pith. sign in

arxiv: 2406.07353 · v1 · submitted 2024-06-11 · 💻 cs.CL · cs.AI· cs.CV· cs.CY· cs.SI

Toxic Memes: A Survey of Computational Perspectives on the Detection and Explanation of Meme Toxicities

Delfina Sol Martinez Pandiani , Erik Tjong Kim Sang , Davide Ceolin This is my paper

Pith reviewed 2026-05-23 23:46 UTC · model grok-4.3

classification 💻 cs.CL cs.AIcs.CVcs.CYcs.SI
keywords toxic memesmeme toxicity detectioncomputational analysistaxonomy of toxicitytarget intent conveyancelarge language modelscontent moderationsurvey
0
0 comments X

The pith

A new taxonomy organizes meme toxicity around target, intent, and conveyance tactics.

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

The survey reviews 158 computational works on toxic memes published through early 2024, extending earlier reviews that stopped at 2022. It introduces a taxonomy to resolve inconsistent definitions of toxicity and identifies three recurring dimensions in the literature: the target of the meme, the apparent intent, and the tactics used to convey the message. The authors build a framework that maps how these dimensions relate to specific toxicity types. This matters for readers because memes function as a primary channel for spreading harmful messages, and computational systems need clearer structures to move past simple yes-or-no labels.

Core claim

After applying PRISMA to locate papers, the authors introduce a taxonomy for categorizing meme toxicity types and identify three content-based dimensions under automatic study—target, intent, and conveyance tactics—while developing a framework that illustrates the relationships between these dimensions and meme toxicities; they also document the shift of computational tasks toward nuanced comprehension of toxicity and the rising use of large language models for both detection and generation.

What carries the argument

The new taxonomy for categorizing meme toxicity types together with the framework that links the three dimensions of target, intent, and conveyance tactics to observed toxicities.

If this is right

  • Computational tasks expand beyond binary toxic versus non-toxic classification toward nuanced comprehension of toxicity.
  • Enhanced cross-modal reasoning and integration of expert and cultural knowledge become necessary for effective systems.
  • Automatic explanation of detected toxicities emerges as a required capability.
  • Dedicated methods are needed to handle meme toxicity in low-resource languages.
  • Large language models are applied both to detect and to generate toxic memes.

Where Pith is reading between the lines

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

  • The three-dimension framework could be tested by re-annotating existing meme datasets and checking consistency of labels across the dimensions.
  • Platforms using automated moderation might adopt the taxonomy to produce more granular removal or warning decisions.
  • The documented rise in generative AI for toxic memes points to a need for separate safeguards against synthetic content.
  • The taxonomy offers a possible bridge between computational work and studies of visual misinformation in other media.

Load-bearing premise

The PRISMA search and inclusion criteria applied to literature through early 2024 captured a representative set of content-based computational works on toxic memes.

What would settle it

Discovery of multiple papers on content-based toxic meme analysis from 2023 or earlier that match the stated inclusion criteria yet were not counted among the 158 works.

Figures

Figures reproduced from arXiv: 2406.07353 by Davide Ceolin, Delfina Sol Martinez Pandiani, Erik Tjong Kim Sang.

Figure 1
Figure 1. Figure 1: Graph depicting the exponential increase in publications within the field of computer science, as indexed by SCOPUS, focusing on research related to toxic memes. The data was gathered using a query targeting specific keywords associated with meme toxicities (see Section 4). independently of their propagation. While topics like meme propagation and contextual dynamics within social networks, such as measuri… view at source ↗
Figure 2
Figure 2. Figure 2: PRISMA 2020 flow diagram for systematic reviews on *SCOPUS and Web of Science (WOS) databases. Registers refers to SCOPUS preprints. Records excluded due to: ** Topic Non-Relevance. *** Computational Non-Relevance. Selection of Database We used Scopus3 and Web of Science (WOS)4 for our literature search because they are two of the largest and most reputable databases, indexing a wide range of high-quality … view at source ↗
Figure 3
Figure 3. Figure 3: Left: Distribution of surveyed papers by publication year. The figure illustrates a steady increase in the number of publications from year to year. Right: Comparison of coverage across previous surveys based on the papers surveyed here. labels, task definitions (e.g., binary, single label multi-class, multi-label multi-class, etc.), and baseline macro F1 scores. This exhaustive examination yielded over 30… view at source ↗
Figure 4
Figure 4. Figure 4: Top: Pie chart showing the distribution of datasets based on the number of memes they contain. Approximately 40% of datasets contain between 1000 to 5000 memes, followed by nearly 30% with 5000 to 10,000 memes. Less than 3% of the datasets have over 15,000 memes. Middle: Pie chart illustrating the distribution of dataset languages. English dominates with nearly 75% of the datasets exclusively in English, w… view at source ↗
Figure 5
Figure 5. Figure 5: provides an overview of research attention across various meme toxicity categories. Notably, there is a clear imbalance, with a significant focus on hateful memes. Nearly half of the research papers concentrate on this aspect, highlighting the prevalence of hate speech in online discourse. This trend indicates the significant impact of the hateful memes challenge proposed by Kiela et al. (2020) [53], as ev… view at source ↗
Figure 6
Figure 6. Figure 6: The taxonomy for meme toxicities that we propose is inspired by the taxonomy presented in [10], while addressing discrepancies, enhancing taxonomical clarity, and including the most recent types of toxicities being computationally studied [PITH_FULL_IMAGE:figures/full_fig_p019_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Venn diagram illustrating taxonomical relations and fuzzy boundaries between meme toxicities, along with examples. Memes in the exploitation category are not depicted due to the absence of corresponding datasets. The diagram is color-coded to match our taxonomy. Caution: Depicted memes contain toxic content, potentially inducing psychological distress. Viewer discretion is advised. These images do not refl… view at source ↗
Figure 8
Figure 8. Figure 8: The toxicity of memes is a complex phenomenon with multiple dimensions, represented by dotted edges. In this survey, we have identified three content-based dimensions of meme toxicity: target, intent, and tactic. These dimensions address the fundamental questions: who is the toxicity directed towards, why the toxic meme is shared (i.e., the underlying goal), and how the toxicity is manifested and conveyed.… view at source ↗
Figure 9
Figure 9. Figure 9: While relationships between attack types, persuasion/propagandistic techniques, and entity roles have not been thoroughly explored, insights can be gleaned from their definitions, suggesting potential connections, equivalences, and other relationships (illustrated by red dotted lines) [PITH_FULL_IMAGE:figures/full_fig_p024_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Intersections between intent and target dimensions hint at specific types of meme toxicity. The y-axis, highlighted in yellow, categorically presents the three identified intents. Meanwhile, the x-axis, depicted in blue, categorizes target types with varying levels of specificity (from less specific at the bottom to more specific at the top). The green plane illustrates the toxicity landscape observed thr… view at source ↗
Figure 11
Figure 11. Figure 11: Illustration of how altering text modifies the interpretation of an image. The left meme conveys a message about water crisis awareness. The right meme alters the text, resulting in a hateful interpretation of the child as “farm equipment”, perpetuating dehumanization and racism. We blurred the face of the child for privacy concerns. Caution: The depicted memes contain toxic content. Viewer discretion is … view at source ↗
Figure 12
Figure 12. Figure 12: Toxicity-related terms derived from our investigation of harmfulness and toxicity in multimodal data, illustrating the complex and overlapping nature of multimodal toxicities. DS Martinez Pandiani et al.: Preprint submitted to Elsevier Page 30 of 39 [PITH_FULL_IMAGE:figures/full_fig_p031_12.png] view at source ↗
read the original abstract

Internet memes, channels for humor, social commentary, and cultural expression, are increasingly used to spread toxic messages. Studies on the computational analyses of toxic memes have significantly grown over the past five years, and the only three surveys on computational toxic meme analysis cover only work published until 2022, leading to inconsistent terminology and unexplored trends. Our work fills this gap by surveying content-based computational perspectives on toxic memes, and reviewing key developments until early 2024. Employing the PRISMA methodology, we systematically extend the previously considered papers, achieving a threefold result. First, we survey 119 new papers, analyzing 158 computational works focused on content-based toxic meme analysis. We identify over 30 datasets used in toxic meme analysis and examine their labeling systems. Second, after observing the existence of unclear definitions of meme toxicity in computational works, we introduce a new taxonomy for categorizing meme toxicity types. We also note an expansion in computational tasks beyond the simple binary classification of memes as toxic or non-toxic, indicating a shift towards achieving a nuanced comprehension of toxicity. Third, we identify three content-based dimensions of meme toxicity under automatic study: target, intent, and conveyance tactics. We develop a framework illustrating the relationships between these dimensions and meme toxicities. The survey analyzes key challenges and recent trends, such as enhanced cross-modal reasoning, integrating expert and cultural knowledge, the demand for automatic toxicity explanations, and handling meme toxicity in low-resource languages. Also, it notes the rising use of Large Language Models (LLMs) and generative AI for detecting and generating toxic memes. Finally, it proposes pathways for advancing toxic meme detection and interpretation.

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

0 major / 4 minor

Summary. The manuscript is a systematic literature survey employing the PRISMA methodology to review content-based computational work on toxic memes through early 2024. It extends prior surveys by incorporating 119 new papers (158 works total), enumerates over 30 datasets and their labeling systems, introduces a new taxonomy for meme toxicity types motivated by observed definitional inconsistencies, identifies three content-based dimensions (target, intent, conveyance tactics) with an accompanying framework, and discusses challenges, trends including LLM usage, and future research pathways.

Significance. If the coverage is representative, the survey supplies a timely consolidation of the field since the 2022 cutoff of earlier reviews, supplies a taxonomy and three-dimension framework that can reduce terminological inconsistency, and flags actionable trends such as cross-modal reasoning, expert knowledge integration, automatic explanation generation, and LLM applications. These elements, together with the explicit dataset enumeration, position the work as a useful reference for standardizing and advancing computational meme toxicity research.

minor comments (4)
  1. [Abstract] The abstract states that the work achieves a 'threefold result' yet enumerates the contributions in a way that blends dataset analysis with the survey count; a clearer three-bullet structure would improve readability.
  2. [Datasets section] The discussion of 'over 30 datasets' would be substantially more usable if accompanied by a summary table that records dataset size, label granularity, public availability, and the toxicity dimensions each covers.
  3. [Framework figure] The framework diagram illustrating relationships among target, intent, and conveyance tactics would benefit from explicit visual cues (arrows, color coding, or call-outs) that link each dimension to concrete toxicity categories discussed in the taxonomy.
  4. [Methods / PRISMA protocol] The PRISMA protocol description would allow readers to assess completeness if the exact search strings, Boolean operators, and the complete list of queried databases were supplied, even if only in an appendix.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive and accurate summary of our PRISMA-based survey, the assessment of its significance in consolidating recent work, and the recommendation for minor revision. No major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity; minor self-citation possible but non-load-bearing

full rationale

This is a systematic literature survey employing PRISMA to review 158 computational works on toxic memes through early 2024, extending three prior surveys. The central contributions—a new taxonomy for toxicity types and a three-dimension framework (target, intent, conveyance tactics)—are derived directly from patterns observed across the reviewed papers rather than from any equations, fitted parameters, or self-referential definitions. No derivation chain reduces to its own inputs by construction, and any potential self-citation of the authors' prior work is not load-bearing for the taxonomy or framework. The survey remains self-contained against external benchmarks of literature coverage.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

This is a literature survey paper. It relies on the PRISMA systematic review methodology as a domain standard and introduces a novel taxonomy based on analysis of existing works. No free parameters or invented physical entities are present.

axioms (1)
  • domain assumption PRISMA methodology provides an appropriate and reproducible way to select and synthesize computational papers on toxic memes
    The paper states it employs the PRISMA methodology to systematically extend previously considered papers.

pith-pipeline@v0.9.0 · 5850 in / 1418 out tokens · 22289 ms · 2026-05-23T23:46:46.861377+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

253 extracted references · 253 canonical work pages · 3 internal anchors

  1. [1]

    Dawkins, The Selfish Gene, new ed., Oxford University Press, 1989

    R. Dawkins, The Selfish Gene, new ed., Oxford University Press, 1989

    work page 1989

  2. [2]

    Dennet, Dangerous memes - a ted talk, Youtube, 2007

    D. Dennet, Dangerous memes - a ted talk, Youtube, 2007. URL:https://www.youtube.com/watch?v=KzGjEkp772s

    work page 2007

  3. [3]

    Koutlis, M

    C. Koutlis, M. Schinas, S. Papadopoulos, Memetector: Enforcing deep focus for meme detection, International Journal of Multimedia Information Retrieval 12 (2023) 11

    work page 2023

  4. [4]

    Peeters, M

    S. Peeters, M. Tuters, T. Willaert, D. De Zeeuw, On the vernacular language games of an antagonistic online subculture, Frontiers in big Data 4 (2021) 718368

    work page 2021

  5. [5]

    Arkenbout, Political meme toolkit: leftist dutch meme makers share their trade secrets, in: Critical meme reader II: memetic tacticality, Institute of Network Cultures, 2022, pp

    C. Arkenbout, Political meme toolkit: leftist dutch meme makers share their trade secrets, in: Critical meme reader II: memetic tacticality, Institute of Network Cultures, 2022, pp. 20–31

    work page 2022

  6. [6]

    Wagener, Semiotic excess in memes: From postdigital creativity to social violence, Internet Pragmatics 6 (2023) 239–258

    A. Wagener, Semiotic excess in memes: From postdigital creativity to social violence, Internet Pragmatics 6 (2023) 239–258

    work page 2023

  7. [7]

    Kiela, H

    D. Kiela, H. Firooz, A. Mohan, V. Goswami, A. Singh, C. Fitzpatrick, P. Bull, G. Lipstein, T. Nelli, R. Zhu, N. Muennighoff, R. Velioglu, J. Rose, P. Lippe, N. Holla, S. Chandra, S. Rajamanickam, G. Antoniou, E. Shutova, H. Yannakoudakis, V. Sandulescu, U. Ozertem, P. Pantel, L. Specia, D. Parikh, The hateful memes challenge: Competition report, in: Proce...

    work page 2020

  8. [8]

    H. Kirk, Y. Jun, P. Rauba, G. Wachtel, R. Li, X. Bai, N. Broestl, M. Doff-Sotta, A. Shtedritski, Y. Asano, Memes in the wild: Assessing the generalizability of the hateful memes challenge dataset, in: Proceedings of WOAH 2021 - 5th Workshop on Online Abuse and Harms, volume 26, 2021

    work page 2021

  9. [9]

    Kumar, K

    G. Kumar, K. Nandakumar, Hate-clipper: Multimodal hateful meme classification based on cross-modal interaction of clip features, in: NLP4PI 2022 - 2nd Workshop on NLP for Positive Impact, Proceedings of the Workshop, volume 171, 2022

    work page 2022

  10. [10]

    Sharma, F

    S. Sharma, F. Alam, M. S. Akhtar, D. Dimitrov, G. Da San Martino, H. Firooz, A. Halevy, F. Silvestri, P. Nakov, T. Chakraborty, Detecting and understanding harmful memes: A survey, arXiv preprint arXiv:2205.04274 1 (2022) 1–9

    work page arXiv 2022

  11. [11]

    Grasso, V

    B. Grasso, V. La Gatta, V. Moscato, G. Sperlì, Kermit: Knowledge-empowered model in harmful meme detection, Information Fusion 106 (2024). DS Martinez Pandiani et al.:Preprint submitted to Elsevier Page 31 of 39 Toxic Memes

    work page 2024

  12. [12]

    C. Yang, F. Zhu, J. Han, S. Hu, Invariant meets specific: A scalable harmful memes detection framework, MM 2023 - Proceedings of the 31st ACM International Conference on Multimedia 4788 (2023)

    work page 2023

  13. [13]

    M. Das, A. Mukherjee, Banglaabusememe: A dataset for bengali abusive meme classification, arXiv preprint arXiv:2310.11748 (2023) 15498–15512

    work page arXiv 2023

  14. [14]

    R. Jain, K. Maity, P. Jha, S. Saha, Generative models vs discriminative models: Which performs better in detecting cyberbullying in memes?, in: Proceedings of the International Joint Conference on Neural Networks, 2023. doi:10.1109/ijcnn54540.2023.10191363

    work page doi:10.1109/ijcnn54540.2023.10191363 2023

  15. [15]

    Shang, Y

    L. Shang, Y. Zhang, Y. Zha, Y. Chen, C. Youn, D. Wang, Aomd: An analogy-aware approach to offensive meme detection on social media, Information Processing & Management (2021)

    work page 2021

  16. [16]

    Chakraborty,V.Pulabaigari,B.Gamback, Semeval-2020 task8:Memotionanalysis–the visuo-lingual metaphor!, arXiv preprint arXiv:2008.03781 (2020) 759–773

    C.Sharma,D.Bhageria,W.Scott,S.Pykl,A.Das,T. Chakraborty,V.Pulabaigari,B.Gamback, Semeval-2020 task8:Memotionanalysis–the visuo-lingual metaphor!, arXiv preprint arXiv:2008.03781 (2020) 759–773

    work page arXiv 2020

  17. [18]

    Williams, M

    A. Williams, M. Dupuis, I don’t always spread disinformation on the web, but when i do i like to use memes: An examination of memes in the spread of disinformation, in: Proceedings of the 11th International Multi-Conferences on Complexity, Informatics and Cybernetics: IMCIC, 2020, pp. 165–172

    work page 2020

  18. [19]

    Dimitrov, B

    D. Dimitrov, B. B. Ali, S. Shaar, F. Alam, F. Silvestri, H. Firooz, G. D. S. Martino, Detecting propaganda techniques in memes, arXiv preprint arXiv:2109.08013 (2021) 6603–6617

    work page arXiv 2021

  19. [20]

    Abdullah, D

    M. Abdullah, D. Abujaber, A. Al-Qarqaz, R. Abbott, M. Hadzikadic, Combating propaganda texts using transfer learning, IAES International Journal of Artificial Intelligence 12 (2023) 956–965

    work page 2023

  20. [21]

    Rodríguez, P

    D. Rodríguez, P. Nakov, V. Dankers, E. Shutova, Paper bullets: Modeling propaganda with the help of metaphor, European Chapter of the Association for Computational Linguistics, Findings of EACL 2023 (2023) 472–489

    work page 2023

  21. [22]

    M. G. Shridara, D. Hládek, M. Pleva, R. Haluška, Identification of trolling in memes using convolutional neural networks, in: 2023 33rd International Conference Radioelektronika (RADIOELEKTRONIKA), IEEE, 2023, pp. 1–6

    work page 2023

  22. [23]

    S.Suryawanshi,B.R.Chakravarthi,M.Arcan,P.Buitelaar, Trollswithopinion:Ataxonomyanddatasetforpredictingdomain-specificopinion manipulation in troll memes, Multimedia Tools and Applications 82 (2023) 9137–9171

    work page 2023

  23. [24]

    Bebić, M

    D. Bebić, M. Volarevic, Do not mess with a meme: the use of viral content in communicating politics, Communication & Society 31 (2018) 43–56

    work page 2018

  24. [25]

    Mazambani, M

    G. Mazambani, M. A. Carlson, S. Reysen, C. F. Hempelmann, Impact of status and meme content on the spread of memes in virtual communities, Human Technology: An Interdisciplinary Journal on Humans in ICT Environments 11 (2015) 148–164

    work page 2015

  25. [26]

    [Accessed 17-04-2024]

    Netanya, The Dangers of Memes — netanyataitague, https://medium.com/@netanyataitague/ the-dangers-of-memes-b1bb67e10083 , 2019. [Accessed 17-04-2024]

    work page 2019

  26. [27]

    Vang, Is Meme Culture Problematic - The Current — thecurrentmsu.com, https://thecurrentmsu.com/2021/08/07/ is-meme-culture-problematic/, 2021

    M. Vang, Is Meme Culture Problematic - The Current — thecurrentmsu.com, https://thecurrentmsu.com/2021/08/07/ is-meme-culture-problematic/, 2021. [Accessed 17-04-2024]

    work page 2021

  27. [28]

    Rojas, The Toxicity of Online Meme Culture: When Is It Too Far? — studybreaks.com,https://studybreaks.com/thoughts/ meme-culture-2/, 2022

    K. Rojas, The Toxicity of Online Meme Culture: When Is It Too Far? — studybreaks.com,https://studybreaks.com/thoughts/ meme-culture-2/, 2022. [Accessed 17-04-2024]

    work page 2022

  28. [29]

    Z. D. Roberts, How the ‘Free Helicopter Rides’ Meme Went Viral — progressive.org,https://progressive.org/magazine/ how-the-free-helicopter-rides-meme-went-viral-roberts-20230907/ , 2023. [Accessed 17-04-2024]

    work page 2023

  29. [30]

    F. J. A. Serna, Los memes como simbolos del discurso de odio: La influencia del humor gráfico en la libertad de expresión y la política, VISUAL REVIEW. International Visual Culture Review/Revista Internacional de Cultura Visual 16 (2024) 241–253

    work page 2024

  30. [31]

    Needham, How the toxic went mainstream — pursuit.unimelb.edu.au, https://pursuit.unimelb.edu.au/articles/ how-the-toxic-went-mainstream , 2019

    L. Needham, How the toxic went mainstream — pursuit.unimelb.edu.au, https://pursuit.unimelb.edu.au/articles/ how-the-toxic-went-mainstream , 2019. [Accessed 17-04-2024]

    work page 2019

  31. [32]

    K. M. Duchscherer, J. F. Dovidio, When memes are mean: Appraisals of and objections to stereotypic memes, Translational Issues in Psychological Science 2 (2016) 335–345

    work page 2016

  32. [33]

    P. M. Bennet, Who Moderates the Social Media Giants? A Call to End Outsourcing, Technical Report, NYU STERN, Center for Business and Human Rights, 2020

    work page 2020

  33. [34]

    Nondo, Facing disturbing content daily, online moderators in africa want better protections and a fair wage, CBC Radio (2023)

    N. Nondo, Facing disturbing content daily, online moderators in africa want better protections and a fair wage, CBC Radio (2023)

    work page 2023

  34. [35]

    Perrigo, Inside facebook’s african sweatshop, Time (2022)

    B. Perrigo, Inside facebook’s african sweatshop, Time (2022)

    work page 2022

  35. [36]

    it’s destroyed me completely

    N. Rowe, "it’s destroyed me completely": Kenyan moderators decry toll of training of ai models, The Guardian (2023)

    work page 2023

  36. [37]

    Mbagathi, In africa, taking on viral hate, Open Society Foundations (2023)

    N. Mbagathi, In africa, taking on viral hate, Open Society Foundations (2023)

    work page 2023

  37. [38]

    1451–1466

    T.H.Afridi,A.Alam,M.N.Khan,J.Khan,Y.-K.Lee, Amultimodalmemesclassification:Asurveyandopenresearchissues, in:Innovations in Smart Cities Applications Volume 4: The Proceedings of the 5th International Conference on Smart City Applications, Springer, 2021, pp. 1451–1466

    work page 2021

  38. [39]

    P. C. d. Q. Hermida, E. M. D. Santos, Detecting hate speech in memes: a review, Artificial Intelligence Review 56 (2023) 1–19

    work page 2023

  39. [40]

    Zannettou, T

    S. Zannettou, T. Caulfield, J. Blackburn, E. De Cristofaro, M. Sirivianos, G. Stringhini, G. Suarez-Tangil, On the origins of memes by means of fringe web communities, in: Proceedings of the internet measurement conference 2018, 2018, pp. 188–202

    work page 2018

  40. [41]

    Chakraborty, S

    T. Chakraborty, S. Masud, Nipping in the bud: detection, diffusion and mitigation of hate speech on social media, ACM SIGWEB Newsletter 2022 (2022) 1–9

    work page 2022

  41. [42]

    Polli, M

    C. Polli, M. G. Sindoni, Multimodal computation or interpretation? automatic vs. critical understanding of text-image relations in racist memes, Ssrn (2023)

    work page 2023

  42. [43]

    Solon, Richard dawkins on the internet’s hijacking of the word ‘meme’, Wired UK 20 (2013)

    O. Solon, Richard dawkins on the internet’s hijacking of the word ‘meme’, Wired UK 20 (2013)

    work page 2013

  43. [44]

    Shifman, Memes in Digital Culture, MIT Press Essential Knowledge Series, MIT Press, 2013

    L. Shifman, Memes in Digital Culture, MIT Press Essential Knowledge Series, MIT Press, 2013

    work page 2013

  44. [45]

    Dynel, The life of covid-19 mask memes: a diachronic study of the pandemic memescape, Comunicar 30 (2022) 73–85

    M. Dynel, The life of covid-19 mask memes: a diachronic study of the pandemic memescape, Comunicar 30 (2022) 73–85. DS Martinez Pandiani et al.:Preprint submitted to Elsevier Page 32 of 39 Toxic Memes

    work page 2022

  45. [46]

    Thakur, F

    A. Thakur, F. Ilievski, H.-A. Sandlin, Z. Sourati, L. Luceri, R. Tommasini, A. Mermoud, Explainable classification of internet memes, in: CEUR Workshop Proceedings, volume 3432, 2023, pp. 395–409

    work page 2023

  46. [47]

    L.Xie,A.Natsev,J.R.Kender,M.Hill,J.R.Smith, Visualmemesinsocialmedia:trackingreal-worldnewsinyoutubevideos, in:Proceedings of the 19th ACM international conference on Multimedia, 2011, pp. 53–62

    work page 2011

  47. [48]

    Y. Du, M. A. Masood, K. Joseph, Understanding visual memes: An empirical analysis of text superimposed on memes shared on twitter, in: Proceedings of the International AAAI Conference on Web and Social Media, volume 14, Aaai, 2020, pp. 153–164

    work page 2020

  48. [49]

    Pramanick, D

    S. Pramanick, D. Dimitrov, R. Mukherjee, S. Sharma, M. S. Akhtar, P. Nakov, T. Chakraborty, Detecting harmful memes and their targets, arXiv preprint arXiv:2110.00413 (2021)

    work page arXiv 2021

  49. [50]

    Knobel, C

    M. Knobel, C. Lankshear, Online memes, affinities, and cultural production, A new literacies sampler 29 (2007) 199–227

    work page 2007

  50. [51]

    Sharma, V

    C. Sharma, V. Pulabaigari, A. Das, Meme vs. non-meme classification using visuo-linguistic association., in: WEBIST, 2020, pp. 353–360

    work page 2020

  51. [52]

    Sherratt, K

    V. Sherratt, K. Pimbblet, N. Dethlefs, Multi-channel convolutional neural network for precise meme classification, in: Proceedings of the 2023 ACM International Conference on Multimedia Retrieval, 2023, pp. 190–198

    work page 2023

  52. [53]

    Kiela, H

    D. Kiela, H. Firooz, A. Mohan, V. Goswami, A. Singh, P. Ringshia, D. Testuggine, The hateful memes challenge: Detecting hate speech in multimodal memes, in: Advances in neural information processing systems, volume 33, 2020, pp. 2611–2624

    work page 2020

  53. [54]

    Kostadinovska-Stojchevska, E

    B. Kostadinovska-Stojchevska, E. Shalevska, Internet memes and their socio-linguistic features, European journal of literature, language and linguistics studies 2 (2018)

    work page 2018

  54. [55]

    Lapidot-Lefler, A

    N. Lapidot-Lefler, A. Barak, Effects of anonymity, invisibility, and lack of eye-contact on toxic online disinhibition, Computers in human behavior 28 (2012) 434–443

    work page 2012

  55. [56]

    Gordeev, V

    D. Gordeev, V. Potapov, Toxicity in texts and images on the internet, in: International Conference on Speech and Computer, Springer, 2020, pp. 156–165

    work page 2020

  56. [57]

    Sheth, V

    A. Sheth, V. L. Shalin, U. Kursuncu, Defining and detecting toxicity on social media: context and knowledge are key, Neurocomputing 490 (2022) 312–318

    work page 2022

  57. [58]

    Carlisle, Toxicity, memes and raids - Power of Zero — powerof0.org,https://powerof0.org/toxicity-memes-and-raids/, 2022

    N. Carlisle, Toxicity, memes and raids - Power of Zero — powerof0.org,https://powerof0.org/toxicity-memes-and-raids/, 2022. [Accessed 17-04-2024]

    work page 2022

  58. [59]

    Mosleh, R

    M. Mosleh, R. Cole, D. G. Rand, Misinformation and harmful language are interconnected, rather than distinct, challenges, PNAS nexus (2024) pgae111

    work page 2024

  59. [60]

    S.Ghosh,S.Lepcha,S.Sakshi,R.R.Shah,S.Umesh, Detoxy:Alarge-scalemultimodaldatasetfortoxicityclassificationinspokenutterances, arXiv preprint arXiv:2110.07592 (2021)

    work page arXiv 2021

  60. [61]

    Understanding Abuse: A Typology of Abusive Language Detection Subtasks

    Z. Waseem, T. Davidson, D. Warmsley, I. Weber, Understanding abuse: A typology of abusive language detection subtasks, arXiv preprint arXiv:1705.09899 (2017)

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  61. [62]

    M.Wiegand,M.Siegel,J.Ruppenhofer, Overviewofthegermeval2018sharedtaskontheidentificationofoffensivelanguage, in:Proceedings of GermEval 2018, 14th Conference on Natural Language Processing (KONVENS 2018), Austrian Academy of Sciences, 2019, pp. 1 – 10. URL: https://nbn-resolving.org/urn:nbn:de:bsz:mh39-84935

    work page 2018

  62. [63]

    SemEval-2019 Task 6: Identifying and Categorizing Offensive Language in Social Media (OffensEval)

    M. Zampieri, S. Malmasi, P. Nakov, S. Rosenthal, N. Farra, R. Kumar, Semeval-2019 task 6: Identifying and categorizing offensive language in social media (offenseval), arXiv preprint arXiv:1903.08983 (2019)

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  63. [64]

    P. Piot, P. Martín-Rodilla, J. Parapar, Metahate: A dataset for unifying efforts on hate speech detection, arXiv preprint arXiv:2401.06526 (2024)

    work page arXiv 2024

  64. [65]

    Adams, J

    C. Adams, J. Sorensen, J. Elliott, L. Dixon, M. McDonald, n., W. Cukierski, Toxic comment classification challenge, 2017. URL: https://kaggle.com/competitions/jigsaw-toxic-comment-classification-challenge

    work page 2017

  65. [66]

    Davidson, D

    T. Davidson, D. Warmsley, M. Macy, I. Weber, Automated hate speech detection and the problem of offensive language, in: Proceedings of the international AAAI conference on web and social media, volume 11, 2017, pp. 512–515

    work page 2017

  66. [67]

    F.E.Ayo,O.Folorunso,F.T.Ibharalu,I.A.Osinuga, Machinelearningtechniquesforhatespeechclassificationoftwitterdata:State-of-the-art, future challenges and research directions, Computer Science Review 38 (2020) 100311

    work page 2020

  67. [68]

    Alkomah, X

    F. Alkomah, X. Ma, A literature review of textual hate speech detection methods and datasets, Information 13 (2022) 273

    work page 2022

  68. [69]

    Chhabra, D

    A. Chhabra, D. K. Vishwakarma, A literature survey on multimodal and multilingual automatic hate speech identification, Multimedia Systems 29 (2023) 1203–1230

    work page 2023

  69. [70]

    Fortuna, J

    P. Fortuna, J. Soler, L. Wanner, Toxic, hateful, offensive or abusive? what are we really classifying? an empirical analysis of hate speech datasets, in: Proceedings of the Twelfth Language Resources and Evaluation Conference, 2020, pp. 6786–6794

    work page 2020

  70. [71]

    Yankoski, W

    M. Yankoski, W. Scheirer, T. Weninger, Meme warfare: Ai countermeasures to disinformation should focus on popular, not perfect, fakes, Bulletin of the atomic scientists 77 (2021) 119–123

    work page 2021

  71. [72]

    Banko, B

    M. Banko, B. MacKeen, L. Ray, A unified taxonomy of harmful content, in: Proceedings of the fourth workshop on online abuse and harms, 2020, pp. 125–137

    work page 2020

  72. [73]

    Nakov, V

    P. Nakov, V. Nayak, K. Dent, A. Bhatawdekar, S. M. Sarwar, M. Hardalov, Y. Dinkov, D. Zlatkova, G. Bouchard, I. Augenstein, Detecting abusive language on online platforms: A critical analysis, arXiv preprint arXiv:2103.00153 (2021)

    work page arXiv 2021

  73. [74]

    Halevy, C

    A. Halevy, C. Canton-Ferrer, H. Ma, U. Ozertem, P. Pantel, M. Saeidi, F. Silvestri, V. Stoyanov, Preserving integrity in online social networks, Communications of the ACM 65 (2022) 92–98

    work page 2022

  74. [75]

    F. Alam, S. Cresci, T. Chakraborty, F. Silvestri, D. Dimitrov, G. D. S. Martino, S. Shaar, H. Firooz, P. Nakov, A survey on multimodal disinformation detection, arXiv preprint arXiv:2103.12541 (2021)

    work page arXiv 2021

  75. [76]

    Anjum, R. Katarya, Hate speech, toxicity detection in online social media: a recent survey of state of the art and opportunities, International Journal of Information Security 23 (2024) 577–608

    work page 2024

  76. [77]

    Lewandowska-Tomaszczyk, A

    B. Lewandowska-Tomaszczyk, A. Bączkowska, C. Liebeskind, G. Valunaite O., S. Žitnik, An integrated explicit and implicit offensive language taxonomy, Lodz Papers in Pragmatics 19 (2023) 7–48

    work page 2023

  77. [78]

    T. Garg, S. Masud, T. Suresh, T. Chakraborty, Handling bias in toxic speech detection: A survey, ACM Computing Surveys 55 (2023) 1–32. DS Martinez Pandiani et al.:Preprint submitted to Elsevier Page 33 of 39 Toxic Memes

    work page 2023

  78. [79]

    H. Lin, Z. Luo, W. Gao, J. Ma, B. Wang, R. Yang, Towards explainable harmful meme detection through multimodal debate between large language models, arXiv preprint arXiv:2401.13298 (2024)

    work page arXiv 2024

  79. [80]

    P. Jha, K. Maity, R. Jain, A. Verma, S. Saha, P. Bhattacharyya, Meme-ingful analysis: Enhanced understanding of cyberbullying in memes through multimodal explanations, arXiv preprint arXiv:2401.09899 (2024)

    work page arXiv 2024

  80. [81]

    H. Lin, Z. Luo, J. Ma, L. Chen, Beneath the surface: Unveiling harmful memes with multimodal reasoning distilled from large language models, arXiv preprint arXiv:2312.05434 9114 (2023)

    work page arXiv 2023

Showing first 80 references.