Pith. sign in

REVIEW 3 major objections 5 minor 64 references

Data-Efficient Hate Speech Detection via Cross-Lingual Nearest Neighbor Retrieval with Limited Labeled Data

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Cross-lingual nearest-neighbor retrieval from a large multilingual pool makes hate speech detection data-efficient: with as few as 20 labeled target examples, fine-tuning on the retrieved neighbors outperforms target-language-only…

desk verdict A clean, useful empirical study of cross-lingual retrieval-augmented fine-tuning for hate speech; the main claim is credible at the average level, but the paper overstates one result and skips variance reporting, so "consistently outperforms" needs qualification. read the letter →

arxiv 2505.14272 v2 pith:NM4HKDWG submitted 2025-05-20 cs.CL cs.CYcs.MM

Faeze Ghorbanpour , Daryna Dementieva , Alexander Fraser This is my paper · ORCID
classification cs.CLcs.CYcs.MM
keywords hatespeechdetectioncross-lingualtransfernearestneighborretrievallow-resourceNLPdata-efficientfine-tuningmultilingualembeddingsF1-macroevaluationretrieval-augmentedtraining
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

The paper claims that a multilingual nearest-neighbor retrieval step can make hate speech detection work well with only a handful of labeled examples in the target language. The authors build a pool of about 265,000 labeled instances from fourteen hate speech datasets, then for each low-resource target language retrieve the closest pool instances to the small labeled training set using Euclidean distance in a multilingual embedding space. Fine-tuning a multilingual classifier on the target set plus the retrieved neighbors consistently beats fine-tuning on the target set alone across eight languages, and in most cases matches or beats the strongest existing data-efficient baseline. The most striking empirical finding is that as few as 20 retrieved instances already help, around 2,000 is near-optimal, and the largest gains appear when the target language has fewer than 50 labels.

What carries the argument

The engine is cross-lingual nearest-neighbor retrieval over a shared embedding space. Every instance in a pool of about 265,000 labeled hate speech examples from fourteen datasets is embedded with the bge-m3 multilingual encoder; for each target-language training instance, the method retrieves the top $k$ pool examples by Euclidean distance using a hierarchical navigable small world (HNSW) index. The retrieved texts are deduplicated and combined with the target training set to fine-tune XLM-T with binary cross-entropy loss, and because source and target share the same label space (hate vs. non-hate), joint fine-tuning is well-defined. This machinery carries the argument because it turns the choice of which source languages or tasks to transfer from into an automatic instance-level similarity search, and the experiments vary the number of retrieved instances from 10 to 100,000 to show where the signal saturates.

What would settle it

If random selection of the same number of cross-lingual examples, with the same hate/non-hate label balance, matches the nearest-neighbor selection's F1-macro, then similarity-based retrieval is not the source of the gains and the central claim would be falsified.

Watch

Extended reading notes

Core claim

The central claim is that fine-tuning a multilingual hate speech classifier on the union of a small target-language training set and the nearest-neighbor instances retrieved from a large multilingual pool gives higher F1-macro than fine-tuning on the target-language data alone. The authors report this consistently across eight languages, with as few as 20 retrieved instances already improving performance and with gains above 10 percentage points in some extreme low-resource settings with fewer than 50 labels. In most languages the retrieved-neighbor model also matches or beats the strongest existing data-efficient baseline, which needs 20,000 instances for intermediate English fine-tuning; the proposed method achieves that with roughly 200 to 2,000 retrieved examples. The authors also find that about 2,000 retrieved instances is near-optimal and that adding more can degrade performance, and that an MMR-based diversity filter helps in some languages when the retrieval budget is small.

Load-bearing premise

The method assumes that a small labeled set of target-language examples exists and that Euclidean distance in the multilingual embedding space is a trustworthy guide to which retrieved cross-lingual examples will actually help, a relevance signal the paper does not directly validate.

Editorial extensions

If this is right

  • With as few as 20 labeled target examples, appending retrieved cross-lingual neighbors improves F1-macro over target-only fine-tuning in every language tested.
  • Retrieving around 2,000 instances is the best operating point on average; adding more retrieved data tends to degrade performance.
  • The method removes the need to manually choose source languages or intermediate tasks, since the retrieval itself selects relevant instances, including culturally or linguistically related ones.
  • The pool is extensible without retraining, so new datasets and languages can be added and immediately used for future retrievals.
  • MMR-based deduplication improves some languages when the retrieval budget is small, by favoring diversity alongside relevance.

Reading between the lines

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

  • Because the paper does not inspect whether retrieved neighbors are topically relevant or correctly labeled, a direct test would be to measure label agreement and topic overlap of the retrieved set; this would show whether the gains come from semantic relevance or simply from extra training data.
  • The same retrieve-and-fine-tune recipe should transfer to other cross-lingual classification tasks that share a label space, such as sentiment or toxicity detection, though the paper evaluates only hate speech.
  • The pool is about two-thirds English, so rebalancing it toward target languages is an untested way to change both the size of the gains and the optimal retrieval count.
  • The rise-then-fall pattern in retrieval count (peak near 2,000) hints at overfitting to a narrow manifold of similar examples; the paper reports the trend but does not explain the mechanism.
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

3 major / 5 minor

Summary. The paper proposes a cross-lingual retrieval-augmented fine-tuning method for hate speech detection in low-resource target languages. The approach embeds a large multilingual pool of labeled hate speech datasets, retrieves nearest neighbors of a small target-language training set in bge-m3 embedding space, and fine-tunes XLM-T on the union of retrieved and target-language instances. The method is evaluated on eight languages with training sizes from 10 to 2,000 and five seeds, reporting F1-macro against a target-only baseline (Mono) and against Röttger et al. (2022). The central claim is that retrieval-augmented fine-tuning consistently outperforms Mono and, in most cases, the state of the art, even with as few as 20 retrieved instances.

Significance. If the empirical claim holds, the method is practically useful: it is simple, scalable, avoids translation or intermediate-task selection, and shows large gains in very low-resource settings. The evaluation design has notable strengths: same-language instances are excluded from retrieval, related translated/English sets are excluded for German and Turkish, test sets are fixed and unseen, and ablations compare English-only versus multilingual retrieval and MMR. However, the statistical foundation is currently incomplete: no variance or significance information is reported, and the paper's own tables contain direct counterexamples to the 'consistently outperforms' statement.

major comments (3)
  1. [Section 5 (Table 5)] The statement 'In all languages, retrieving as few as 20 instances for fine-tuning already outperforms the Mono setting' is directly contradicted by Table 5. For San20 it at training size 100, Mono is 70.96 and retrieved-20 is 70.01; for Ous19 fr at training size 20, Mono is 47.26 and retrieved-20 is 47.21; for Gahd24 de at training size 50, Mono is 57.85 and retrieved-20 is 54.53; for For19 pt at training size 50, Mono is 60.25 and retrieved-20 is 59.26. The abstract's 'consistently outperforms' is therefore too strong as stated. Please revise the claim to refer to average performance, or to specific retrieval/training sizes, or explicitly report the counterexamples.
  2. [Section 4.3 (Evaluation Details) and Table 5] All F1-macro scores are means over 5 seeds, but no standard deviations, confidence intervals, or significance tests are reported anywhere in the paper. In the 10-50 training-example regime, seed variance is typically large, so the reader cannot tell whether the reported improvements over Mono, or the differences against Röttger et al. (2022), are reliable or within noise. This is load-bearing for the central empirical claim. Please provide per-seed results, or at least standard deviations plus a paired significance test across the five seeds (e.g., Wilcoxon signed-rank) for the main comparisons.
  3. [Section 5.2 (Table 2)] The comparison between English-only and multilingual retrieval is used to conclude that 'incorporating even a small amount of multilingual data can be beneficial,' but the observed differences are often tiny and no variance is reported. For example, Bas19 es AVG at 20 retrieved is identical (67.27) in both settings, and Ous19 fr AVG at 20 retrieved differs by only 0.56. Without per-seed statistics or a significance test, this secondary conclusion is not supported and should be either substantiated or softened.
minor comments (5)
  1. [Abstract and Section 1] The abstract says 'retrieving as small as 200 instances' while the body repeatedly emphasizes 'as few as 20 instances'; please harmonize the wording.
  2. [Table 4] The language column contains the typo 'Purtegues' for Portuguese; please fix.
  3. [Section 3.2] The symbol R is used both for the desired number of retrieved instances and for the retrieved set; please use distinct notation, for example R for the count and \mathcal{R} for the set.
  4. [Appendix G (Table 7)] Table 7 is hard to parse because the language row labels are not cleanly aligned with the sub-tables; please reformat so that each language block is clearly separated and the with/without MMR columns are visually distinct.
  5. [Appendix C] The statement that 'All models and datasets used in this study are licensed for academic research purposes' is imprecise, since some licenses (e.g., CC BY-NC-SA) restrict commercial use; consider rewording to avoid overstating licensing compatibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the evaluation is held-out and the retrieval pipeline does not use target test labels, so the reported gains are empirical rather than constructed.

full rationale

The paper's claimed result—retrieval-augmented fine-tuning outperforms target-only training—is an empirical comparison, not a derivation from an assumed outcome. The target-language test set is fixed and unseen during retrieval and fine-tuning (Section 4.1), and instances from the target language are excluded from the retrieval pool, as are related translated sets for the German and Turkish targets. Retrieval selection uses Euclidean distance in fixed bge-m3 embeddings and does not use labels, so the retrieved set is not constructed to match the test labels. Hyperparameters (learning rate 5e-5, 10/5 epochs, MMR lambda = 0.5) are fixed a priori. The claim that 2,000 retrieved instances is near-optimal is a post-hoc description of the plotted curves, not a fitted parameter reused as a prediction. The few self-citations (Dementieva et al. 2024; Bigoulaeva et al. 2021, 2022) appear only as related-work context and are not load-bearing for the method. The inconsistency between the '20 retrieved beats Mono' statement and specific cells of Table 5 is a statistical-correctness concern (no per-seed variance or significance tests), not circularity, because the compared values are independently measured on held-out test data.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The method introduces no new entities, forces, or conserved quantities. The free-parameter count is low; the main hand-chosen value is the MMR lambda. The load-bearing assumptions are the compatibility of labels across datasets and the cross-lingual validity of the embedding space, both acknowledged or unexamined in the paper.

free parameters (1)
  • MMR lambda = 0.5
    Hand-set trade-off between relevance and diversity in Section 6; not tuned per language or dataset, and not fit to test performance.
assumptions (4)
  • standard math FAISS HNSW approximate nearest neighbor search approximates exact kNN well for the 265k pool
    Relied on in Section 3.2 for retrieval; no analysis of approximation error is given.
  • domain assumption Hate vs non-hate labels are compatible across the 14 datasets
    Section 3.3 states joint training on combined source and target data is well-defined and coherent; the paper acknowledges cultural and annotation inconsistencies in the Limitations.
  • domain assumption Hate speech semantics transfer across languages via multilingual embeddings (bge-m3)
    The whole retrieval pipeline (Sections 3.1 to 3.2) assumes Euclidean distance in embedding space is a valid cross-lingual relevance signal; not directly validated.
  • domain assumption A small labeled target set (10 to 2,000 instances) is available
    Stated as the paper's operating assumption in the Abstract and in the Limitations; may fail in extremely low-resource settings.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Data-Efficient Hate Speech Detection via Cross-Lingual Nearest Neighbor Retrieval with Limited Labeled Data." pith.science (2026). https://pith.science/paper/NM4HKDWG

@misc{pith2026250514272,
  author       = {Pith},
  title        = {Pith review of: Data-Efficient Hate Speech Detection via Cross-Lingual Nearest Neighbor Retrieval with Limited Labeled Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NM4HKDWG}},
  note         = {Machine review of arXiv:2505.14272}
}
read the original abstract

Considering the importance of detecting hateful language, labeled hate speech data is expensive and time-consuming to collect, particularly for low-resource languages. Prior work has demonstrated the effectiveness of cross-lingual transfer learning and data augmentation in improving performance on tasks with limited labeled data. To develop an efficient and scalable cross-lingual transfer learning approach, we leverage nearest-neighbor retrieval to augment minimal labeled data in the target language, thereby enhancing detection performance. Specifically, we assume access to a small set of labeled training instances in the target language and use these to retrieve the most relevant labeled examples from a large multilingual hate speech detection pool. We evaluate our approach on eight languages and demonstrate that it consistently outperforms models trained solely on the target language data. Furthermore, in most cases, our method surpasses the current state-of-the-art. Notably, our approach is highly data-efficient, retrieving as small as 200 instances in some cases while maintaining superior performance. Moreover, it is scalable, as the retrieval pool can be easily expanded, and the method can be readily adapted to new languages and tasks. We also apply maximum marginal relevance to mitigate redundancy and filter out highly similar retrieved instances, resulting in improvements in some languages.

Figures

Figures reproduced from arXiv: 2505.14272 by the authors.

Figure 1
Figure 1. Overview of the proposed method. Given a small number of examples from a target language, we search [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Performance across different numbers of retrieved instances (10 to 100,000, log-scaled) for four languages. [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Sankey diagram of the distribution of the top [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Sankey diagrams of the distribution of the top four retrieved source tasks (left), languages (middle), and [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

64 extracted references · 26 canonical work pages

  1. [1]

    AlEmadi and Wajdi Zaghouani

    Maryam M. AlEmadi and Wajdi Zaghouani. 2024. https://aclanthology.org/2024.legal-1.10/ Emotional toll and coping strategies: Navigating the effects of annotating hate speech data . In Proceedings of the Workshop on Legal and Ethical Issues in Human Language Technologies @ LREC-COLING 2024, pages 66--72, Torino, Italia. ELRA and ICCL

    work page 2024

  2. [3]

    Sai Saketh Aluru, Binny Mathew, Punyajoy Saha, and Animesh Mukherjee. 2020 b . https://doi.org/10.1007/978-3-030-67670-4_26 Deep learning models for multilingual hate speech detection . In Machine Learning and Knowledge Discovery in Databases. ECML PKDD 2020, volume 12461 of Lecture Notes in Computer Science, pages 528--544. Springer

    doi:10.1007/978-3-030-67670-4_26 2020

  3. [4]

    Dimosthenis Antypas and Jose Camacho-Collados. 2023. https://doi.org/10.18653/v1/2023.woah-1.25 Robust hate speech detection in social media: A cross-dataset empirical evaluation . In The 7th Workshop on Online Abuse and Harms (WOAH), pages 231--242, Toronto, Canada. ACL

    doi:10.18653/v1/2023.woah-1.25 2023

  4. [5]

    Francesco Barbieri, Luis Espinosa Anke, and Jose Camacho-Collados. 2022. https://aclanthology.org/2022.lrec-1.27/ XLM - T : Multilingual language models in T witter for sentiment analysis and beyond . In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 258--266, Marseille, France. ELRA

    work page 2022

  5. [6]

    Valerio Basile, Cristina Bosco, Elisabetta Fersini, Debora Nozza, Viviana Patti, Francisco Manuel Rangel Pardo, Paolo Rosso, and Manuela Sanguinetti. 2019. https://doi.org/10.18653/v1/S19-2007 S em E val-2019 task 5: Multilingual detection of hate speech against immigrants and women in T witter . In Proceedings of the 13th International Workshop on Semant...

    doi:10.18653/v1/s19-2007 2019

  6. [7]

    Djamila Romaissa Beddiar, Md Saroar Jahan, and Mourad Oussalah. 2021. https://doi.org/https://doi.org/10.1016/j.osnem.2021.100153 Data expansion using back translation and paraphrasing for hate speech detection . Online Social Networks and Media, 24:100153

    arXiv 2021

  7. [8]

    Irina Bigoulaeva, Viktor Hangya, and Alexander Fraser. 2021. https://aclanthology.org/2021.ltedi-1.3/ Cross-lingual transfer learning for hate speech detection . In Proceedings of the First Workshop on Language Technology for Equality, Diversity and Inclusion, pages 15--25, Kyiv. ACL

    work page 2021

  8. [9]

    Irina Bigoulaeva, Viktor Hangya, Iryna Gurevych, and Alexander Fraser. 2022. https://arxiv.org/abs/2201.05922 Addressing the challenges of cross-lingual hate speech detection . arXiv preprint arXiv:2201.05922

    arXiv 2022

Show all 64 references
  1. [10]

    Jaime Carbonell and Jade Goldstein. 1998. https://doi.org/10.1145/290941.291025 The use of mmr, diversity-based reranking for reordering documents and producing summaries . In Proceedings of the 21st Annual International ACM SIGIR Conference on Research and Development in Info...

    1998

  2. [11]

    Tommaso Caselli, Valerio Basile, Jelena Mitrovi \'c , Inga Kartoziya, and Michael Granitzer. 2020. https://aclanthology.org/2020.lrec-1.760/ I feel offended, don`t be abusive! implicit/explicit messages in offensive and abusive language . In Proceedings of the Twelfth Language...

    2020

  3. [12]

    Jianlyu Chen, Shitao Xiao, Peitian Zhang, Kun Luo, Defu Lian, and Zheng Liu. 2024. https://doi.org/10.18653/v1/2024.findings-acl.137 M 3-embedding: Multi-linguality, multi-functionality, multi-granularity text embeddings through self-knowledge distillation . In Findings of the...

    2024 doi

  4. [13]

    Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzm \'a n, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. https://doi.org/10.18653/v1/2020.acl-main.747 Unsupervised cross-lingual representation learning ...

    2020 doi

  5. [14]

    Devleena Das and Vivek Khetan. 2024. https://doi.org/10.18653/v1/2024.emnlp-main.1132 DEFT - UCS : Data efficient fine-tuning for pre-trained language models via unsupervised core-set selection for text-editing . In Proceedings of the 2024 Conference on Empirical Methods in Na...

    2024 doi

  6. [15]

    Daryna Dementieva, Valeriia Khylenko, Nikolay Babakov, and Georg Groh. 2024. https://doi.org/10.18653/v1/2024.woah-1.19 Toxicity classification in U krainian . In Proceedings of the 8th Workshop on Online Abuse and Harms (WOAH 2024), pages 244--255, Mexico City, Mexico. ACL

    2024 doi

  7. [16]

    Matthijs Douze, Alexandr Guzhva, Chengqi Deng, Jeff Johnson, Gergely Szilvasy, Pierre-Emmanuel Mazaré, Maria Lomeli, Lucas Hosseini, and Hervé Jégou. 2024. https://arxiv.org/abs/2401.08281 The faiss library . arXiv preprint arXiv:2401.08281

    2024 arXiv

  8. [17]

    Mai ElSherief, Caleb Ziems, David Muchlinski, Vaishnavi Anupindi, Jordyn Seybolt, Munmun De Choudhury, and Diyi Yang. 2021. https://doi.org/10.18653/v1/2021.emnlp-main.29 Latent hatred: A benchmark for understanding implicit hate speech . In Proceedings of the 2021 Conference ...

    2021 doi

  9. [18]

    Anderson Almeida Firmino, Cl\' a udio de Souza Baptista, and Anselmo Cardoso de Paiva. 2024. https://doi.org/10.1016/j.eswa.2023.121115 Improving hate speech detection using cross-lingual learning . Expert Syst. Appl., 235(C)

    2024

  10. [19]

    Paula Fortuna and S\' e rgio Nunes. 2018. https://doi.org/10.1145/3232676 A survey on automatic detection of hate speech in text . ACM Computing Surveys, 51(4)

    2018 doi

  11. [20]

    Paula Fortuna, Jo \ a o Rocha da Silva, Juan Soler-Company, Leo Wanner, and S \'e rgio Nunes. 2019. https://doi.org/10.18653/v1/W19-3510 A hierarchically-labeled P ortuguese hate speech dataset . In Proceedings of the Third Workshop on Abusive Language Online, pages 94--104, F...

    2019 doi

  12. [21]

    Antigoni Founta, Constantinos Djouvas, Despoina Chatzakou, Ilias Leontiadis, Jeremy Blackburn, Gianluca Stringhini, Athena Vakali, Michael Sirivianos, and Nicolas Kourtellis. 2018. https://doi.org/10.1609/icwsm.v12i1.14991 Large scale crowdsourcing and characterization of twit...

    2018 doi

  13. [22]

    Hassan Gharoun, Fereshteh Momenifar, Fang Chen, and Amir H. Gandomi. 2024. https://doi.org/10.1145/3659943 Meta-learning approaches for few-shot learning: A survey of recent advances . ACM Comput. Surv., 56(12)

    2024 doi

  14. [23]

    Janis Goldzycher, Moritz Preisig, Chantal Amrhein, and Gerold Schneider. 2023. https://doi.org/10.18653/v1/2023.woah-1.19 Evaluating the effectiveness of natural language inference for hate speech detection in languages with limited labeled data . In The 7th Workshop on Online...

    2023 doi

  15. [24]

    Janis Goldzycher, Paul R \"o ttger, and Gerold Schneider. 2024. https://doi.org/10.18653/v1/2024.naacl-long.248 Improving adversarial data collection by supporting annotators: Lessons from GAHD , a G erman hate speech dataset . In Proceedings of the 2024 Conference of the Nort...

    2024 doi

  16. [25]

    Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Ming-Wei Chang. 2020. https://proceedings.mlr.press/v119/guu20a/guu20a.pdf Realm: retrieval-augmented language model pre-training . In Proceedings of the 37th International Conference on Machine Learning, ICML'20. JMLR.org

    2020

  17. [26]

    Xiaochuang Han and Yulia Tsvetkov. 2022. https://arxiv.org/abs/2205.12600 Orca: Interpreting prompted language models via locating supporting data evidence in the ocean of pretraining data . arXiv preprint arXiv:2205.12600

    2022 arXiv

  18. [27]

    Ehtesham Hashmi, Sule Yildirim Yayilgan, and Mohamed Abomhara. 2025. https://doi.org/10.1007/s40747-025-01808-w Metalinguist: Enhancing hate speech detection with cross-lingual meta-learning . Complex & Intelligent Systems, 11(1):179

    2025 doi

  19. [28]

    Pengcheng He, Jianfeng Gao, and Weizhu Chen. 2021. https://arxiv.org/abs/2111.09543 Debertav3: Improving deberta using electra-style pre-training with gradient-disentangled embedding sharing . arXiv preprint arXiv:2111.09543

    2021 arXiv

  20. [29]

    Smith, Hannaneh Hajishirzi, and Pradeep Dasigi

    Hamish Ivison, Noah A. Smith, Hannaneh Hajishirzi, and Pradeep Dasigi. 2023. https://doi.org/10.18653/v1/2023.findings-acl.576 Data-efficient finetuning using cross-task nearest neighbors . In Findings of the Association for Computational Linguistics: ACL 2023, pages 9036--906...

    2023 doi

  21. [30]

    Jeff Johnson, Matthijs Douze, and Hervé Jégou. 2021. https://doi.org/10.1109/TBDATA.2019.2921572 Billion-scale similarity search with gpus . IEEE Transactions on Big Data, 7(3):535--547

    2021

  22. [31]

    Brendan Kennedy, Xisen Jin, Aida Mostafazadeh Davani, Morteza Dehghani, and Xiang Ren. 2020 a . https://doi.org/10.18653/v1/2020.acl-main.483 Contextualizing hate speech classifiers with post-hoc explanation . In Proceedings of the 58th Annual Meeting of the Association for Co...

    2020 doi

  23. [32]

    Kennedy, Geoff Bacon, Alexander Sahn, and Claudia von Vacano

    Chris J. Kennedy, Geoff Bacon, Alexander Sahn, and Claudia von Vacano. 2020 b . https://arxiv.org/abs/2009.10277 Constructing interval variables via faceted rasch measurement and multitask deep learning: a hate speech application . arXiv preprint arXiv:2009.10277

    2020 arXiv

  24. [33]

    Urvashi Khandelwal, Omer Levy, Dan Jurafsky, Luke Zettlemoyer, and Mike Lewis. 2020. https://openreview.net/forum?id=HgxsSjxZt7 Generalization through memorization: Nearest neighbor language models . In Proceedings of the 8th International Conference on Learning Representations (ICLR)

    2020

  25. [34]

    Bill Yuchen Lin, Kangmin Tan, Chris Miller, Beiwen Tian, and Xiang Ren. 2022. https://proceedings.neurips.cc/paper_files/paper/2022/file/8a0d3ae989a382ce6e50312bc35bf7e1-Paper-Conference.pdf Unsupervised cross-task generalization via retrieval augmentation . In Advances in Neu...

    2022

  26. [35]

    Malkov and D

    Yu A. Malkov and D. A. Yashunin. 2020. https://doi.org/10.1109/TPAMI.2018.2889473 Efficient and robust approximate nearest neighbor search using hierarchical navigable small world graphs . IEEE Transactions on Pattern Analysis and Machine Intelligence, 42(4):824–836

    2020

  27. [36]

    Thomas Mandl, Sandip Modha, Gautam Kishore Shahi, Hiren Madhu, Shrey Satapara, Prasenjit Majumder, Johannes Schäfer, Tharindu Ranasinghe, Marcos Zampieri, Durgesh Nandini, and Amit Kumar Jaiswal. 2021. https://ceur-ws.org/Vol-3159/T1-1.pdf Overview of the hasoc subtrack at fir...

    2021

  28. [37]

    Binny Mathew, Punyajoy Saha, Seid Muhie Yimam, Chris Biemann, Pawan Goyal, and Animesh Mukherjee. 2021. https://ojs.aaai.org/index.php/AAAI/article/view/17745 Hatexplain: A benchmark dataset for explainable hate speech detection . In Proceedings of the AAAI Conference on Artif...

    2021

  29. [38]

    Ayme Arango Monnar, Jorge Pérez Rojas, and Barbara Poblete. 2024. https://doi.org/10.1371/journal.pone.0306521 Cross-lingual hate speech detection using domain-specific word embeddings . PLOS ONE, 19(7):e0306521

    2024 doi

  30. [39]

    Marzieh Mozafari, Reza Farahbakhsh, and Noel Crespi. 2022. https://doi.org/10.1109/ACCESS.2022.3147588 Cross-lingual few-shot hate speech and offensive language detection using meta learning . IEEE Access, 10:14880--14896

    2022

  31. [40]

    Masayasu Muraoka, Bishwaranjan Bhattacharjee, Michele Merler, Graeme Blackwood, Yulong Li, and Yang Zhao. 2023. https://doi.org/10.1145/3581783.3611992 Cross-lingual transfer of large language model by visually-derived supervision toward low-resource languages . In Proceedings...

    2023

  32. [41]

    Rachna Narula and Poonam Chaudhary. 2024. https://doi.org/10.1007/s13278-024-01401-y A comprehensive review on detection of hate speech for multi-lingual data . Social Network Analysis and Mining, 14(244)

    2024 doi

  33. [42]

    Nedjma Ousidhoum, Zizheng Lin, Hongming Zhang, Yangqiu Song, and Dit-Yan Yeung. 2019. https://doi.org/10.18653/v1/D19-1474 Multilingual and multi-aspect hate speech analysis . In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th...

    2019 doi

  34. [43]

    Endang Wahyu Pamungkas, Valerio Basile, and Viviana Patti. 2021. https://doi.org/10.1016/j.ipm.2021.102544 A joint learning approach with knowledge injection for zero-shot cross-lingual hate speech detection . Information Processing and Management, 58(4)

    2021

  35. [44]

    Endang Wahyu Pamungkas and Viviana Patti. 2019. https://doi.org/10.18653/v1/P19-2051 Cross-domain and cross-lingual abusive language detection: A hybrid approach with deep learning and a multilingual lexicon . In Proceedings of the 57th Annual Meeting of the Association for Co...

    2019 doi

  36. [45]

    Marinela Parovic, Alan Ansell, Ivan Vuli \'c , and Anna Korhonen. 2023. https://doi.org/10.18653/v1/2023.findings-acl.13 Cross-lingual transfer with target language-ready task adapters . In Findings of the Association for Computational Linguistics: ACL 2023, pages 176--193, To...

    2023 doi

  37. [46]

    Andraž Pelicon, Ravi Shekhar, Blaž Škrlj, Matthew Purver, and Senja Pollak. 2021. https://doi.org/10.7717/peerj-cs.559 Investigating cross-lingual training for offensive language detection . PeerJ Computer Science, 7:e559

    2021 doi

  38. [47]

    Trinh Pham, Khoi Le, and Anh Tuan Luu. 2024. https://doi.org/10.18653/v1/2024.acl-long.174 U ni B ridge: A unified approach to cross-lingual transfer learning for low-resource languages . In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistic...

    2024 doi

  39. [48]

    Fabio Poletto, Valerio Basile, Manuela Sanguinetti, Cristina Bosco, and Viviana Patti. 2021. https://doi.org/10.1007/s10579-020-09502-8 Resources and benchmark corpora for hate speech detection: a systematic review . Language Resources and Evaluation, 55(2):477--523

    2021 doi

  40. [49]

    Garima Pruthi, Frederick Liu, Satyen Kale, and Mukund Sundararajan. 2020. https://proceedings.neurips.cc/paper/2020/file/e6385d39ec9394f2f3a354d9d2b88eec-Paper.pdf Estimating training data influence by tracing gradient descent . In Advances in Neural Information Processing Sys...

    2020

  41. [50]

    Nils Reimers and Iryna Gurevych. 2020. https://doi.org/10.18653/v1/2020.emnlp-main.365 Making monolingual sentence embeddings multilingual using knowledge distillation . In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 45...

    2020 doi

  42. [51]

    Paul R \"o ttger, Debora Nozza, Federico Bianchi, and Dirk Hovy. 2022. https://doi.org/10.18653/v1/2022.emnlp-main.383 Data-efficient strategies for expanding hate speech detection into under-resourced languages . In Proceedings of the 2022 Conference on Empirical Methods in N...

    2022 doi

  43. [52]

    Paul R \"o ttger, Bertie Vidgen, Dong Nguyen, Zeerak Waseem, Helen Margetts, and Janet Pierrehumbert. 2021. https://doi.org/10.18653/v1/2021.acl-long.4 H ate C heck: Functional tests for hate speech detection models . In Proceedings of the 59th Annual Meeting of the Associatio...

    2021 doi

  44. [53]

    Sumegh Roychowdhury and Vikram Gupta. 2023. https://doi.org/10.18653/v1/2023.findings-eacl.9 Data-efficient methods for improving hate speech detection . In Findings of the Association for Computational Linguistics: EACL 2023, pages 125--132, Dubrovnik, Croatia. ACL

    2023 doi

  45. [54]

    Manuela Sanguinetti, Gloria Comandini, Elisa Di Nuovo, Simona Frenda, Marco Stranisci, Cristina Bosco, Tommaso Caselli, Viviana Patti, and Irene Russo. 2020. https://ceur-ws.org/Vol-2765/paper162.pdf Haspeede 2 @ evalita2020: Overview of the evalita 2020 hate speech detection ...

    2020

  46. [55]

    Weijia Shi, Julian Michael, Suchin Gururangan, and Luke Zettlemoyer. 2022. https://doi.org/10.18653/v1/2022.emnlp-main.214 Nearest neighbor zero-shot inference . In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 3254--3265, Abu Dh...

    2022 doi

  47. [56]

    Cagri Toraman, Furkan S ahinu c , and Eyup Yilmaz. 2022. https://aclanthology.org/2022.lrec-1.238/ Large-scale hate speech detection with cross-domain transfer . In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 2215--2225, Marseille, France. ELRA

    2022

  48. [57]

    Bertie Vidgen, Dong Nguyen, Helen Margetts, Patricia Rossini, and Rebekah Tromble. 2021 a . https://doi.org/10.18653/v1/2021.naacl-main.182 Introducing CAD : the contextual abuse dataset . In Proceedings of the 2021 Conference of the North American Chapter of the Association f...

    2021 doi

  49. [58]

    Bertie Vidgen, Tristan Thrush, Zeerak Waseem, and Douwe Kiela. 2021 b . https://doi.org/10.18653/v1/2021.acl-long.132 Learning from the worst: Dynamically generated datasets to improve online hate detection . In Proceedings of the 59th Annual Meeting of the Association for Com...

    2021 doi

  50. [59]

    Michael Wiegand, Josef Ruppenhofer, and Thomas Kleinbauer. 2019. https://doi.org/10.18653/v1/N19-1060 D etection of A busive L anguage: the P roblem of B iased D atasets . In Proceedings of the 2019 Conference of the North A merican Chapter of the Association for Computational...

    2019 doi

  51. [60]

    Richard Ashby Wilson. 2019. https://doi.org/10.2139/ssrn.3465225 The digital ethnography of law: Studying online hate speech online and offline . Journal of Legal Anthropology, 3(1):1--20

    2019 doi

  52. [61]

    Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Clement Delangue, Anthony Moi, Pierric Cistac, Tim Rault, Remi Louf, Morgan Funtowicz, Joe Davison, Sam Shleifer, Patrick von Platen, Clara Ma, Yacine Jernite, Julien Plu, Canwen Xu, Teven Le Scao, Sylvain Gugger, Mari...

    2020 doi

  53. [62]

    Wenjie Yin and Arkaitz Zubiaga. 2021. https://doi.org/10.7717/peerj-cs.598 Towards generalisable hate speech detection: a review on obstacles and solutions . PeerJ Computer Science, 7:e598

    2021 doi

  54. [63]

    Li Zhou, Laura Cabello, Yong Cao, and Daniel Hershcovich. 2023. https://doi.org/10.18653/v1/2023.c3nlp-1.2 Cross-cultural transfer learning for C hinese offensive language detection . In Proceedings of the First Workshop on Cross-Cultural Considerations in NLP (C3NLP), pages 8...

    2023 doi

  55. [64]

    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...

  56. [65]

    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 gl...

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.