Exposing the Achilles’ heel of textual hate speech classifiers using indistinguishable adversarial examples
The accessibility of online hate speech has increased significantly, making it crucial for social-media companies to prioritize efforts to curb its spread. Although deep learning models demonstrate vulnerability to adversarial attacks, whether models fine-tuned for hate speech detection exhibit simi...
Uložené v:
| Vydané v: | Expert systems with applications Ročník 254; s. 124278 |
|---|---|
| Hlavní autori: | , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Elsevier Ltd
15.11.2024
|
| Predmet: | |
| ISSN: | 0957-4174 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | The accessibility of online hate speech has increased significantly, making it crucial for social-media companies to prioritize efforts to curb its spread. Although deep learning models demonstrate vulnerability to adversarial attacks, whether models fine-tuned for hate speech detection exhibit similar susceptibility remains underexplored. Textual adversarial attacks involve making subtle alterations to the original samples. These alterations are designed so that the adversarial examples produced can effectively deceive the target model, even when correctly classified by human observers. Though many approaches have been proposed to conduct word-level adversarial attacks on textual data, they face the obstacle of preserving the semantic coherence of texts during the generation of adversarial counterparts. Moreover, the adversarial examples produced are often easily distinguishable by human observers. This work presents a novel methodology that uses visually confusable glyphs and invisible characters to generate semantically and visually similar adversarial examples in a black-box setting. In the hate speech detection task context, our attack was effectively applied to several state-of-the-art deep learning models, fine-tuned on two benchmark datasets. The major contributions of this study are: (1) demonstrating the vulnerability of deep learning models fine-tuned for hate speech detection; (2) a novel attack framework based on a simple yet potent modification strategy; (3) superior outcomes in terms of accuracy degradation, attack success rate, average perturbation, semantic similarity, and perplexity when compared to existing baselines; (4) strict adherence to prescribed linguistic constraints while formulating adversarial samples; and (5) preservation of ground truth label while perturbing original input using imperceptible adversarial examples. |
|---|---|
| AbstractList | The accessibility of online hate speech has increased significantly, making it crucial for social-media companies to prioritize efforts to curb its spread. Although deep learning models demonstrate vulnerability to adversarial attacks, whether models fine-tuned for hate speech detection exhibit similar susceptibility remains underexplored. Textual adversarial attacks involve making subtle alterations to the original samples. These alterations are designed so that the adversarial examples produced can effectively deceive the target model, even when correctly classified by human observers. Though many approaches have been proposed to conduct word-level adversarial attacks on textual data, they face the obstacle of preserving the semantic coherence of texts during the generation of adversarial counterparts. Moreover, the adversarial examples produced are often easily distinguishable by human observers. This work presents a novel methodology that uses visually confusable glyphs and invisible characters to generate semantically and visually similar adversarial examples in a black-box setting. In the hate speech detection task context, our attack was effectively applied to several state-of-the-art deep learning models, fine-tuned on two benchmark datasets. The major contributions of this study are: (1) demonstrating the vulnerability of deep learning models fine-tuned for hate speech detection; (2) a novel attack framework based on a simple yet potent modification strategy; (3) superior outcomes in terms of accuracy degradation, attack success rate, average perturbation, semantic similarity, and perplexity when compared to existing baselines; (4) strict adherence to prescribed linguistic constraints while formulating adversarial samples; and (5) preservation of ground truth label while perturbing original input using imperceptible adversarial examples. |
| ArticleNumber | 124278 |
| Author | Vishwakarma, Dinesh Kumar Aggarwal, Sajal |
| Author_xml | – sequence: 1 givenname: Sajal surname: Aggarwal fullname: Aggarwal, Sajal email: sajalaggarwal_it20b11_08@dtu.ac.in – sequence: 2 givenname: Dinesh Kumar surname: Vishwakarma fullname: Vishwakarma, Dinesh Kumar email: dinesh@dtu.ac.in |
| BookMark | eNp9kE1OwzAQhb0oEm3hAqx8gZSxk9SJxKaqyo9UiQ2sLcceE1dpEsVuKTuuwfU4CS5lxaKredLT96T5JmTUdi0ScsNgxoDNbzcz9O9qxoFnM8YzLooRGUOZiyRjIrskE-83AEwAiDHZrg595137RkONdKFr1zTovz-_aI3Y0M7SgIewUw2tVUDqe0RdU90o7511OHi6-6Vda5wPMe2cr1XVIFVmH2s1uMjiQW37uHtFLqxqPF7_3Sl5vV-9LB-T9fPD03KxTnQKEBKubFlAJRhyBTBnqSiE5dqkJcxTpnUFhWJYWRELW5R5VuZQZcbkOVpTcZNOCT_t6qHzfkAr-8Ft1fAhGcijJLmRR0nyKEmeJEWo-AdpF1RwXRsG5Zrz6N0JxfjUPmqRXjtsNRo3oA7SdO4c_gNxNIqV |
| CitedBy_id | crossref_primary_10_1007_s10207_024_00925_w crossref_primary_10_1038_s41598_024_76632_2 crossref_primary_10_1016_j_neucom_2024_128263 crossref_primary_10_1016_j_neunet_2025_107850 crossref_primary_10_1080_07421222_2025_2520173 crossref_primary_10_1016_j_knosys_2024_112532 crossref_primary_10_1016_j_neucom_2025_129660 crossref_primary_10_1016_j_knosys_2025_113350 crossref_primary_10_1016_j_neunet_2024_106512 |
| Cites_doi | 10.1016/j.eswa.2023.120898 10.18653/v1/2023.eacl-main.149 10.14722/ndss.2019.23138 10.1016/j.inffus.2023.101869 10.3115/v1/D14-1162 10.1049/cit2.12028 10.1609/icwsm.v11i1.14955 10.18653/v1/E17-1040 10.1016/j.ins.2023.119237 10.1162/tacl_a_00300 10.1016/j.eswa.2023.122223 10.1016/j.neucom.2021.05.103 10.1109/SP.2017.49 10.1201/9781351251389-8 10.1016/j.aiopen.2022.10.001 10.18653/v1/P19-1103 10.1016/j.neucom.2023.126787 10.1109/EuroSP.2016.36 10.18653/v1/2020.sustainlp-1.17 10.18653/v1/2022.naacl-main.125 10.1016/j.eswa.2021.115458 10.1007/s00530-023-01051-8 10.18653/v1/2023.trustnlp-1.24 10.1063/1.1699114 10.1016/j.neucom.2023.01.071 10.1007/s11633-019-1211-x 10.18653/v1/W19-4824 10.18653/v1/2021.emnlp-main.374 10.18653/v1/2020.emnlp-demos.16 10.1093/biomet/57.1.97 10.1109/ICNN.1995.488968 10.1016/j.knosys.2023.110515 10.18653/v1/P19-1561 10.18653/v1/2020.emnlp-main.495 10.1016/j.eswa.2022.119342 10.1145/3052973.3053009 10.1007/s40747-021-00608-2 10.1016/j.eswa.2023.122894 10.1080/08839514.2023.2166719 10.18653/v1/S19-2102 10.18653/v1/2021.naacl-main.423 |
| ContentType | Journal Article |
| Copyright | 2024 Elsevier Ltd |
| Copyright_xml | – notice: 2024 Elsevier Ltd |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.eswa.2024.124278 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| ExternalDocumentID | 10_1016_j_eswa_2024_124278 S0957417424011448 |
| GroupedDBID | --K --M .DC .~1 0R~ 13V 1B1 1RT 1~. 1~5 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN 9JO AAAKF AABNK AACTN AAEDT AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AARIN AAXUO AAYFN ABBOA ABFNM ABMAC ABMVD ABUCO ACDAQ ACGFS ACHRH ACNTT ACRLP ACZNC ADBBV ADEZE ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGUMN AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJOXV AKRWK ALEQD ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD APLSM AXJTR BJAXD BKOJK BLXMC BNSAS CS3 DU5 EBS EFJIC EO8 EO9 EP2 EP3 F5P FDB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HAMUX IHE J1W JJJVA KOM LG9 LY1 LY7 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. PQQKQ Q38 RIG ROL RPZ SDF SDG SDP SDS SES SEW SPC SPCBC SSB SSD SSL SST SSV SSZ T5K TN5 ~G- 29G 9DU AAAKG AAQXK AATTM AAXKI AAYWO AAYXX ABJNI ABKBG ABUFD ABWVN ABXDB ACLOT ACNNM ACRPL ACVFH ADCNI ADJOM ADMUD ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKYEP ANKPU APXCP ASPBG AVWKF AZFZN CITATION EFKBS EFLBG EJD FEDTE FGOYB G-2 HLZ HVGLF HZ~ R2- SBC SET WUQ XPP ZMT ~HD |
| ID | FETCH-LOGICAL-c300t-2af980b71e2a00613787f2cd390631ccb08a1ebf7378f8954950b4dd55efdb2d3 |
| ISICitedReferencesCount | 9 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001263673200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 0957-4174 |
| IngestDate | Sat Nov 29 03:07:38 EST 2025 Tue Nov 18 22:31:01 EST 2025 Sat Aug 31 16:00:37 EDT 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Transformers Adversarial attack Glyphs Hate speech Offensive language Natural Language Processing (NLP) |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c300t-2af980b71e2a00613787f2cd390631ccb08a1ebf7378f8954950b4dd55efdb2d3 |
| ParticipantIDs | crossref_primary_10_1016_j_eswa_2024_124278 crossref_citationtrail_10_1016_j_eswa_2024_124278 elsevier_sciencedirect_doi_10_1016_j_eswa_2024_124278 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-11-15 |
| PublicationDateYYYYMMDD | 2024-11-15 |
| PublicationDate_xml | – month: 11 year: 2024 text: 2024-11-15 day: 15 |
| PublicationDecade | 2020 |
| PublicationTitle | Expert systems with applications |
| PublicationYear | 2024 |
| Publisher | Elsevier Ltd |
| Publisher_xml | – name: Elsevier Ltd |
| References | Li, Ma, Guo, Xue, Qiu (b0235) 2020 119–126. https://doi.org/10.18653/v1/2020.emnlp-demos.16. Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated Hate Speech Detection and the Problem of Offensive Language. Qi, F., Chen, Y., Zhang, X., Li, M., Liu, Z., & Sun, M. (2021). Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer. Liu, H., Cai, C., & Qi, Y. (2023). Expanding Scope: Adapting English Adversarial Attacks to Chinese. 5582–5591. https://doi.org/10.18653/v1/P19-1561. Sharma, Kandasamy, Kandasamy (b0410) 2021; 185 124–135. https://doi.org/10.18653/v1/2020.sustainlp-1.17. 276–286. https://doi.org/10.18653/v1/2023.trustnlp-1.24. Qi, F., Yang, C., Liu, Z., Dong, Q., Sun, M., & Dong, Z. (2019). Kim (b0190) 2014 Radford, A., Metz, L., & Chintala, S. (2015). Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. In Zhan, Yang, Wang, Zheng, Huang, Wang (b0495) 2023; 2023 Kennedy, J., & Eberhart, R. (1995). Particle swarm optimization. 1942–1948 vol.4. https://doi.org/10.1109/ICNN.1995.488968. Madry, Makelov, Schmidt, Tsipras, Vladu (b0280) 2018 Kurakin, A., Goodfellow, I. J., & Bengio, S. (2018). Adversarial Examples in the Physical World. In R. V. Yampolskiy (Ed.) Clark, K., Luong, M.-T., Le, Q. V., & Manning, C. D. (2020, April). Lauriola, Lavelli, Aiolli (b0210) 2022; 470 Iandola, F., Shaw, A., Krishna, R., & Keutzer, K. (2020). SqueezeBERT: What can computer vision teach NLP about efficient neural networks? Lei, Cao, Li, Zhou, Fang, Pechenizkiy (b0220) 2022; 2022 (n.d.). Retrieved 26 December 2023, from https://perspectiveapi.com/. Formento, Foo, Tuan, Ng (b0120) 2023; 2023 Eighth International Conference on Learning Representations. https://iclr.cc/virtual_2020/poster_r1xMH1BtvB.html. https://www.semanticscholar.org/paper/RoBERTa%3A-A-Robustly-Optimized-BERT-Pretraining-Liu-Ott/077f8329a7b6fa3b7c877a57b81eb6c18b5f87de. Madhu, Satapara, Modha, Mandl, Majumder (b0275) 2023; 215 Ren, S., Deng, Y., He, K., & Che, W. (2019). Generating Natural Language Adversarial Examples through Probability Weighted Word Saliency. Wang, B., Pei, H., Pan, B., Chen, Q., Wang, S., & Li, B. (2020). T3: Tree-Autoencoder Constrained Adversarial Text Generation for Targeted Attack. Vaswani, Shazeer, Parmar, Uszkoreit, Jones, Gomez, Kaiser, Polosukhin (b0425) 2017; 30 Zheng, Zhu (b0515) 2023 , Choi, Kim, Lee (b0080) 2022 . Mamta, Ekbal (b0005) 2022; 2022 Zang, Qi, Yang, Liu, Zhang, Liu, Sun (b0490) 2020 Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2020). Chen, Duan, Houthooft, Schulman, Sutskever, Abbeel (b0055) 2016 Carlini, N., & Wagner, D. (2017). 4569–4580. https://doi.org/10.18653/v1/2021.emnlp-main.374. 39–57. https://doi.org/10.1109/SP.2017.49. Chang, Gao, Yao, Xiong (b0050) 2023; 529 (arXiv:1901.09957). arXiv. https://doi.org/10.48550/arXiv.1901.09957. Chakraborty, Alam, Dey, Chattopadhyay, Mukhopadhyay (b0045) 2021; 6 Bajaj, Kumar Vishwakarma (b0020) 2023; 558 Lees, Tran, Tay, Sorensen, Gupta, Metzler, Vasserman (b0215) 2022 Pandey, Vishwakarma (b0335) 2023; 269 Moosavi-Dezfooli, Fawzi, Frossard (b0310) 2016; 2016 Zhao, Zhang, Xu, Yuan (b0505) 2022; 2022 Pavlopoulos, J., Thain, N., Dixon, L., & Androutsopoulos, I. (2019). ConvAI at SemEval-2019 Task 6: Offensive Language Identification and Categorization with Perspective and BERT. In J. May, E. Shutova, A. Herbelot, X. Zhu, M. Apidianaki, & S. M. Mohammad (Eds.) Cheng, Jiang, Macherey (b0065) 2019 Chhabra, Vishwakarma (b0070) 2023; 29 Devlin, Chang, Lee, Toutanova (b0100) 2019 Modas, Moosavi-Dezfooli, Frossard (b0295) 2019; 2019 Yang, Qi, Chen, Liu, Liu (b0465) 2023; 644 del Valle-Cano, Quijano-Sánchez, Liberatore, Gómez (b0420) 2023; 216 (pp. 571–576). Association for Computational Linguistics. https://doi.org/10.18653/v1/S19-2102. Papernot, N., McDaniel, P., Goodfellow, I., Jha, S., Celik, Z. B., & Swami, A. (2017). Practical Black-Box Attacks against Machine Learning. Mollas, Chrysopoulou, Karlos, Tsoumakas (b0300) 2022; 8 Macas, Wu, Fuertes (b0270) 2024; 238 Ebrahimi, Rao, Lowd, Dou (b0105) 2018 (pp. 1532–1543). Association for Computational Linguistics. https://doi.org/10.3115/v1/D14-1162. 1735–1746. https://doi.org/10.18653/v1/2022.naacl-main.125. Iyyer, Wieting, Gimpel, Zettlemoyer (b0165) 2018 Wang, Xu, Liu, Cheng, Li (b0445) 2022; 2022 Zhang, Zhou, Miao, Li (b0500) 2019 Mehrish, Majumder, Bharadwaj, Mihalcea, Poria (b0285) 2023; 99 Gao, Lanchantin, Soffa, Qi (b0130) 2018; 2018 Bajaj, Vishwakarma (b0025) 2023 Jiao, Yin, Shang, Jiang, Chen, Li, Wang, Liu (b0170) 2020; 2020 Cer, Yang, Kong, Hua, Limtiaco, St. John, Constant, Guajardo-Cespedes, Yuan, Tar, Strope, Kurzweil (b0040) 2018 Morris, J., Lifland, E., Yoo, J. Y., Grigsby, J., Jin, D., & Qi, Y. (2020). TextAttack: A Framework for Adversarial Attacks, Data Augmentation, and Adversarial Training in NLP. 6134–6150. https://doi.org/10.18653/v1/2020.emnlp-main.495. Zhao, Yuanzhe, Zhongtao, Yiming, Jun, Kang (b0510) 2022 Nguyen-Son, Ung, Hidano, Fukushima, Kiyomoto (b0325) 2022; 2022 Hayet, Yao, Luo (b0155) 2022; 2022 Yoo, Qi (b0480) 2021; 2021 Pruthi, D., Dhingra, B., & Lipton, Z. C. (2019). Combating Adversarial Misspellings with Robust Word Recognition. Gaiński, P., & Ba\lazy, K. (2023). Step by Step Loss Goes Very Far: Multi-Step Quantization for Adversarial Text Attacks. Joshi, Chen, Liu, Weld, Zettlemoyer, Levy (b0180) 2020; 8 Garg, Ramakrishnan (b0135) 2020 Oseledets, Khrulkov (b0330) 2018; 2018 Xu, Ma, Liu, Deb, Liu, Tang, Jain (b0450) 2020; 17 Zhou, Li, Min (b0525) 2022 Network and Distributed System Security Symposium, San Diego, CA. https://doi.org/10.14722/ndss.2019.23138. (1), Article 1. https://doi.org/10.1609/icwsm.v11i1.14955. Goodfellow, I. J., Shlens, J., & Szegedy, C. (2015). Deng, C., Liu, M., Qin, Y., Zhang, J., Duan, H.-X., & Sun, D. (2022). ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model. Yuan, Zhang, Chen, Wei (b0485) 2023; 2023 (1st ed., pp. 99–112). Chapman and Hall/CRC. https://doi.org/10.1201/9781351251389-8. Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., & Sutskever, I. (n.d.). 233–240. https://doi.org/10.18653/v1/W19-4824. Lin, Wang, Liu, Qiu (b0245) 2022; 3 Xu, He, Lyu, Qu, Haffari (b0455) 2022 506–519. https://doi.org/10.1145/3052973.3053009. Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019). RoBERTa: A Robustly Optimized BERT Pretraining Approach. Pennington, J., Socher, R., & Manning, C. (2014). GloVe: Global Vectors for Word Representation. In A. Moschitti, B. Pang, & W. Daelemans (Eds.) Mondal, I. (2021). BBAEG: Towards BERT-based Biomedical Adversarial Example Generation for Text Classification. Chen, Su, Wei (b0060) 2021 5378–5384. https://doi.org/10.18653/v1/2021.naacl-main.423. Morris, Lifland, Lanchantin, Ji, Qi (b0315) 2020; 2020 Lin, Gao, Yan, Moreno, Ren (b0240) 2021 Chiang, Lee (b0075) 2023; 2023 Yang, Dai, Yang, Carbonell, Salakhutdinov, Le (b0470) 2019; 32 (arXiv:1910.01108; Version 4). arXiv. http://arxiv.org/abs/1910.01108. Fang, Cheng, Liu, Wang (b0115) 2023; 2023 Zhou, P., Qi, Z., Zheng, S., Xu, J., Bao, H., & Xu, B. (2016). Text Classification Improved by Integrating Bidirectional LSTM with Two-dimensional Max Pooling. In Y. Matsumoto & R. Prasad (Eds.) Saleh, Alhothali, Moria (b0395) 2023; 37 Kumar, Maheshwary, Pudi (b0195) 2021; 2021 (arXiv:1412.6572). arXiv. https://doi.org/10.48550/arXiv.1412.6572. Li, J., Ji, S., Du, T., Li, B., & Wang, T. (2019). TextBugger: Generating Adversarial Text Against Real-world Applications. Hastings (b0150) 1970; 57 (arXiv:1909.11942; Version 6). arXiv. http://arxiv.org/abs/1909.11942. (pp. 3485–3495). The COLING 2016 Organizing Committee. https://aclanthology.org/C16-1329. https://doi.org/10.48550/arXiv.1511.06434. Gupta, Yadav, Vishwakarma (b0145) 2024; 244 Li, Zhang, Peng, Chen, Brockett, Sun, Dolan (b0225) 2021 Aggarwal, Vishwakarma (b0015) 2023 Liu, Chen, Liu, Song (b0260) 2016 Tsai, Y.-T., Yang, M.-C., & Chen, H.-Y. (2019). Adversarial Attack on Sentiment Classification. 5085–5097. https://aclanthology.org/2022.coling-1.451. (pp. 417–427). Association for Computational Linguistics. https://aclanthology.org/E17-1040. Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., & Soricut, R. (2020). Wan, J., Yang, J., Ma, S., Zhang, D., Zhang, W., Yu, Y., & Li, Z. (2022). PAEG: Phrase-level Adversarial Example Generation for Neural Machine Translation. Liu, Yu, Hu, Li, Lin, Ma, Yang, Wen (b0250) 2022 Papernot, McDaniel, Jha, Fredrikson, Celik, Swami (b0345) 2016; 2016 1085–1097. https://doi.org/10.18653/v1/P19-1103. Bao, Wang, Zhao (b0030) 2021; 2021 Eger, Benz (b0110) 2020 Zhu, Cheng, Gan, Sun, Goldstein, Liu (b0530) 2020 Yadav, Vishwakarma (b0460) 2023; 232 Ye, Zhang, Dong, Ji (b0475) 2021; 2021 Metropolis, Rosenbluth, Rosenbluth, Teller, Teller (b0290) 2004; 21 Salimans, Goodfellow, Zaremba, Cheung, Radford, Chen, Chen (b0400) 2016; 29 2038–2048. https://doi.org/10.18653/v1/2023.eacl-main.149. Verwimp, L., Pelemans, J., Van hamme, H., & Wambacq, P. (2017). Character-Word LSTM Language Models. In M. Lapata, P. Blunsom, & A. Koller (Eds.) Jin, Jin, Zhou, Szolovits (b0175) 2020 Aggarwal, Pandey, Vishwakarma (b0010) 2023; 2023 Bajaj (10.1016/j.eswa.2024.124278_b0020) 2023; 558 Zhu (10.1016/j.eswa.2024.124278_b0530) 2020 Lei (10.1016/j.eswa.2024.124278_b0220) 2022; 2022 Ebrahimi (10.1016/j.eswa.2024.124278_b0105) 2018 Sharma (10.1016/j.eswa.2024.124278_b0410) 2021; 185 10.1016/j.eswa.2024.124278_b0340 Yang (10.1016/j.eswa.2024.124278_b0470) 2019; 32 Joshi (10.1016/j.eswa.2024.124278_b0180) 2020; 8 10.1016/j.eswa.2024.124278_b0185 Zang (10.1016/j.eswa.2024.124278_b0490) 2020 Kumar (10.1016/j.eswa.2024.124278_b0195) 2021; 2021 Jin (10.1016/j.eswa.2024.124278_b0175) 2020 Hayet (10.1016/j.eswa.2024.124278_b0155) 2022; 2022 Eger (10.1016/j.eswa.2024.124278_b0110) 2020 Zhang (10.1016/j.eswa.2024.124278_b0500) 2019 Lin (10.1016/j.eswa.2024.124278_b0240) 2021 10.1016/j.eswa.2024.124278_b0205 Li (10.1016/j.eswa.2024.124278_b0235) 2020 Bajaj (10.1016/j.eswa.2024.124278_b0025) 2023 Garg (10.1016/j.eswa.2024.124278_b0135) 2020 Nguyen-Son (10.1016/j.eswa.2024.124278_b0325) 2022; 2022 Iyyer (10.1016/j.eswa.2024.124278_b0165) 2018 Chakraborty (10.1016/j.eswa.2024.124278_b0045) 2021; 6 Mollas (10.1016/j.eswa.2024.124278_b0300) 2022; 8 Macas (10.1016/j.eswa.2024.124278_b0270) 2024; 238 Chang (10.1016/j.eswa.2024.124278_b0050) 2023; 529 Zheng (10.1016/j.eswa.2024.124278_b0515) 2023 Zhan (10.1016/j.eswa.2024.124278_b0495) 2023; 2023 Choi (10.1016/j.eswa.2024.124278_b0080) 2022 10.1016/j.eswa.2024.124278_b0435 Cheng (10.1016/j.eswa.2024.124278_b0065) 2019 Lin (10.1016/j.eswa.2024.124278_b0245) 2022; 3 10.1016/j.eswa.2024.124278_b0440 Formento (10.1016/j.eswa.2024.124278_b0120) 2023; 2023 10.1016/j.eswa.2024.124278_b0320 10.1016/j.eswa.2024.124278_b0200 Lees (10.1016/j.eswa.2024.124278_b0215) 2022 Madry (10.1016/j.eswa.2024.124278_b0280) 2018 Ye (10.1016/j.eswa.2024.124278_b0475) 2021; 2021 10.1016/j.eswa.2024.124278_b0160 Vaswani (10.1016/j.eswa.2024.124278_b0425) 2017; 30 Yuan (10.1016/j.eswa.2024.124278_b0485) 2023; 2023 Yadav (10.1016/j.eswa.2024.124278_b0460) 2023; 232 del Valle-Cano (10.1016/j.eswa.2024.124278_b0420) 2023; 216 Li (10.1016/j.eswa.2024.124278_b0225) 2021 Gupta (10.1016/j.eswa.2024.124278_b0145) 2024; 244 Zhou (10.1016/j.eswa.2024.124278_b0525) 2022 Gao (10.1016/j.eswa.2024.124278_b0130) 2018; 2018 10.1016/j.eswa.2024.124278_b0305 Wang (10.1016/j.eswa.2024.124278_b0445) 2022; 2022 Jiao (10.1016/j.eswa.2024.124278_b0170) 2020; 2020 Fang (10.1016/j.eswa.2024.124278_b0115) 2023; 2023 10.1016/j.eswa.2024.124278_b0430 10.1016/j.eswa.2024.124278_b0035 Bao (10.1016/j.eswa.2024.124278_b0030) 2021; 2021 Moosavi-Dezfooli (10.1016/j.eswa.2024.124278_b0310) 2016; 2016 10.1016/j.eswa.2024.124278_b0390 Liu (10.1016/j.eswa.2024.124278_b0250) 2022 10.1016/j.eswa.2024.124278_b0415 Oseledets (10.1016/j.eswa.2024.124278_b0330) 2018; 2018 Chiang (10.1016/j.eswa.2024.124278_b0075) 2023; 2023 10.1016/j.eswa.2024.124278_b0385 10.1016/j.eswa.2024.124278_b0140 Cer (10.1016/j.eswa.2024.124278_b0040) 2018 10.1016/j.eswa.2024.124278_b0265 10.1016/j.eswa.2024.124278_b0380 Yoo (10.1016/j.eswa.2024.124278_b0480) 2021; 2021 Pandey (10.1016/j.eswa.2024.124278_b0335) 2023; 269 Xu (10.1016/j.eswa.2024.124278_b0455) 2022 Modas (10.1016/j.eswa.2024.124278_b0295) 2019; 2019 Salimans (10.1016/j.eswa.2024.124278_b0400) 2016; 29 Metropolis (10.1016/j.eswa.2024.124278_b0290) 2004; 21 10.1016/j.eswa.2024.124278_b0405 Papernot (10.1016/j.eswa.2024.124278_b0345) 2016; 2016 Xu (10.1016/j.eswa.2024.124278_b0450) 2020; 17 Aggarwal (10.1016/j.eswa.2024.124278_b0010) 2023; 2023 10.1016/j.eswa.2024.124278_b0375 10.1016/j.eswa.2024.124278_b0255 10.1016/j.eswa.2024.124278_b0090 10.1016/j.eswa.2024.124278_b0095 10.1016/j.eswa.2024.124278_b0370 Morris (10.1016/j.eswa.2024.124278_b0315) 2020; 2020 Madhu (10.1016/j.eswa.2024.124278_b0275) 2023; 215 Saleh (10.1016/j.eswa.2024.124278_b0395) 2023; 37 Mehrish (10.1016/j.eswa.2024.124278_b0285) 2023; 99 Mamta (10.1016/j.eswa.2024.124278_b0005) 2022; 2022 10.1016/j.eswa.2024.124278_b0085 10.1016/j.eswa.2024.124278_b0360 Devlin (10.1016/j.eswa.2024.124278_b0100) 2019 10.1016/j.eswa.2024.124278_b0520 10.1016/j.eswa.2024.124278_b0125 10.1016/j.eswa.2024.124278_b0365 Kim (10.1016/j.eswa.2024.124278_b0190) 2014 Aggarwal (10.1016/j.eswa.2024.124278_b0015) 2023 Hastings (10.1016/j.eswa.2024.124278_b0150) 1970; 57 Chen (10.1016/j.eswa.2024.124278_b0055) 2016 10.1016/j.eswa.2024.124278_b0230 Zhao (10.1016/j.eswa.2024.124278_b0510) 2022 Chen (10.1016/j.eswa.2024.124278_b0060) 2021 10.1016/j.eswa.2024.124278_b0350 10.1016/j.eswa.2024.124278_b0355 Zhao (10.1016/j.eswa.2024.124278_b0505) 2022; 2022 Liu (10.1016/j.eswa.2024.124278_b0260) 2016 Chhabra (10.1016/j.eswa.2024.124278_b0070) 2023; 29 Lauriola (10.1016/j.eswa.2024.124278_b0210) 2022; 470 Yang (10.1016/j.eswa.2024.124278_b0465) 2023; 644 |
| References_xml | – reference: , 124–135. https://doi.org/10.18653/v1/2020.sustainlp-1.17. – reference: (pp. 3485–3495). The COLING 2016 Organizing Committee. https://aclanthology.org/C16-1329. – reference: Mondal, I. (2021). BBAEG: Towards BERT-based Biomedical Adversarial Example Generation for Text Classification. – volume: 2016 start-page: 372 year: 2016 end-page: 387 ident: b0345 article-title: The Limitations of Deep Learning in Adversarial Settings publication-title: IEEE European Symposium on Security and Privacy (EuroS&P) – start-page: 469 year: 2022 end-page: 470 ident: b0525 article-title: Adversarial example generation via genetic algorithm: A preliminary result publication-title: Proceedings of the Genetic and Evolutionary Computation Conference Companion – start-page: 786 year: 2020 end-page: 803 ident: b0110 article-title: From Hero to Zéroe: A Benchmark of Low-Level Adversarial Attacks publication-title: Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing – volume: 8 start-page: 64 year: 2020 end-page: 77 ident: b0180 article-title: SpanBERT: Improving Pre-training by Representing and Predicting Spans publication-title: Transactions of the Association for Computational Linguistics – start-page: 1746 year: 2014 end-page: 1751 ident: b0190 article-title: Convolutional Neural Networks for Sentence Classification publication-title: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) – reference: Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., & Sutskever, I. (n.d.). – volume: 6 start-page: 25 year: 2021 end-page: 45 ident: b0045 article-title: A survey on adversarial attacks and defences publication-title: CAAI Transactions on Intelligence Technology – volume: 232 year: 2023 ident: b0460 article-title: MRT-Net: Auto-adaptive weighting of manipulation residuals and texture clues for face manipulation detection publication-title: Expert Systems with Applications – volume: 29 start-page: 1203 year: 2023 end-page: 1230 ident: b0070 article-title: A literature survey on multimodal and multilingual automatic hate speech identification publication-title: Multimedia Systems – volume: 2018 start-page: 50 year: 2018 end-page: 56 ident: b0130 article-title: Black-Box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers publication-title: IEEE Security and Privacy Workshops (SPW) – reference: Pruthi, D., Dhingra, B., & Lipton, Z. C. (2019). Combating Adversarial Misspellings with Robust Word Recognition. – start-page: 169 year: 2018 end-page: 174 ident: b0040 article-title: Universal Sentence Encoder for English publication-title: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing: System Demonstrations – volume: 2023 start-page: 7322 year: 2023 end-page: 7336 ident: b0115 article-title: Modeling Adversarial Attack on Pre-trained Language Models as Sequential Decision Making publication-title: Findings of the Association for Computational Linguistics: ACL – reference: . Eighth International Conference on Learning Representations. https://iclr.cc/virtual_2020/poster_r1xMH1BtvB.html. – start-page: 1875 year: 2018 end-page: 1885 ident: b0165 article-title: Adversarial Example Generation with Syntactically Controlled Paraphrase Networks publication-title: Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers) – reference: Pennington, J., Socher, R., & Manning, C. (2014). GloVe: Global Vectors for Word Representation. In A. Moschitti, B. Pang, & W. Daelemans (Eds.), – reference: Davidson, T., Warmsley, D., Macy, M., & Weber, I. (2017). Automated Hate Speech Detection and the Problem of Offensive Language. – year: 2018 ident: b0280 article-title: Towards Deep Learning Models Resistant to Adversarial Attacks publication-title: International Conference on Learning Representations – year: 2023 ident: b0025 article-title: A state-of-the-art review on adversarial machine learning in image classification publication-title: Multimedia Tools and Applications – reference: , 506–519. https://doi.org/10.1145/3052973.3053009. – reference: . 39–57. https://doi.org/10.1109/SP.2017.49. – volume: 57 start-page: 97 year: 1970 end-page: 109 ident: b0150 article-title: Monte Carlo sampling methods using Markov chains and their applications publication-title: Biometrika – reference: . https://doi.org/10.48550/arXiv.1511.06434. – volume: 32 year: 2019 ident: b0470 article-title: XLNet: Generalized Autoregressive Pretraining for Language Understanding publication-title: Advances in Neural Information Processing Systems – reference: Gaiński, P., & Ba\lazy, K. (2023). Step by Step Loss Goes Very Far: Multi-Step Quantization for Adversarial Text Attacks. – reference: Qi, F., Yang, C., Liu, Z., Dong, Q., Sun, M., & Dong, Z. (2019). – start-page: 4171 year: 2019 end-page: 4186 ident: b0100 article-title: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding publication-title: Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) – reference: , 6134–6150. https://doi.org/10.18653/v1/2020.emnlp-main.495. – start-page: 5564 year: 2019 end-page: 5569 ident: b0500 article-title: Generating Fluent Adversarial Examples for Natural Languages publication-title: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics – reference: (1), Article 1. https://doi.org/10.1609/icwsm.v11i1.14955. – reference: , 276–286. https://doi.org/10.18653/v1/2023.trustnlp-1.24. – volume: 244 year: 2024 ident: b0145 article-title: HumanPoseNet: An all-transformer architecture for pose estimation with efficient patch expansion and attentional feature refinement publication-title: Expert Systems with Applications – start-page: 2180 year: 2016 end-page: 2188 ident: b0055 article-title: InfoGAN: Interpretable representation learning by information maximizing generative adversarial nets publication-title: Proceedings of the 30th International Conference on Neural Information Processing Systems – reference: Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2020). – volume: 216 year: 2023 ident: b0420 article-title: SocialHaterBERT: A dichotomous approach for automatically detecting hate speech on Twitter through textual analysis and user profiles publication-title: Expert Systems with Applications – volume: 2021 start-page: 2705 year: 2021 end-page: 2712 ident: b0195 article-title: Adversarial Examples for Evaluating Math Word Problem Solvers publication-title: Findings of the Association for Computational Linguistics: EMNLP – reference: (arXiv:1901.09957). arXiv. https://doi.org/10.48550/arXiv.1901.09957. – volume: 2021 start-page: 39 year: 2021 end-page: 40 ident: b0475 article-title: Heuristic-word-selection Genetic Algorithm for Generating Natural Language Adversarial Examples publication-title: IEEE International Conference on Artificial Intelligence Testing (AITest) – reference: Kurakin, A., Goodfellow, I. J., & Bengio, S. (2018). Adversarial Examples in the Physical World. In R. V. Yampolskiy (Ed.), – reference: . https://www.semanticscholar.org/paper/RoBERTa%3A-A-Robustly-Optimized-BERT-Pretraining-Liu-Ott/077f8329a7b6fa3b7c877a57b81eb6c18b5f87de. – volume: 2021 start-page: 945 year: 2021 end-page: 956 ident: b0480 article-title: Towards Improving Adversarial Training of NLP Models publication-title: Findings of the Association for Computational Linguistics: EMNLP – reference: Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., & Stoyanov, V. (2019). RoBERTa: A Robustly Optimized BERT Pretraining Approach. – volume: 2022 start-page: 176 year: 2022 end-page: 205 ident: b0445 article-title: SemAttack: Natural Textual Attacks via Different Semantic Spaces publication-title: Findings of the Association for Computational Linguistics: NAACL – reference: , – reference: (arXiv:1910.01108; Version 4). arXiv. http://arxiv.org/abs/1910.01108. – start-page: 5053 year: 2021 end-page: 5069 ident: b0225 article-title: Contextualized Perturbation for Textual Adversarial Attack publication-title: Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies – reference: (arXiv:1412.6572). arXiv. https://doi.org/10.48550/arXiv.1412.6572. – volume: 17 start-page: 151 year: 2020 end-page: 178 ident: b0450 article-title: Adversarial Attacks and Defenses in Images, Graphs and Text: A Review publication-title: International Journal of Automation and Computing – reference: Pavlopoulos, J., Thain, N., Dixon, L., & Androutsopoulos, I. (2019). ConvAI at SemEval-2019 Task 6: Offensive Language Identification and Categorization with Perspective and BERT. In J. May, E. Shutova, A. Herbelot, X. Zhu, M. Apidianaki, & S. M. Mohammad (Eds.), – reference: Liu, H., Cai, C., & Qi, Y. (2023). Expanding Scope: Adapting English Adversarial Attacks to Chinese. – start-page: 2849 year: 2022 end-page: 2860 ident: b0455 article-title: Student Surpasses Teacher: Imitation Attack for Black-Box NLP APIs publication-title: Proceedings of the 29th International Conference on Computational Linguistics – volume: 2018 start-page: 8562 year: 2018 end-page: 8570 ident: b0330 article-title: Art of Singular Vectors and Universal Adversarial Perturbations publication-title: IEEE/CVF Conference on Computer Vision and Pattern Recognition – volume: 558 year: 2023 ident: b0020 article-title: Evading text based emotion detection mechanism via adversarial attacks publication-title: Neurocomputing – year: 2020 ident: b0530 article-title: FreeLB: Enhanced Adversarial Training for Natural Language Understanding publication-title: Eighth International Conference on Learning Representations – reference: , 119–126. https://doi.org/10.18653/v1/2020.emnlp-demos.16. – reference: Tsai, Y.-T., Yang, M.-C., & Chen, H.-Y. (2019). Adversarial Attack on Sentiment Classification. – volume: 2022 start-page: 5009 year: 2022 end-page: 5018 ident: b0155 article-title: Invernet: An Inversion Attack Framework to Infer Fine-Tuning Datasets through Word Embeddings publication-title: Findings of the Association for Computational Linguistics: EMNLP – start-page: 7664 year: 2022 end-page: 7676 ident: b0250 article-title: Character-level White-Box Adversarial Attacks against Transformers via Attachable Subwords Substitution publication-title: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing – reference: Ren, S., Deng, Y., He, K., & Che, W. (2019). Generating Natural Language Adversarial Examples through Probability Weighted Word Saliency. – reference: (arXiv:1909.11942; Version 6). arXiv. http://arxiv.org/abs/1909.11942. – reference: Kennedy, J., & Eberhart, R. (1995). Particle swarm optimization. – reference: Clark, K., Luong, M.-T., Le, Q. V., & Manning, C. D. (2020, April). – start-page: 4511 year: 2021 end-page: 4526 ident: b0060 article-title: Multi-granularity Textual Adversarial Attack with Behavior Cloning publication-title: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing – reference: Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., & Soricut, R. (2020). – volume: 30 year: 2017 ident: b0425 article-title: Attention is All you Need publication-title: Advances in Neural Information Processing Systems – volume: 185 year: 2021 ident: b0410 article-title: Deep Learning for predicting neutralities in Offensive Language Identification Dataset publication-title: Expert Systems with Applications – reference: Wan, J., Yang, J., Ma, S., Zhang, D., Zhang, W., Yu, Y., & Li, Z. (2022). PAEG: Phrase-level Adversarial Example Generation for Neural Machine Translation. – reference: Deng, C., Liu, M., Qin, Y., Zhang, J., Duan, H.-X., & Sun, D. (2022). ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model. – reference: , 233–240. https://doi.org/10.18653/v1/W19-4824. – volume: 2016 start-page: 2574 year: 2016 end-page: 2582 ident: b0310 article-title: DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks publication-title: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – reference: Radford, A., Metz, L., & Chintala, S. (2015). Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks. In – volume: 2020 start-page: 3829 year: 2020 end-page: 3839 ident: b0315 article-title: Reevaluating Adversarial Examples in Natural Language publication-title: Findings of the Association for Computational Linguistics: EMNLP – reference: . (n.d.). Retrieved 26 December 2023, from https://perspectiveapi.com/. – volume: 2023 start-page: 7132 year: 2023 end-page: 7146 ident: b0485 article-title: Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Framework publication-title: Findings of the Association for Computational Linguistics: ACL – volume: 470 start-page: 443 year: 2022 end-page: 456 ident: b0210 article-title: An introduction to Deep Learning in Natural Language Processing: Models, techniques, and tools publication-title: Neurocomputing – start-page: 3197 year: 2022 end-page: 3207 ident: b0215 article-title: A New Generation of Perspective API: Efficient Multilingual Character-level Transformers publication-title: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining – reference: Wang, B., Pei, H., Pan, B., Chen, Q., Wang, S., & Li, B. (2020). T3: Tree-Autoencoder Constrained Adversarial Text Generation for Targeted Attack. – start-page: 6174 year: 2020 end-page: 6181 ident: b0135 article-title: BAE: BERT-based Adversarial Examples for Text Classification publication-title: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) – volume: 2022 start-page: 4599 year: 2022 end-page: 4606 ident: b0505 article-title: Generating Textual Adversaries with Minimal Perturbation publication-title: Findings of the Association for Computational Linguistics: EMNLP – reference: Morris, J., Lifland, E., Yoo, J. Y., Grigsby, J., Jin, D., & Qi, Y. (2020). TextAttack: A Framework for Adversarial Attacks, Data Augmentation, and Adversarial Training in NLP. – reference: , 5085–5097. https://aclanthology.org/2022.coling-1.451. – volume: 2022 start-page: 1095 year: 2022 end-page: 1112 ident: b0220 article-title: Phrase-level Textual Adversarial Attack with Label Preservation publication-title: Findings of the Association for Computational Linguistics: NAACL – start-page: 6193 year: 2020 end-page: 6202 ident: b0235 article-title: BERT-ATTACK: Adversarial Attack Against BERT Using BERT publication-title: Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) – volume: 2021 start-page: 3248 year: 2021 end-page: 3258 ident: b0030 article-title: Defending Pre-trained Language Models from Adversarial Word Substitution Without Performance Sacrifice publication-title: Findings of the Association for Computational Linguistics: ACL-IJCNLP – volume: 529 start-page: 190 year: 2023 end-page: 203 ident: b0050 article-title: TextGuise: Adaptive adversarial example attacks on text classification model publication-title: Neurocomputing – start-page: 31 year: 2018 end-page: 36 ident: b0105 article-title: HotFlip: White-Box Adversarial Examples for Text Classification publication-title: Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) – volume: 2022 start-page: 478 year: 2022 end-page: 492 ident: b0005 article-title: Adversarial Sample Generation for Aspect based Sentiment Classification publication-title: Findings of the Association for Computational Linguistics: AACL-IJCNLP – start-page: 1 year: 2023 end-page: 5 ident: b0015 article-title: Protecting our Children from the Dark Corners of YouTube: A Cutting-Edge Analysis publication-title: 2023 4th IEEE Global Conference for Advancement in Technology (GCAT) – reference: (pp. 571–576). Association for Computational Linguistics. https://doi.org/10.18653/v1/S19-2102. – volume: 215 year: 2023 ident: b0275 article-title: Detecting offensive speech in conversational code-mixed dialogue on social media: A contextual dataset and benchmark experiments publication-title: Expert Systems with Applications – reference: , 2038–2048. https://doi.org/10.18653/v1/2023.eacl-main.149. – reference: (pp. 1532–1543). Association for Computational Linguistics. https://doi.org/10.3115/v1/D14-1162. – start-page: 6066 year: 2020 end-page: 6080 ident: b0490 article-title: Word-level Textual Adversarial Attacking as Combinatorial Optimization publication-title: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics – reference: (1st ed., pp. 99–112). Chapman and Hall/CRC. https://doi.org/10.1201/9781351251389-8. – volume: 2023 start-page: 1853 year: 2023 end-page: 1878 ident: b0075 article-title: Are Synonym Substitution Attacks Really Synonym Substitution Attacks? publication-title: Findings of the Association for Computational Linguistics: ACL – year: 2020 ident: b0175 article-title: Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment publication-title: Proceedings of the AAAI Conference on Artificial Intelligence – reference: Qi, F., Chen, Y., Zhang, X., Li, M., Liu, Z., & Sun, M. (2021). Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer. – start-page: 4324 year: 2019 end-page: 4333 ident: b0065 publication-title: Robust Neural Machine Translation with Doubly Adversarial Inputs – volume: 3 start-page: 111 year: 2022 end-page: 132 ident: b0245 article-title: A survey of transformers publication-title: AI Open – start-page: 932 year: 2022 end-page: 944 ident: b0510 article-title: Can we Really Trust Explanations? Evaluating the Stability of Feature Attribution Explanation Methods via Adversarial Attack publication-title: Proceedings of the 21st Chinese National Conference on Computational Linguistics – volume: 2023 start-page: 1 year: 2023 end-page: 5 ident: b0010 article-title: Multimodal Sarcasm Recognition by Fusing Textual, Visual and Acoustic content via Multi-Headed Attention for Video Dataset publication-title: World Conference on Communication & Computing (WCONF) – reference: Verwimp, L., Pelemans, J., Van hamme, H., & Wambacq, P. (2017). Character-Word LSTM Language Models. In M. Lapata, P. Blunsom, & A. Koller (Eds.), – volume: 2019 start-page: 9079 year: 2019 end-page: 9088 ident: b0295 article-title: SparseFool: A Few Pixels Make a Big Difference publication-title: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) – reference: (pp. 417–427). Association for Computational Linguistics. https://aclanthology.org/E17-1040. – start-page: 5490 year: 2022 end-page: 5498 ident: b0080 article-title: TABS: Efficient Textual Adversarial Attack for Pre-trained NL Code Model Using Semantic Beam Search publication-title: Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing – reference: , 5582–5591. https://doi.org/10.18653/v1/P19-1561. – reference: , 4569–4580. https://doi.org/10.18653/v1/2021.emnlp-main.374. – reference: Goodfellow, I. J., Shlens, J., & Szegedy, C. (2015). – volume: 37 start-page: 2166719 year: 2023 ident: b0395 article-title: Detection of Hate Speech using BERT and Hate Speech Word Embedding with Deep Model publication-title: Applied Artificial Intelligence – volume: 2022 start-page: 2903 year: 2022 end-page: 2913 ident: b0325 article-title: CheckHARD: Checking Hard Labels for Adversarial Text Detection, Prediction Correction, and Perturbed Word Suggestion publication-title: Findings of the Association for Computational Linguistics: EMNLP – reference: Papernot, N., McDaniel, P., Goodfellow, I., Jha, S., Celik, Z. B., & Swami, A. (2017). Practical Black-Box Attacks against Machine Learning. – reference: Carlini, N., & Wagner, D. (2017). – volume: 238 year: 2024 ident: b0270 article-title: Adversarial examples: A survey of attacks and defenses in deep learning-enabled cybersecurity systems publication-title: Expert Systems with Applications – reference: , 1942–1948 vol.4. https://doi.org/10.1109/ICNN.1995.488968. – volume: 21 start-page: 1087 year: 2004 end-page: 1092 ident: b0290 article-title: Equation of State Calculations by Fast Computing Machines publication-title: The Journal of Chemical Physics – start-page: 3728 year: 2021 end-page: 3737 ident: b0240 article-title: RockNER: A Simple Method to Create Adversarial Examples for Evaluating the Robustness of Named Entity Recognition Models publication-title: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing – reference: , 1735–1746. https://doi.org/10.18653/v1/2022.naacl-main.125. – volume: 644 year: 2023 ident: b0465 article-title: Generation-based parallel particle swarm optimization for adversarial text attacks publication-title: Information Sciences – volume: 2020 start-page: 4163 year: 2020 end-page: 4174 ident: b0170 article-title: TinyBERT: Distilling BERT for Natural Language Understanding publication-title: Findings of the Association for Computational Linguistics: EMNLP – reference: , 5378–5384. https://doi.org/10.18653/v1/2021.naacl-main.423. – reference: Zhou, P., Qi, Z., Zheng, S., Xu, J., Bao, H., & Xu, B. (2016). Text Classification Improved by Integrating Bidirectional LSTM with Two-dimensional Max Pooling. In Y. Matsumoto & R. Prasad (Eds.), – reference: . Network and Distributed System Security Symposium, San Diego, CA. https://doi.org/10.14722/ndss.2019.23138. – start-page: 9960 year: 2023 end-page: 9976 ident: b0515 article-title: NatLogAttack: A Framework for Attacking Natural Language Inference Models with Natural Logic publication-title: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) – volume: 2023 start-page: 7891 year: 2023 end-page: 7906 ident: b0495 article-title: Similarizing the Influence of Words with Contrastive Learning to Defend Word-level Adversarial Text Attack publication-title: Findings of the Association for Computational Linguistics: ACL – reference: . – reference: Iandola, F., Shaw, A., Krishna, R., & Keutzer, K. (2020). SqueezeBERT: What can computer vision teach NLP about efficient neural networks? – reference: Li, J., Ji, S., Du, T., Li, B., & Wang, T. (2019). TextBugger: Generating Adversarial Text Against Real-world Applications. – year: 2016 ident: b0260 article-title: Delving into Transferable Adversarial Examples and Black-box Attacks publication-title: International Conference on Learning Representations – volume: 8 start-page: 4663 year: 2022 end-page: 4678 ident: b0300 article-title: ETHOS: A multi-label hate speech detection dataset publication-title: Complex & Intelligent Systems – reference: , 1085–1097. https://doi.org/10.18653/v1/P19-1103. – volume: 269 year: 2023 ident: b0335 article-title: VABDC-Net: A framework for Visual-Caption Sentiment Recognition via spatio-depth visual attention and bi-directional caption processing publication-title: Knowledge-Based Systems – volume: 29 year: 2016 ident: b0400 article-title: Improved Techniques for Training GANs publication-title: Advances in Neural Information Processing Systems – volume: 2023 start-page: 1 year: 2023 end-page: 34 ident: b0120 article-title: Using Punctuation as an Adversarial Attack on Deep Learning-Based NLP Systems: An Empirical Study publication-title: Findings of the Association for Computational Linguistics: EACL – volume: 99 year: 2023 ident: b0285 article-title: A review of deep learning techniques for speech processing publication-title: Information Fusion – volume: 232 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0460 article-title: MRT-Net: Auto-adaptive weighting of manipulation residuals and texture clues for face manipulation detection publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2023.120898 – start-page: 9960 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0515 article-title: NatLogAttack: A Framework for Attacking Natural Language Inference Models with Natural Logic – ident: 10.1016/j.eswa.2024.124278_b0125 doi: 10.18653/v1/2023.eacl-main.149 – ident: 10.1016/j.eswa.2024.124278_b0230 doi: 10.14722/ndss.2019.23138 – start-page: 1 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0015 article-title: Protecting our Children from the Dark Corners of YouTube: A Cutting-Edge Analysis – ident: 10.1016/j.eswa.2024.124278_b0265 – volume: 99 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0285 article-title: A review of deep learning techniques for speech processing publication-title: Information Fusion doi: 10.1016/j.inffus.2023.101869 – start-page: 7664 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0250 article-title: Character-level White-Box Adversarial Attacks against Transformers via Attachable Subwords Substitution – volume: 2021 start-page: 39 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0475 article-title: Heuristic-word-selection Genetic Algorithm for Generating Natural Language Adversarial Examples publication-title: IEEE International Conference on Artificial Intelligence Testing (AITest) – ident: 10.1016/j.eswa.2024.124278_b0355 doi: 10.3115/v1/D14-1162 – volume: 2022 start-page: 176 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0445 article-title: SemAttack: Natural Textual Attacks via Different Semantic Spaces publication-title: Findings of the Association for Computational Linguistics: NAACL – start-page: 786 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0110 article-title: From Hero to Zéroe: A Benchmark of Low-Level Adversarial Attacks – start-page: 469 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0525 article-title: Adversarial example generation via genetic algorithm: A preliminary result – ident: 10.1016/j.eswa.2024.124278_b0085 – ident: 10.1016/j.eswa.2024.124278_b0360 – start-page: 2849 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0455 article-title: Student Surpasses Teacher: Imitation Attack for Black-Box NLP APIs – volume: 6 start-page: 25 issue: 1 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0045 article-title: A survey on adversarial attacks and defences publication-title: CAAI Transactions on Intelligence Technology doi: 10.1049/cit2.12028 – volume: 2020 start-page: 4163 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0170 article-title: TinyBERT: Distilling BERT for Natural Language Understanding publication-title: Findings of the Association for Computational Linguistics: EMNLP – ident: 10.1016/j.eswa.2024.124278_b0405 – ident: 10.1016/j.eswa.2024.124278_b0090 doi: 10.1609/icwsm.v11i1.14955 – ident: 10.1016/j.eswa.2024.124278_b0430 doi: 10.18653/v1/E17-1040 – volume: 2023 start-page: 7322 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0115 article-title: Modeling Adversarial Attack on Pre-trained Language Models as Sequential Decision Making publication-title: Findings of the Association for Computational Linguistics: ACL – volume: 2022 start-page: 478 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0005 article-title: Adversarial Sample Generation for Aspect based Sentiment Classification publication-title: Findings of the Association for Computational Linguistics: AACL-IJCNLP – volume: 644 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0465 article-title: Generation-based parallel particle swarm optimization for adversarial text attacks publication-title: Information Sciences doi: 10.1016/j.ins.2023.119237 – volume: 2021 start-page: 3248 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0030 article-title: Defending Pre-trained Language Models from Adversarial Word Substitution Without Performance Sacrifice publication-title: Findings of the Association for Computational Linguistics: ACL-IJCNLP – year: 2023 ident: 10.1016/j.eswa.2024.124278_b0025 article-title: A state-of-the-art review on adversarial machine learning in image classification publication-title: Multimedia Tools and Applications – volume: 8 start-page: 64 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0180 article-title: SpanBERT: Improving Pre-training by Representing and Predicting Spans publication-title: Transactions of the Association for Computational Linguistics doi: 10.1162/tacl_a_00300 – ident: 10.1016/j.eswa.2024.124278_b0520 – volume: 2023 start-page: 7891 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0495 article-title: Similarizing the Influence of Words with Contrastive Learning to Defend Word-level Adversarial Text Attack publication-title: Findings of the Association for Computational Linguistics: ACL – volume: 238 year: 2024 ident: 10.1016/j.eswa.2024.124278_b0270 article-title: Adversarial examples: A survey of attacks and defenses in deep learning-enabled cybersecurity systems publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2023.122223 – volume: 470 start-page: 443 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0210 article-title: An introduction to Deep Learning in Natural Language Processing: Models, techniques, and tools publication-title: Neurocomputing doi: 10.1016/j.neucom.2021.05.103 – ident: 10.1016/j.eswa.2024.124278_b0035 doi: 10.1109/SP.2017.49 – ident: 10.1016/j.eswa.2024.124278_b0200 doi: 10.1201/9781351251389-8 – volume: 3 start-page: 111 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0245 article-title: A survey of transformers publication-title: AI Open doi: 10.1016/j.aiopen.2022.10.001 – ident: 10.1016/j.eswa.2024.124278_b0390 doi: 10.18653/v1/P19-1103 – volume: 558 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0020 article-title: Evading text based emotion detection mechanism via adversarial attacks publication-title: Neurocomputing doi: 10.1016/j.neucom.2023.126787 – volume: 216 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0420 article-title: SocialHaterBERT: A dichotomous approach for automatically detecting hate speech on Twitter through textual analysis and user profiles publication-title: Expert Systems with Applications – volume: 2016 start-page: 372 year: 2016 ident: 10.1016/j.eswa.2024.124278_b0345 article-title: The Limitations of Deep Learning in Adversarial Settings publication-title: IEEE European Symposium on Security and Privacy (EuroS&P) doi: 10.1109/EuroSP.2016.36 – volume: 2022 start-page: 1095 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0220 article-title: Phrase-level Textual Adversarial Attack with Label Preservation publication-title: Findings of the Association for Computational Linguistics: NAACL – ident: 10.1016/j.eswa.2024.124278_b0160 doi: 10.18653/v1/2020.sustainlp-1.17 – start-page: 6066 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0490 article-title: Word-level Textual Adversarial Attacking as Combinatorial Optimization – ident: 10.1016/j.eswa.2024.124278_b0095 doi: 10.18653/v1/2022.naacl-main.125 – volume: 2018 start-page: 8562 year: 2018 ident: 10.1016/j.eswa.2024.124278_b0330 article-title: Art of Singular Vectors and Universal Adversarial Perturbations publication-title: IEEE/CVF Conference on Computer Vision and Pattern Recognition – start-page: 4324 year: 2019 ident: 10.1016/j.eswa.2024.124278_b0065 – volume: 185 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0410 article-title: Deep Learning for predicting neutralities in Offensive Language Identification Dataset publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2021.115458 – volume: 29 start-page: 1203 issue: 3 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0070 article-title: A literature survey on multimodal and multilingual automatic hate speech identification publication-title: Multimedia Systems doi: 10.1007/s00530-023-01051-8 – ident: 10.1016/j.eswa.2024.124278_b0255 doi: 10.18653/v1/2023.trustnlp-1.24 – volume: 21 start-page: 1087 issue: 6 year: 2004 ident: 10.1016/j.eswa.2024.124278_b0290 article-title: Equation of State Calculations by Fast Computing Machines publication-title: The Journal of Chemical Physics doi: 10.1063/1.1699114 – start-page: 169 year: 2018 ident: 10.1016/j.eswa.2024.124278_b0040 article-title: Universal Sentence Encoder for English – start-page: 6193 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0235 article-title: BERT-ATTACK: Adversarial Attack Against BERT Using BERT – volume: 529 start-page: 190 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0050 article-title: TextGuise: Adaptive adversarial example attacks on text classification model publication-title: Neurocomputing doi: 10.1016/j.neucom.2023.01.071 – start-page: 4511 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0060 article-title: Multi-granularity Textual Adversarial Attack with Behavior Cloning – volume: 17 start-page: 151 issue: 2 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0450 article-title: Adversarial Attacks and Defenses in Images, Graphs and Text: A Review publication-title: International Journal of Automation and Computing doi: 10.1007/s11633-019-1211-x – volume: 2022 start-page: 5009 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0155 article-title: Invernet: An Inversion Attack Framework to Infer Fine-Tuning Datasets through Word Embeddings publication-title: Findings of the Association for Computational Linguistics: EMNLP – volume: 2019 start-page: 9079 year: 2019 ident: 10.1016/j.eswa.2024.124278_b0295 article-title: SparseFool: A Few Pixels Make a Big Difference publication-title: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) – ident: 10.1016/j.eswa.2024.124278_b0415 doi: 10.18653/v1/W19-4824 – start-page: 5564 year: 2019 ident: 10.1016/j.eswa.2024.124278_b0500 article-title: Generating Fluent Adversarial Examples for Natural Languages – start-page: 3197 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0215 article-title: A New Generation of Perspective API: Efficient Multilingual Character-level Transformers – ident: 10.1016/j.eswa.2024.124278_b0370 doi: 10.18653/v1/2021.emnlp-main.374 – ident: 10.1016/j.eswa.2024.124278_b0320 doi: 10.18653/v1/2020.emnlp-demos.16 – start-page: 3728 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0240 article-title: RockNER: A Simple Method to Create Adversarial Examples for Evaluating the Robustness of Named Entity Recognition Models – year: 2020 ident: 10.1016/j.eswa.2024.124278_b0175 article-title: Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment – start-page: 5490 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0080 article-title: TABS: Efficient Textual Adversarial Attack for Pre-trained NL Code Model Using Semantic Beam Search – volume: 57 start-page: 97 issue: 1 year: 1970 ident: 10.1016/j.eswa.2024.124278_b0150 article-title: Monte Carlo sampling methods using Markov chains and their applications publication-title: Biometrika doi: 10.1093/biomet/57.1.97 – volume: 2021 start-page: 945 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0480 article-title: Towards Improving Adversarial Training of NLP Models publication-title: Findings of the Association for Computational Linguistics: EMNLP – ident: 10.1016/j.eswa.2024.124278_b0185 doi: 10.1109/ICNN.1995.488968 – volume: 269 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0335 article-title: VABDC-Net: A framework for Visual-Caption Sentiment Recognition via spatio-depth visual attention and bi-directional caption processing publication-title: Knowledge-Based Systems doi: 10.1016/j.knosys.2023.110515 – ident: 10.1016/j.eswa.2024.124278_b0435 – year: 2018 ident: 10.1016/j.eswa.2024.124278_b0280 article-title: Towards Deep Learning Models Resistant to Adversarial Attacks – ident: 10.1016/j.eswa.2024.124278_b0365 doi: 10.18653/v1/P19-1561 – volume: 2020 start-page: 3829 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0315 article-title: Reevaluating Adversarial Examples in Natural Language publication-title: Findings of the Association for Computational Linguistics: EMNLP – ident: 10.1016/j.eswa.2024.124278_b0375 – ident: 10.1016/j.eswa.2024.124278_b0440 doi: 10.18653/v1/2020.emnlp-main.495 – year: 2020 ident: 10.1016/j.eswa.2024.124278_b0530 article-title: FreeLB: Enhanced Adversarial Training for Natural Language Understanding – volume: 2018 start-page: 50 year: 2018 ident: 10.1016/j.eswa.2024.124278_b0130 article-title: Black-Box Generation of Adversarial Text Sequences to Evade Deep Learning Classifiers publication-title: IEEE Security and Privacy Workshops (SPW) – volume: 215 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0275 article-title: Detecting offensive speech in conversational code-mixed dialogue on social media: A contextual dataset and benchmark experiments publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2022.119342 – start-page: 4171 year: 2019 ident: 10.1016/j.eswa.2024.124278_b0100 article-title: BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding – ident: 10.1016/j.eswa.2024.124278_b0340 doi: 10.1145/3052973.3053009 – volume: 2023 start-page: 1853 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0075 article-title: Are Synonym Substitution Attacks Really Synonym Substitution Attacks? publication-title: Findings of the Association for Computational Linguistics: ACL – start-page: 2180 year: 2016 ident: 10.1016/j.eswa.2024.124278_b0055 article-title: InfoGAN: Interpretable representation learning by information maximizing generative adversarial nets – volume: 8 start-page: 4663 issue: 6 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0300 article-title: ETHOS: A multi-label hate speech detection dataset publication-title: Complex & Intelligent Systems doi: 10.1007/s40747-021-00608-2 – volume: 2023 start-page: 1 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0010 article-title: Multimodal Sarcasm Recognition by Fusing Textual, Visual and Acoustic content via Multi-Headed Attention for Video Dataset publication-title: World Conference on Communication & Computing (WCONF) – volume: 2023 start-page: 1 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0120 article-title: Using Punctuation as an Adversarial Attack on Deep Learning-Based NLP Systems: An Empirical Study publication-title: Findings of the Association for Computational Linguistics: EACL – ident: 10.1016/j.eswa.2024.124278_b0205 – start-page: 31 year: 2018 ident: 10.1016/j.eswa.2024.124278_b0105 article-title: HotFlip: White-Box Adversarial Examples for Text Classification – volume: 2022 start-page: 2903 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0325 article-title: CheckHARD: Checking Hard Labels for Adversarial Text Detection, Prediction Correction, and Perturbed Word Suggestion publication-title: Findings of the Association for Computational Linguistics: EMNLP – start-page: 6174 year: 2020 ident: 10.1016/j.eswa.2024.124278_b0135 article-title: BAE: BERT-based Adversarial Examples for Text Classification – volume: 29 year: 2016 ident: 10.1016/j.eswa.2024.124278_b0400 article-title: Improved Techniques for Training GANs publication-title: Advances in Neural Information Processing Systems – ident: 10.1016/j.eswa.2024.124278_b0140 – ident: 10.1016/j.eswa.2024.124278_b0385 – volume: 2021 start-page: 2705 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0195 article-title: Adversarial Examples for Evaluating Math Word Problem Solvers publication-title: Findings of the Association for Computational Linguistics: EMNLP – volume: 2023 start-page: 7132 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0485 article-title: Bridge the Gap Between CV and NLP! A Gradient-based Textual Adversarial Attack Framework publication-title: Findings of the Association for Computational Linguistics: ACL – volume: 244 year: 2024 ident: 10.1016/j.eswa.2024.124278_b0145 article-title: HumanPoseNet: An all-transformer architecture for pose estimation with efficient patch expansion and attentional feature refinement publication-title: Expert Systems with Applications doi: 10.1016/j.eswa.2023.122894 – start-page: 1746 year: 2014 ident: 10.1016/j.eswa.2024.124278_b0190 article-title: Convolutional Neural Networks for Sentence Classification – volume: 2016 start-page: 2574 year: 2016 ident: 10.1016/j.eswa.2024.124278_b0310 article-title: DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks publication-title: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) – volume: 37 start-page: 2166719 issue: 1 year: 2023 ident: 10.1016/j.eswa.2024.124278_b0395 article-title: Detection of Hate Speech using BERT and Hate Speech Word Embedding with Deep Model publication-title: Applied Artificial Intelligence doi: 10.1080/08839514.2023.2166719 – ident: 10.1016/j.eswa.2024.124278_b0350 doi: 10.18653/v1/S19-2102 – volume: 32 year: 2019 ident: 10.1016/j.eswa.2024.124278_b0470 article-title: XLNet: Generalized Autoregressive Pretraining for Language Understanding publication-title: Advances in Neural Information Processing Systems – volume: 2022 start-page: 4599 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0505 article-title: Generating Textual Adversaries with Minimal Perturbation publication-title: Findings of the Association for Computational Linguistics: EMNLP – year: 2016 ident: 10.1016/j.eswa.2024.124278_b0260 article-title: Delving into Transferable Adversarial Examples and Black-box Attacks – ident: 10.1016/j.eswa.2024.124278_b0380 – start-page: 932 year: 2022 ident: 10.1016/j.eswa.2024.124278_b0510 article-title: Can we Really Trust Explanations? Evaluating the Stability of Feature Attribution Explanation Methods via Adversarial Attack – volume: 30 year: 2017 ident: 10.1016/j.eswa.2024.124278_b0425 article-title: Attention is All you Need publication-title: Advances in Neural Information Processing Systems – start-page: 1875 year: 2018 ident: 10.1016/j.eswa.2024.124278_b0165 article-title: Adversarial Example Generation with Syntactically Controlled Paraphrase Networks – ident: 10.1016/j.eswa.2024.124278_b0305 doi: 10.18653/v1/2021.naacl-main.423 – start-page: 5053 year: 2021 ident: 10.1016/j.eswa.2024.124278_b0225 article-title: Contextualized Perturbation for Textual Adversarial Attack |
| SSID | ssj0017007 |
| Score | 2.5080295 |
| Snippet | The accessibility of online hate speech has increased significantly, making it crucial for social-media companies to prioritize efforts to curb its spread.... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 124278 |
| SubjectTerms | Adversarial attack Glyphs Hate speech Natural Language Processing (NLP) Offensive language Transformers |
| Title | Exposing the Achilles’ heel of textual hate speech classifiers using indistinguishable adversarial examples |
| URI | https://dx.doi.org/10.1016/j.eswa.2024.124278 |
| Volume | 254 |
| WOSCitedRecordID | wos001263673200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 issn: 0957-4174 databaseCode: AIEXJ dateStart: 19950101 customDbUrl: isFulltext: true dateEnd: 99991231 titleUrlDefault: https://www.sciencedirect.com omitProxy: false ssIdentifier: ssj0017007 providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Pb9MwFLbKxoEL46cYDOQDtyhT4iQ4PlaoEyA0ITFQb5Hj2O26kVVNu_bIv8G_x1-y92I7yzaY2IGLVaWJE_V9ev1iv-99hLxNuMlVEsswYZyFqdBpKIRRYcaFfhcrpkXVCoU_88PDfDwWXwaDtdfCnJ_yus43GzH_r6GGYxBslM7eIdzdpHAAPkPQYYSww_hPgR9tsBDLiaCGKNbGujdX0yCAGeqWfmLFB0pHpkA2g2autZoGCqn0sUF37GDVWLVLhUmgnqzQpRlVVhIdnBvZmn3ojcTmws2V5X3snbx0HaK9dq63S94hbDKRi3VrNxB8lTPZVXp8hzut5QkutbcJEevyp0FbCt5fomApavWsSLNba-RhGls7Hp92WZb2EifQDGa9fG7kdLu8MNvXzRobRbF0__Lkqw20r_2xdeWGvpJtVuAcBc5R2DnukW3GMwHpcHv4cTT-1G1A8cgq7f2TO72VLQ28_iR_5jQ9nnL0iDx0Lxh0aIHxmAx0_YTsePMO6nL5U_LD44QCTqjHye-fvygihJ4Z6hBCESHUIoT2EEJbhNAbCKE9hFCPkGfk28Ho6P2H0HlvhCqJomXIpBF5VPJYM4k0N4HEbpiqEgGcNlaqjHIZ69Jw-MLkuFecRWVaVVmmTVWyKnlOtuqzWr8gFN4QsKWQMhGMpsplAhQ11irPk9YBb5fE_tcrlGtMj_4op8Xf47ZLgu6auW3LcuvZmQ9K4YilJYwFYOyW617e6S6vyINL8O-RreVipV-T--p8edws3jiAXQBg8p44 |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Exposing+the+Achilles%E2%80%99+heel+of+textual+hate+speech+classifiers+using+indistinguishable+adversarial+examples&rft.jtitle=Expert+systems+with+applications&rft.au=Aggarwal%2C+Sajal&rft.au=Vishwakarma%2C+Dinesh+Kumar&rft.date=2024-11-15&rft.issn=0957-4174&rft.volume=254&rft.spage=124278&rft_id=info:doi/10.1016%2Fj.eswa.2024.124278&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_eswa_2024_124278 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0957-4174&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0957-4174&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0957-4174&client=summon |