Pith. sign in

REVIEW 5 major objections 4 minor 34 references

Cyberbullying Detection via Aggression-Enhanced Prompting

T0 review · 5 major / 4 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Enriching cyberbullying prompts with predicted aggression labels consistently beats standard LoRA fine-tuning, raising macro-F1 to 0.99 on two of three language models.

desk verdict A plausible idea with a confounded comparison: the EPP gain over LoRA is not attributable to the aggression label until a neutral-prefix control and label-quality check are run. read the letter →

arxiv 2508.06360 v2 pith:LXRI7XIB submitted 2025-08-08 cs.CL

Aisha Saeid , Anu Sabu , Girish A. Koushik , Ferrante Neri , Diptesh Kanojia This is my paper
classification cs.CL
keywords cyberbullyingdetectionaggressionpromptenrichmentLoRAfine-tuningmulti-tasklearninglargelanguagemodelscontextualaugmentationsocialmediasafety
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to establish that adding a predicted aggression label to the prompt improves cyberbullying detection across three instruction-tuned language models, compared with fine-tuning the same models on the cyberbullying data alone. It tests zero-shot, few-shot, LoRA fine-tuning, joint multi-task learning, and an enriched prompt pipeline (EPP) that prepends the predicted aggression category—overtly aggressive, covertly aggressive, or not aggressive—to each post before classification. On macro-F1 for cyberbullying, EPP reaches 0.99 for two models and 0.86 for a third, beating the corresponding LoRA baselines of 0.84, 0.93, and 0.84. The paper's point is that cheap prompt-level context can substitute for extra training signal in a low-resource safety task, though the results are presented as preliminary.

What carries the argument

The central mechanism is the Enriched Prompt Pipeline (EPP), a two-stage inference procedure: a LoRA-tuned aggression model first labels each post as Overtly Aggressive, Covertly Aggressive, or Not-Aggressive; that label is inserted into a fixed prompt template that states the predicted aggression and asks for cyberbullying classification; then a second LoRA-tuned cyberbullying classifier reads the enriched prompt. This converts single-task text classification into a contextual reasoning problem, with aggression serving as a guiding signal, and it does so without changing model weights or adding training time beyond the original LoRA adapters.

What would settle it

Run the same EPP pipeline with the aggression label replaced by a neutral or randomly assigned label (for example, always 'Not-Aggressive' or a permuted label) and compare macro-F1 on the cyberbullying test set. If performance stays near 0.99, the gain is not from the aggression signal. As a second check, measure the aggression model's accuracy on the cyberbullying test posts; near-chance accuracy would show the labels cannot be carrying the semantic information the paper attributes to them.

Watch

Extended reading notes

Core claim

The paper's central discovery is that aggression-informed contextual prompting—embedding a predicted overt/covert/not-aggressive label into the cyberbullying classification prompt—consistently improves macro-F1 over standard LoRA fine-tuning on the same models. In the reported numbers, the enriched prompt pipeline raises Gemma-2-2B from 0.84 to 0.99, Gemma-2-9B from 0.93 to 0.99, and Gemma-3-4B from 0.84 to 0.86. The authors interpret this as evidence that auxiliary aggression signals, injected at inference time, help models recognize subtle and indirect bullying that they miss when classifying raw text alone. They also report that multi-task learning gives inconsistent results across models

Load-bearing premise

The paper assumes that the predicted aggression labels placed into the prompts are accurate enough to be informative on the cyberbullying test posts, and that the F1 gains come from the label content rather than from the prompt template alone; it reports no test-set accuracy for the aggression predictor and no control prompt with a neutral or random label.

Editorial extensions

If this is right

  • EPP raises cyberbullying macro-F1 to 0.99 on two of three models and 0.86 on the third, all above their LoRA baselines, without retraining the cyberbullying classifier at inference time.
  • The largest gain appears on the smallest model, suggesting prompt enrichment may be most useful when model capacity or annotated data is limited.
  • Because the aggression scores are identical between LoRA and EPP, the improvement is attributed entirely to the prompt context rather than to any change in the model weights.
  • Joint multi-task learning produced inconsistent results across models, so the paper argues that sequential prompt enrichment is a more reliable way to transfer aggression knowledge than joint training.
  • The same two-stage enrichment idea could be applied to other safety-critical text classification tasks where a coarse auxiliary label is cheap to predict.

Reading between the lines

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

  • A neutral-prefix control—replacing the aggression label with a fixed or random token—would test whether the F1 gain comes from the label's semantics or simply from a longer prompt template.
  • The aggression labeler is trained on five datasets spanning English, code-mixed Hindi-English, and Bengali transliteration, but the paper does not measure cross-domain transfer, so whether EPP generalizes beyond the Kaggle test set remains open.
  • Errors in the aggression predictor propagate into the prompt: a covertly aggressive post mislabeled 'Not-Aggressive' may suppress the cyberbullying classifier, so the pipeline is only as good as the auxiliary model on the target domain.
  • With F1 at 0.99, the Kaggle benchmark is near saturation; testing on a more imbalanced or out-of-domain set would reveal whether the enrichment effect persists at lower label quality.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

5 major / 4 minor

Summary. The paper investigates whether aggression detection can serve as an auxiliary signal for cyberbullying detection in instruction-tuned LLMs. It evaluates zero-shot, few-shot, LoRA-based supervised fine-tuning, joint multi-task learning, and a proposed Enriched Prompt Pipeline (EPP) in which predicted aggression labels are embedded into cyberbullying-classification prompts. Experiments use five aggression datasets (D1–D5) and one cyberbullying dataset (D6) with Gemma-2-2B, Gemma-2-9B, and Gemma-3-4B. Table 2 reports EPP macro-F1 scores of 0.99, 0.99, and 0.86 versus 0.84, 0.93, and 0.84 for standard LoRA on cyberbullying detection. The authors conclude that aggression-informed context significantly boosts detection.

Significance. If validated, the proposed EPP is simple, parameter-efficient, and practically appealing for low-resource moderation settings. The paper uses publicly available benchmarks and covers three model scales, which are strengths. However, the central causal claim is not currently established: the EPP comparison confounds prompt-format change with the semantic content of the predicted aggression label, the quality of the predicted labels on the target set is never reported, and the results lack statistical grounding. The contribution is therefore promising but preliminary.

major comments (5)
  1. [§3.3, §4.4, Table 2] The central claim that 'aggression-informed context significantly boosts cyberbullying detection' is confounded. EPP differs from LoRA in two ways: it prepends the sentence 'This post was predicted as [...]' and it includes a predicted label value. There is no control condition with a neutral or fixed prefix (e.g., 'This post was predicted as [not-applicable].') or with a shuffled/random label. Since the cyberbullying LoRA adapter was not trained on enriched prompts, the test-time input distribution is shifted. The reported F1 gains (0.84→0.99, 0.93→0.99, 0.84→0.86) could therefore be due to the prompt-format change alone. The paper must include a neutral-prefix ablation to support its causal interpretation.
  2. [§3.3, Step 1; §4.4] Step 1 passes D6 test posts through a LoRA aggression model, but no accuracy, F1, or predicted-label distribution is reported for that model on D6. If the predicted aggression labels are near-random on this out-of-domain data, the 'enrichment' is effectively an uninformative or constant prefix. The authors should report the aggression model's performance on D6 and ideally compare EPP with oracle/ground-truth aggression labels to verify that the semantic signal, not the template, drives the improvement.
  3. [§3.4, Table 2] All results are single-point estimates without standard deviations, confidence intervals, or significance tests. In particular, the Gemma-3-4B improvement (0.84 vs. 0.86) is only two macro-F1 points and may be within run-to-run noise given the fixed hyperparameters (e.g., lr=1e-4, r=8, 1 epoch). The 'consistently outperforms' claim requires repeated runs or at least a randomization test.
  4. [§3.1, Table 1, Table 2] The five aggression datasets D1–D5 are collapsed into a single 'Aggression Detection' column in Table 2, but the methodology never specifies whether the aggression LoRA was trained on pooled D1–D5, on a selected subset, or separately per dataset with averaged results. This matters because the datasets differ in language (English, Hindi-English, Bengali), source, and annotation conventions; an unexplained aggregation could bias the predicted labels used in EPP and make reproduction impossible. The exact training protocol must be described.
  5. [§4.4, Table 2] The generalization claim rests on a single cyberbullying dataset (D6). No second cyberbullying corpus, cross-platform evaluation, or domain-shift experiment is provided. EPP's advantage may be dataset-specific, so the paper should either temper the generalization language or add an external validation set.
minor comments (4)
  1. [§3.3, Step 2] The prompt template is described only schematically ('This post was predicted as [...]'). Please provide the exact full prompt, including the original post and any instruction formatting, to enable reproducibility.
  2. [§3.4] Footnote 1 ('Transformer Heads Library') lacks a URL or citation. Also, the model repository for Gemma-3-4B is named as 'unsloth/gemma-3-4b-it-unsloth-bnb-4bit'; please specify the quantization procedure and whether it was applied to all models.
  3. [§3.2] The MTL description is vague: 'joint loss (sum) from different heads to both LoRA adapters' and training epochs 'between 3 and 6' are underspecified. Exact loss weighting, epoch counts, and head architecture should be stated.
  4. [General] There is no code or data availability statement. Even a pointer to a repository and the exact Kaggle version/DOI for D6 would improve reproducibility. Also, the paper has minor typographical and formatting inconsistencies (e.g., 'gemma' capitalization, no in-text reference to Figure 1).

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: EPP is a sequential pipeline with independently trained aggression labels as input; the only self-citations are non-load-bearing dataset references.

full rationale

The paper's central claim is that the Enriched Prompt Pipeline (EPP) outperforms standard LoRA for cyberbullying detection. The EPP is a sequential pipeline (Section 3.3, Steps 1-3): a LoRA-tuned aggression model predicts OAG/CAG/NAG on each input; that predicted label is embedded in a fixed prompt template; and the enriched prompt is passed to a second LoRA-tuned cyberbullying classifier. None of these steps defines the cyberbullying output in terms of itself. The aggression labels come from a separately trained model on five aggression datasets (D1-D5); they are not fit to the cyberbullying test set (D6), so the reported EPP F1 scores are not numerically forced by construction. The comparison against standard LoRA is a between-condition comparison with the same adapter family; it may be confounded by the prompt-format change (no neutral-prefix control) and by unmeasured aggression-label accuracy on D6, but those are validity threats, not circular reductions. The paper does cite datasets from the authors' own group (Rawat et al. 2023 for D1; Nafis et al. 2023 for D2), and these citations are used as data sources, not as an argument that the EPP must work. The self-citations are therefore not load-bearing for the central empirical result. Overall, no derivation step reduces to its own input, and no fitted parameter is renamed as a prediction; the score reflects only the minor, non-load-bearing self-citation in dataset provenance.

Assumptions & free parameters 6 free parameters · 3 assumptions · 0 invented entities

The paper's central claim rests on the assumption that predicted aggression labels from a separate model are accurate and informative on the cyberbullying test posts, and that the EPP improvement is not an artifact of prompt format. No ablation or label-quality check is provided. The hyperparameters are manually chosen and not swept, and the MTL epoch variation is unjustified.

free parameters (6)
  • LoRA rank r = 8
    Chosen manually for all fine-tuning experiments.
  • Learning rate = 1e-4
    Chosen manually.
  • Batch size = 8
    Chosen manually.
  • Epochs for LoRA SFT = 1
    Chosen manually; the paper notes this is lightweight.
  • Epochs for MTL = 3-6
    Chosen manually and varied without a clear criterion; this may affect MTL results.
  • Few-shot k = 3
    Number of examples per class in few-shot evaluations.
assumptions (3)
  • domain assumption Aggression datasets and cyberbullying dataset are representative and accurately labeled enough for transfer.
    The EPP relies on predicted aggression labels being informative; the paper never evaluates aggression accuracy on the cyberbullying test set.
  • domain assumption The LoRA-fine-tuned cyberbullying model responds to prompt enrichment at inference time without being trained with enriched prompts.
    EPP modifies the prompt for a model trained on standard prompts; the paper assumes the model can exploit this new prefix.
  • domain assumption Macro-F1 is the appropriate primary metric for the imbalanced multiclass cyberbullying task.
    Stated in Section 3.2, but no alternative metrics are examined or justified beyond a brief note on class imbalance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Cyberbullying Detection via Aggression-Enhanced Prompting." pith.science (2026). https://pith.science/paper/LXRI7XIB

@misc{pith2026250806360,
  author       = {Pith},
  title        = {Pith review of: Cyberbullying Detection via Aggression-Enhanced Prompting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LXRI7XIB}},
  note         = {Machine review of arXiv:2508.06360}
}
read the original abstract

Detecting cyberbullying on social media remains a critical challenge due to its subtle and varied expressions. This study investigates whether integrating aggression detection as an auxiliary task within a unified training framework can enhance the generalisation and performance of large language models (LLMs) in cyberbullying detection. Experiments are conducted on five aggression datasets and one cyberbullying dataset using instruction-tuned LLMs. We evaluated multiple strategies: zero-shot, few-shot, independent LoRA fine-tuning, and multi-task learning (MTL). Given the inconsistent results of MTL, we propose an enriched prompt pipeline approach in which aggression predictions are embedded into cyberbullying detection prompts to provide contextual augmentation. Preliminary results show that the enriched prompt pipeline consistently outperforms standard LoRA fine-tuning, indicating that aggression-informed context significantly boosts cyberbullying detection. This study highlights the potential of auxiliary tasks, such as aggression detection, to improve the generalisation of LLMs for safety-critical applications on social networks.

Figures

Figures reproduced from arXiv: 2508.06360 by the authors.

Figure 1
Figure 1. An overview of the proposed system architecture for cyberbullying detection. The diagram [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

34 extracted references · 23 canonical work pages

  1. [1]

    URL: " 'urlintro :=

    ENTRY address author booktitle chapter edition editor howpublished institution journal key month note number organization pages publisher school series title type volume year eprint doi pubmed url lastchecked label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block STRINGS urlintro eprinturl eprintpr...

  2. [2]

    write newline

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

  3. [3]

    Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin...

    arXiv 2020

  4. [4]

    Yupeng Chang, Xu Wang, Jindong Wang, Yuan Wu, Linyi Yang, Kaijie Zhu, Hao Chen, Xiaoyuan Yi, Cunxiang Wang, and Yidong Wang. 2024. https://doi.org/10.1145/3641289 A survey on evaluation of large language models . ACM Transactions on Intelligent Systems and Technology, 15(3):1--45

    doi:10.1145/3641289 2024

  5. [5]

    Suliman Mohamed Fati, Mohammed A Mahdi, Mohamed AG Hazber, Shahanawaj Ahamad, Sawsan A Saad, Mohammed Gamal Ragab, and Mohammed Al-Shalabi. 2025. Enhancing multi-class cyberbullying classification with hybrid feature extraction and transformer-based models. Computer Modeling in Engineering & Sciences (CMES), 143(2)

    work page 2025

  6. [6]

    Gemma Team . 2024. Gemma: Open models for responsible ai. https://ai.google.dev/gemma. Google DeepMind

    work page 2024

  7. [7]

    Karel Guti \'e rrez-Batista, Jesica G \'o mez-S \'a nchez, and Carlos Fernandez-Basso. 2024. Improving automatic cyberbullying detection in social network environments by fine-tuning a pre-trained sentence transformer language model. Social Network Analysis and Mining, 14(1):136

    work page 2024

  8. [8]

    Huimin Han, Muhammad Asif, Emad Mahrous Awwad, Nadia Sarhan, Yazeed Yasid Ghadi, and Bo Xu. 2024. Innovative deep learning techniques for monitoring aggressive behavior in social media posts. Journal of Cloud Computing, 13(1):19

    work page 2024

Show all 34 references
  1. [9]

    Sameer Hinduja and Justin W. Patchin. 2010. https://doi.org/10.1080/13811118.2010.494133 Bullying, cyberbullying, and suicide . Archives of Suicide Research, 14(3):206--221

    2010

  2. [10]

    Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin de Laroussilhe, Andrea Gesmundo, Mona Attariyan, and Sylvain Gelly. 2019. http://arxiv.org/abs/1902.00751 Parameter-efficient transfer learning for nlp

    2019 arXiv

  3. [11]

    Jamjoom, and Nagwan Abdel Samee

    Umair Khan, Salabat Khan, Atif Rizwan, Ghada Atteia, Mona M. Jamjoom, and Nagwan Abdel Samee. 2022. https://doi.org/10.3390/app12105083 Aggression detection in social media from textual data using deep learning models . Applied Sciences, 12(10):5083

    2022 doi

  4. [12]

    Kowalski, Susan P

    Robin M. Kowalski, Susan P. Limber, and Patricia W. Agatston. 2014. Bullying in the digital age: A critical review and meta-analysis of cyberbullying research among youth. Psychological Bulletin, 140(4):1073--1137. Available at: https://pubmed.ncbi.nlm.nih.gov/24512111/

    2014

  5. [13]

    Ritesh Kumar, Atul Kr Ojha, Shervin Malmasi, and Marcos Zampieri. 2018. Benchmarking aggression identification in social media. In Proceedings of the first workshop on trolling, aggression and cyberbullying (TRAC-2018), pages 1--11

    2018

  6. [14]

    Ritesh Kumar, Atul Kr Ojha, Shervin Malmasi, and Marcos Zampieri. 2020. Evaluating aggression identification in social media. In Proceedings of the second workshop on trolling, aggression and cyberbullying, pages 1--5

    2020

  7. [15]

    Kai Luo, Ce Zheng, and Zhenyu Guan. 2025. Reinforced multi-modal cyberbullying detection with subgraph neural networks. International Journal of Machine Learning and Cybernetics, 16(3):2161--2180

    2025

  8. [16]

    explain thyself bully

    Krishanu Maity, Prince Jha, Raghav Jain, Sriparna Saha, and Pushpak Bhattacharyya. 2023. “explain thyself bully”: Sentiment aided cyberbullying detection with explanation. In International Conference on Document Analysis and Recognition, pages 132--148. Springer

    2023

  9. [17]

    Amel Muminovic. 2025. https://doi.org/10.5120/ijca2025925403 Moderating harm: Benchmarking large language models for cyberbullying detection in youtube comments . International Journal of Computer Applications, 187(25):1--9

    2025 doi

  10. [18]

    Nazia Nafis, Diptesh Kanojia, Naveen Saini, and Rudra Murthy. 2023. Towards safer communities: Detecting aggression and offensive language in code-mixed tweets to combat cyberbullying. In The 7th Workshop on Online Abuse and Harms (WOAH), pages 29--41

    2023

  11. [19]

    Adamu Gaston Philipo, Doreen Sebastian Sarwatt, Jianguo Ding, Mahmoud Daneshmand, and Huansheng Ning. 2025. Assessing text classification methods for cyberbullying detection on social media platforms. IEEE Transactions on Information Forensics and Security

    2025

  12. [20]

    Jing Qian, Anna Bethke, Yinyin Liu, Elizabeth Belding, and William Yang Wang. 2019. https://doi.org/10.18653/v1/D19-1482 A benchmark dataset for learning to intervene in online hate speech . In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Process...

    2019 doi

  13. [21]

    Tharindu Ranasinghe and Marcos Zampieri. 2023. https://doi.org/10.48550/arXiv.2312.03379 A text-to-text model for multilingual offensive language identification . In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2023. Accepted to Findings of IJCNLP-AACL 2023

    2023 doi arXiv

  14. [22]

    Akash Rawat, Nazia Nafis, Dnyaneshwar Bhadane, Diptesh Kanojia, and Rudra Murthy. 2023. Modelling political aggression on social media platforms. In Proceedings of the 13th workshop on computational approaches to subjectivity, sentiment, & social media analysis, pages 497--510

    2023

  15. [24]

    Chowdhury, Milica Milenkovic, Soon gyo Jung, Haewoon Kwak, and Bernard J

    Joni Salminen, Matthias Hopf, Sami A. Chowdhury, Milica Milenkovic, Soon gyo Jung, Haewoon Kwak, and Bernard J. Jansen. 2020 b . https://doi.org/10.1186/s13673-019-0205-6 Developing an online hate classifier for multiple social media platforms . Human-centric Computing and Inf...

    2020 doi

  16. [25]

    Niloofar Safi Samghabadi, Parth Patwa, Srinivas Pykl, Prerana Mukherjee, Amitava Das, and Thamar Solorio. 2020. Aggression and misogyny detection using bert: A multi-task approach. In Proceedings of the second workshop on trolling, aggression and cyberbullying, pages 126--131

    2020

  17. [26]

    Timo Schick and Hinrich Schütze. 2021. https://arxiv.org/abs/2001.07676 Exploiting cloze questions for few shot text classification and natural language inference . arXiv preprint arXiv:2001.07676

    2021 arXiv

  18. [27]

    Smith, and Ann Frisén

    Robert Slonje, Peter K. Smith, and Ann Frisén. 2013. https://doi.org/10.1016/j.chb.2012.05.024 The nature of cyberbullying, and strategies for prevention . Computers in Human Behavior, 29(1):26--32

    2013 doi

  19. [28]

    Bertie Vidgen, Alex Harris, Dong Nguyen, Rebekah Tromble, Scott Hale, and Helen Margetts. 2019. Challenges and frontiers in abusive content detection. In Proceedings of the third workshop on abusive language online. Association for Computational Linguistics

    2019

  20. [29]

    Peng Wang, Xinqiang Wang, Xiaoyan Yang, and Li Lei. 2019. https://doi.org/10.3390/sym12091536 Affective and stress consequences of cyberbullying: The mediating role of social anxiety . Symmetry, 12(9):1536

    2019 doi

  21. [30]

    Zeerak Waseem, Thomas Davidson, Dana Warmsley, and Ingmar Weber. 2017. http://arxiv.org/abs/1705.09899 Understanding abuse: A typology of abusive language detection subtasks . To appear in the proceedings of the 1st Workshop on Abusive Language Online

    2017 arXiv

  22. [31]

    Jun-Ming Xu, Kwang-Sung Jun, Xiaojin Zhu, and Amy Bellmore. 2012. Learning from bullying traces in social media. In Proceedings of the 2012 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 656--666

    2012

  23. [32]

    Peiling Yi and Arkaitz Zubiaga. 2022. Cyberbullying detection across social media platforms via platform-aware adversarial encoding. In Proceedings of the International AAAI Conference on Web and Social Media, volume 16, pages 1430--1434

    2022

  24. [33]

    Peiling Yi, Arkaitz Zubiaga, and Yunfei Long. 2025. https://doi.org/10.48550/arXiv.2501.16925 Detecting harassment and defamation in cyberbullying with emotion-adaptive training . arXiv preprint arXiv:2501.16925

    2025 doi arXiv

  25. [34]

    Marcos Zampieri, Preslav Nakov, Sara Rosenthal, Pepa Atanasova, Georgi Karadzhov, Hamdy Mubarak, Leon Derczynski, Zeses Pitenis, and Çağrı Çöltekin. 2020. https://doi.org/10.18653/v1/2020.semeval-1.188 Semeval-2020 task 12: Multilingual offensive language identification in soc...

    2020 doi

  26. [35]

    Yongchao Zhou, Andrei Ioan Muresanu, Ziwen Han, Keiran Paster, Silviu Pitis, Harris Chan, and Jimmy Ba. 2022. https://arxiv.org/abs/2211.01910 Large language models are human-level prompt engineers . arXiv preprint arXiv:2211.01910

    2022 arXiv

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.