, Australia (2018).https://doi.org/10.18653...

    work page doi:10.18653/v1/p18-1076 2018

  73. [75]

    In: Baden-Fuller, C., Mangematin, V

    Mikhalkina, T., Cabantous, L.: Business Model Innovation: How Iconic Business Models Emerge. In: Baden-Fuller, C., Mangematin, V. (eds.) Advances in Strate- gic Management, vol. 33, pp. 59–95. Emerald Group Publishing Limited (Oct 2015). https://doi.org/10.1108/S0742-332220150000033024

    work page doi:10.1108/s0742-332220150000033024 2015

  74. [77]

    In: Proceedings of the 56th An- nual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

    Min, S., Zhong, V., Socher, R., Xiong, C.: Efficient and Robust Question An- swering from Minimal Context over Documents. In: Proceedings of the 56th An- nual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). pp. 1725–1735. Association for Computational Linguistics, Melbourne, Australia (2018).https://doi.org/10.18653/v1/P18-1160

    work page doi:10.18653/v1/p18-1160 2018

  75. [78]

    In: 2016 International Conference on Circuit, power and computing technologies (IC- CPCT)

    Moratanch, N., Chitrakala, S.: A survey on abstractive text summarization. In: 2016 International Conference on Circuit, power and computing technologies (IC- CPCT). pp. 1–7. IEEE (2016)

    work page 2016

  76. [79]

    Nadeem, M., Bethke, A., Reddy, S.: StereoSet: Measuring stereotypical bias in pretrained language models (Apr 2020)

    work page 2020

  77. [80]

    In: Proceedings of the AAAI conference on artificial intelligence

    Nallapati,R.,Zhai,F.,Zhou,B.:Summarunner:Arecurrentneuralnetworkbased sequence model for extractive summarization of documents. In: Proceedings of the AAAI conference on artificial intelligence. vol. 31 (2017)

    work page 2017

  78. [81]

    In: Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning

    Nallapati,R., Zhou, B., dos Santos, C., Xiang, B.: Abstractive text summarization using sequence-to-sequence rnns and beyond. In: Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning. pp. 280–290 (2016)

    work page 2016

  79. [82]

    Nie, Y., Wang, S., Bansal, M.: Revealing the Importance of Semantic Re- trieval for Machine Reading at Scale. In: Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th Interna- tional Joint Conference on Natural Language Processing (EMNLP-IJCNLP). pp. 2553–2566. Association for Computational Linguistics, Hong K...

    work page doi:10.18653/v1/d19-1258 2019

  80. [83]

    Strategic Entrepreneurship Journal 9(1), 58–78 (Mar 2015).https: //doi.org/10.1002/sej.1192

    Osiyevskyy, O., Dewald, J.: Explorative Versus Exploitative Business Model Change: The Cognitive Antecedents of Firm-Level Responses to Disruptive In- novation. Strategic Entrepreneurship Journal 9(1), 58–78 (Mar 2015).https: //doi.org/10.1002/sej.1192

    work page doi:10.1002/sej.1192 2015

Showing first 80 references.