Elamur Research⌕
☰
home / papers / 2025 / Pragmatic Inference Chain (PIC) Improving LLMs' Reasoning of Authentic Implicit Toxic Language

Pragmatic Inference Chain (PIC) Improving LLMs' Reasoning of Authentic Implicit Toxic Language

Summary

The authors examine how LLMs recognize implicit toxic language. In real online conversations, such language is disguised to evade censorship. Understanding it requires a chain of pragmatic inferences rather than purely logical reasoning. For evaluation, they compiled a Chinese dataset of gender-targeted "post–comment" pairs from Weibo and RedNote. Expert linguists annotated these pairs and, for a subset of the toxic examples, wrote down step-by-step reasoning. The authors took Sperber and Wilson's Relevance Theory as their basis and, drawing on their experiments, turned its stages into a prompting method, Pragmatic Inference Chain (PIC). Several PIC variants were compared with five baseline prompts, including CoT and rule-based, on GPT-4o, Llama-3.1-70B-Instruct, DeepSeek-v2.5, DeepSeek-v3 and QwQ32b. Without PIC, the models barely exceed chance level. With PIC, their performance improves by 12–20%, and their reasoning becomes more explicit and coherent. The work shows that LLMs' weaknesses in pragmatics can be partially remedied through in-context learning. The method is likely applicable to other pragmatic inference tasks as well, such as understanding humor or metaphors.

Key points

  • The data were collected using 55 gender-related keywords. These words were taken from Zhihu users' self-reports, meaning that no explicitly toxic terms are among them. For each keyword, 10 posts were collected, and for each post, the 10 most popular comments. After manual cleaning, 4000 pairs remained, of which 3097 were used in the experiments.
  • The annotation was carried out by two linguist annotators specializing in pragmatics. Each pair was assigned to one of four classes: non-toxic, toxicity toward women, toxicity toward men, anti-toxicity. For 400 toxic texts, the annotators manually wrote inference chains following Relevance Theory.
  • PIC was tested in four variants: one-shot, step-by-step instructions only, instructions + 3 PIC examples, and instructions + rules. Baseline prompts: zero-shot, three-shot, CoT, rule, rule + CoT.
  • PIC yields a 12–20% performance gain over the baseline prompts. The main effect comes from the step-by-step instructions themselves, while additional examples change almost nothing.
  • The authors propose distilling or fine-tuning models on machine-generated PICs in future work. However, such chains are sometimes incomplete: they do not account for all linguistic cues, even though their conclusion is correct.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Topics Language models & pretraining Reasoning Bias, fairness & safety

Cite

Cite as Xi Chen and Shuo Wang. 2025. Pragmatic Inference Chain (PIC) Improving LLMs' Reasoning of Authentic Implicit Toxic Language. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2025.emnlp-main.296 · arXiv 2503.01539

@inproceedings{chen2025pragmatic,
  title = {{Pragmatic Inference Chain (PIC) Improving LLMs' Reasoning of Authentic Implicit Toxic Language}},
  author = {Xi Chen and Shuo Wang},
  booktitle = {Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing},
  year = {2025},
  doi = {10.18653/v1/2025.emnlp-main.296},
  eprint = {2503.01539},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/2025.emnlp-main.296}
}

References

80 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Josh Achiam, Steven Adler, Sandhini Agarwal, Lama Ahmad, Ilge Akkaya, Florencia Leoni Aleman, Diogo Almeida, Janko Altenschmidt, Sam Altman, Shyamal Anadkat, et al. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774. our card
  2. Abdullah Albanyan and Eduardo Blanco. 2022. Pinpointing Fine-Grained Relationships between Hateful Tweets and Replies. Proceedings of the AAAI Conference on Artificial Intelligence, 36(10):10418–10426. Number: 10. find
  3. Abdullah Albanyan, Ahmed Hassan, and Eduardo Blanco. 2023. Not All Counterhate Tweets Elicit the Same Replies: A Fine-Grained Analysis. In Proceedings of the 12th Joint Conference on Lexical and Computational Semantics (*SEM 2023), pages 71– 88, Toronto, Canada. Association for Computational Linguis… find
  4. Mohammad Atari, Mona J. Xue, Peter S. Park, Damián Blasi, and Joseph Henrich. 2023. Which Humans? find
  5. Chiara Barattieri di SanPietro, Federico Frau, Veronica Mangiaterra, and Valentina Bambini. 2023. The pragmatic profile of ChatGPT: Assessing the communicative skills of a conversational agent. find
  6. Andrew Blair-Stanek, Nils Holzenberger, and Benjamin Van Durme. 2023. Can gpt-3 perform statutory reasoning? In Proceedings of the Nineteenth International Conference on Artificial Intelligence and Law, ICAIL ’23, page 22–31, New York, NY, USA. Association for Computing Machinery. our card
  7. Jan Blommaert. 2005. Discourse: A Critical Introduction. Key Topics in Sociolinguistics. Cambridge University Press, Cambridge. find
  8. Tolga Bolukbasi, Kai-Wei Chang, James Y Zou, Venkatesh Saligrama, and Adam T Kalai. 2016. Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddings. In Advances in Neural Information Processing Systems, volume 29. Curran Associates, Inc. our card
  9. Ioana Buhnila, Georgeta Cislaru, and Amalia Todirascu. 2024. Chain-of-MetaWriting: Linguistic and Textual Analysis of How Small Language Models Write Young Students Texts. ArXiv:2412.14986 [cs]. find
  10. Judith Butler. 2007. Gender trouble: feminism and the subversion of identity. Routledge classics. Routledge, New York. find
  11. Tommaso Caselli, Valerio Basile, Jelena Mitrović, and Michael Granitzer. 2021. HateBERT: Retraining BERT for abusive language detection in English. In Proceedings of the 5th Workshop on Online Abuse and Harms (WOAH 2021), pages 17–25, Online. Association for Computational Linguistics. our card
  12. Xi Chen and Lucien Brown. 2024. L2 Pragmatic Development in Constructing and Negotiating Contextual Meanings. Applied Linguistics, page amae049. find
  13. Xi Chen and Jungmin Lee. 2021. The relationship between stereotypical meaning and contextual meaning of Korean honorifics. Journal of Pragmatics, 171:118–130. find
  14. Patricia W Cheng and Keith J Holyoak. 1985. Pragmatic reasoning schemas. Cognitive Psychology, 17(4):391–416. find
  15. James Chua and Owain Evans. 2025. Are deepseek r1 and other reasoning models more faithful? find
  16. DeepSeek-AI, Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, Damai Dai, Daya Guo, Dejian Yang, Deli Chen, Dongjie Ji, Erhang Li, Fangyun Lin, Fucong Dai, Fuli Luo, Guangbo Hao, Guanting Chen, Guowei Li, H. Zhang, Han Bao, H… find
  17. Jwala Dhamala, Tony Sun, Varun Kumar, Satyapriya Krishna, Yada Pruksachatkun, Kai-Wei Chang, and Rahul Gupta. 2021. Bold: Dataset and metrics for measuring biases in open-ended language generation. In Proceedings of the 2021 ACM conference on fairness, accountability, and transparency, pages 862–872… find
  18. Lucas Dixon, John Li, Jeffrey Sorensen, Nithum Thain, and Lucy Vasserman. 2018. Measuring and Mitigating Unintended Bias in Text Classification. In Proceedings of the 2018 AAAI/ACM Conference on AI, Ethics, and Society, pages 67–73, New Orleans LA USA. ACM. find
  19. Cho Hyeonsu, Dooyoung Kim, and Youngjoong Ko. 2025. DAPI: Domain adaptive toxicity probe vector intervention, for fine-grained detoxification. In Findings of the Association for Computational Linguistics: ACL 2025, pages 15059–15069, Vienna, Austria. Association for Computational Linguistics. find
  20. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, et al. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783. find
  21. Martín Iglesias, Oscar Araque, and Carlos Á. Iglesias. 2023. A toxic style transfer method based on the delete–retrieve–generate framework exploiting toxic lexicon semantic similarity. Applied Sciences, 13(15). find
  22. Chi-Hé Elder. 2024. Pragmatic Inference: Misunderstandings, Accountability, Deniability. Cambridge University Press. Google-Books-ID: okn8EAAAQBAJ. find
  23. Vaibhav Jain and Mina Naghshnejad. 2021. Entity at SemEval-2021 task 5: Weakly supervised token labelling for toxic spans detection. In Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021), pages 935–940, Online. Association for Computational Linguistics. find
  24. Mai ElSherief, Caleb Ziems, David Muchlinski, Vaish-navi Anupindi, Jordyn Seybolt, Munmun De Choudhury, and Diyi Yang. 2021. Latent hatred: A benchmark for understanding implicit hate speech. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 345–363, On… our card
  25. Yuanheng Fang, Guoqing Chao, Wenqiang Lei, Shaobo Li, and Dianhui Chu. 2025. CDW-CoT: Clustered Distance-Weighted Chain-of-Thoughts Reasoning. ArXiv:2501.12226 [cs]. — [arXiv:2501.12226](https://arxiv.org/abs/2501.12226) find
  26. Samuel Gehman, Suchin Gururangan, Maarten Sap, Yejin Choi, and Noah A. Smith. 2020. Realtoxicityprompts: Evaluating neural toxic degeneration in language models. our card
  27. Erving Goffman. 1981. Forms of Talk. University of Pennsylvania Press. Google-Books-ID: Z3bvx_T4Zu8C. find
  28. H. P. Grice. 1975. Logic and Conversation. Brill. Pages: 41-58 Section: Speech Acts. find
  29. Thomas Hartvigsen, Saadia Gabriel, Hamid Palangi, Maarten Sap, Dipankar Ray, and Ece Kamar. 2022. ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1:… our card
  30. Jennifer Hu, Sammy Floyd, Olessia Jouravlev, Evelina Fedorenko, and Edward Gibson. 2023. A fine-grained comparison of pragmatic language understanding in humans and language models. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pa… our card
  31. Xin Huang, Tarun Kumar Vangani, Zhengyuan Liu, Bowei Zou, and Ai Ti Aw. 2025. AdaCoT: Rethinking Cross-Lingual Factual Reasoning through Adaptive Chain-of-Thought. ArXiv:2501.16154 [cs]. — [arXiv:2501.16154](https://arxiv.org/abs/2501.16154) find
  32. Shihao Ji, Zihui Song, Fucheng Zhong, Jisen Jia, Zhaobo Wu, Zheyi Cao, and Tianhao Xu. 2025. MyGO Multiplex CoT: A Method for Self-Reflection in Large Language Models via Double Chain of Thought Thinking. ArXiv:2501.13117 [cs]. — [arXiv:2501.13117](https://arxiv.org/abs/2501.13117) find
  33. Hoang Anh Just, Mahavir Dabas, Lifu Huang, Ming Jin, and Ruoxi Jia. 2024. Dipt: Enhancing llm reasoning through diversified perspective-taking. find
  34. Ecem Kavaz, Anna Puig, Inmaculada Rodriguez, Mariona Taule, and Montserrat Nofre. 2021. Data Visualization for Supporting Linguists in the Analysis of Toxic Messages. find
  35. Yumin Kim and Hwanhee Lee. 2025. Selective demonstration retrieval for improved implicit hate speech detection. find
  36. Andrew Konya, Aviv Ovadya, Kevin Feng, Quan Ze Chen, Lisa Schirch, Colin Irwin, and Amy X. Zhang. 2024. Chain of Alignment: Integrating Public Will with Expert Intelligence for Language Model Alignment. ArXiv:2411.10534 [cs]. — [arXiv:2411.10534](https://arxiv.org/abs/2411.10534) find
  37. Alyssa Lees, Vinh Q. Tran, Yi Tay, Jeffrey Sorensen, Jai Gupta, Donald Metzler, and Lucy Vasserman. 2022. A new generation of perspective api: Efficient multilingual character-level transformers. In Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD ’22, page 3… find
  38. Alina Leidinger and Richard Rogers. 2024. How are llms mitigating stereotyping harms? learning from search engine studies. find
  39. Xun Liang, Hanyu Wang, Yezhaohui Wang, Shichao Song, Jiawei Yang, Simin Niu, Jie Hu, Dan Liu, Shunyu Yao, Feiyu Xiong, and Zhiyu Li. 2024. Controllable text generation for large language models: A survey. find
  40. Yuanyuan Liang, Jianing Wang, Hanlun Zhu, Lei Wang, Weining Qian, and Yunshi Lan. 2023. Prompting large language models with chain-of-thought for few-shot knowledge base question generation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 4329– 4343,… find
  41. Zizheng Lin, Chunkit Chan, Yangqiu Song, and Xin Liu. 2024. Constrained Reasoning Chains for Enhancing Theory-of-Mind in Large Language Models. ArXiv:2409.13490 [cs]. — [arXiv:2409.13490](https://arxiv.org/abs/2409.13490) find
  42. Aixin Liu, Bei Feng, Bin Wang, Bingxuan Wang, Bo Liu, Chenggang Zhao, Chengqi Dengr, Chong Ruan, Damai Dai, Daya Guo, et al. 2024a. Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model. arXiv preprint arXiv:2405.04434. find
  43. Guangliang Liu, Lei Jiang, Xitong Zhang, and Kristen Marie Johnson. 2025. Diagnosing moral reasoning acquisition in language models: Pragmatics and generalization. arXiv preprint arXiv:2502.16600. find
  44. Yi Liu, Junzhe Yu, Huijia Sun, Ling Shi, Gelei Deng, Yuqi Chen, and Yang Liu. 2024b. Efficient detection of toxic prompts in large language models. find
  45. Yifan Lu, Jing Li, Yigeng Zhou, Yihui Zhang, Wenya Wang, Xiucheng Li, Meishan Zhang, Fangming Liu, Jun Yu, and Min Zhang. 2025. Adaptive detoxification: Safeguarding general capabilities of LLMs through toxicity-aware knowledge editing. In Findings of the Association for Computational Linguistics: A… find
  46. Xilai Ma, Jing Li, and Min Zhang. 2023. Chain of thought with explicit evidence reasoning for few-shot relation extraction. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 2334–2352, Singapore. Association for Computational Linguistics. find
  47. Chengting Mao. 2020. Feminist activism via social media in China. Asian Journal of Women’s Studies, 26(2):245–258. Publisher: Routledge. find
  48. Marco Mazzone. 2011. Schemata and associative processes in pragmatics. Journal of Pragmatics, 43(8):2148–2159. find
  49. Arianna Muti, Federico Ruggeri, Khalid Al Khatib, Alberto Barrón-Cedeño, and Tommaso Caselli. 2024. Language is scary when over-analyzed: Unpacking implied misogynistic reasoning with argumentation theory-driven prompts. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language… find
  50. Saeel Sandeep Nachane, Ojas Gramopadhye, Prateek Chanda, Ganesh Ramakrishnan, Kshitij Sharad Jad-hav, Yatin Nandwani, Dinesh Raghu, and Sachin-dra Joshi. 2024. Few shot chain-of-thought driven reasoning to prompt LLMs for open-ended medical question answering. In Findings of the Association for Comp… find
  51. Tarek Naous, Michael J. Ryan, Alan Ritter, and Wei Xu. 2024. Having Beer after Prayer? Measuring Cultural Bias in Large Language Models. ArXiv:2305.14456 [cs]. our card
  52. Fuqiang Niu, Minghuan Tan, Bowen Zhang, Min Yang, and Ruifeng Xu. 2024. DualCoTs: Dual Chain-of-Thoughts Prompting for Sentiment Lexicon Expansion of Idioms. ArXiv:2409.17588 [cs]. — [arXiv:2409.17588](https://arxiv.org/abs/2409.17588) find
  53. Nicolás Benjamín Ocampo, Ekaterina Sviridova, Elena Cabrio, and Serena Villata. 2023. An in-depth analysis of implicit and subtle hate speech messages. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 1997–2013, Dubrovnik, Croatia.… find
  54. Elinor Ochs. 1988. Culture and Language Development: Language Acquisition and Language Socialization in a Samoan Village. CUP Archive. Google-Books-ID: Zwc5AAAAIAAJ. find
  55. Alexis Palmer, Christine Carr, Melissa Robinson, and Jordan Sanders. 2020. COLD: Annotation scheme and evaluation data set for complex offensive language in English. Journal for Language Technology and Computational Linguistics, 34(1):1–28. Number: 1. find
  56. Jianfeng Pan, Senyou Deng, and Shaomang Huang. 2025. CoAT: Chain-of-Associated-Thoughts Framework for Enhancing Large Language Models Reasoning. ArXiv:2502.02390 [cs]. find
  57. Alicia Parrish, Angelica Chen, Nikita Nangia, Vishakh Padmakumar, Jason Phang, Jana Thompson, Phu Mon Htut, and Samuel R Bowman. 2021. Bbq: A hand-built bias benchmark for question answering. arXiv preprint arXiv:2110.08193. our card
  58. Flor Miriam Plaza-del arco, Debora Nozza, and Dirk Hovy. 2023. Respectful or toxic? using zero-shot learning with language models to detect hate speech. In The 7th Workshop on Online Abuse and Harms (WOAH), pages 60–68, Toronto, Canada. Association for Computational Linguistics. find
  59. Jérôme Prado, Nicola Spotorno, Eric Koun, Emily Hewitt, Jean-Baptiste Van der Henst, Dan Sperber, and Ira A. Noveck. 2015. Neural Interaction between Logical Reasoning and Pragmatic Processing in Narrative Discourse. Journal of Cognitive Neuroscience, 27(4):692–704. find
  60. Zhuang Qiu, Xufeng Duan, and Zhenguang Garry Cai. 2023. Pragmatic Implicature Processing in ChatGPT. find
  61. QwenTeam. 2025. Qwq-32b: Embracing the power of reinforcement learning. https://qwenlm.github.io/blog/qwq-32b/. Accessed: 2025-05-13. ↗
  62. Laura Ruis, Akbir Khan, Stella Biderman, Sara Hooker, Tim Rocktäschel, and Edward Grefenstette. 2023. The Goldilocks of Pragmatic Understanding: Fine-Tuning Strategy Matters for Implicature Resolution by LLMs. ArXiv:2210.14986 [cs]. our card
  63. Maarten Sap, Dallas Card, Saadia Gabriel, Yejin Choi, and Noah A. Smith. 2019. The Risk of Racial Bias in Hate Speech Detection. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1668–1678, Florence, Italy. Association for Computational Linguistics. our card
  64. Richard W. Schmidt. 1990. The Role of Consciousness in Second Language Learning1. Applied Linguistics, 11(2):129–158. Publisher: Oxford Academic. find
  65. Sergio Servantez, Joe Barrow, Kristian Hammond, and Rajiv Jain. 2024. Chain of logic: Rule-based reasoning with large language models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 2721–2733, Bangkok, Thailand. Association for Computational Linguistics. find
  66. Guillermo Villate-Castillo, Javier Del Ser, and Borja Sanz Urquijo. 2024. A Systematic Review of Toxicity in Large Language Models: Definitions, Datasets, Detectors, Detoxification Methods and Challenges. ISSN: 2693-5015. find
  67. Hongbo Wang, LiMingDa LiMingDa, Junyu Lu, Hebin Xia, Liang Yang, Bo Xu, Ruizhu Liu, and Hongfei Lin. 2024. PclGPT: A large language model for patronizing and condescending language detection. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 6913–6928, Miami, Florida, U… find
  68. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2023. Chain-of-thought prompting elicits reasoning in large language models. our card
  69. Jiaxin Wen, Pei Ke, Hao Sun, Zhexin Zhang, Chengfei Li, Jinfeng Bai, and Minlie Huang. 2023. Unveiling the Implicit Toxicity in Large Language Models. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 1322–1338, Singapore. Association for Computational… find
  70. Dan Sperber and Deirdre Wilson. 1995. Relevance: communication and cognition., 2nd ed. edition. Blackwell. find
  71. Thilini Wijesiriwardene, Hale Inan, Ugur Kursuncu, Manas Gaur, Valerie L. Shalin, Krishnaprasad Thirunarayan, Amit Sheth, and I. Budak Arpinar. 2020. ALONE: A Dataset for Toxic Behavior among Adolescents on Twitter. volume 12467, pages 427– 439. ArXiv:2008.06465 [cs]. — [arXiv:2008.06465](https://arxiv.org/abs/2008.06465) find
  72. Dan Sperber and Deirdre Wilson. 1997. Remarks on relevance theory and the social sciences. Multilin-gua - Journal of Cross-Cultural and Interlanguage Communication, 16(2):145–152. find
  73. Alex Wilf, Sihyun Shawn Lee, Paul Pu Liang, and Louis-Philippe Morency. 2023. Think twice: Perspective-taking improves large language models’ theory-of-mind capabilities. find
  74. Nicola Spotorno, Corey T. McMillan, Katya Rascov-sky, David J. Irwin, Robin Clark, and Murray Grossman. 2015. Beyond words: Pragmatic inference in behavioral variant of frontotemporal degeneration. Neuropsychologia, 75:556–564. find
  75. Deirdre Wilson and Dan Sperber. 1993. Linguistic form and relevance. Lingua, 90(1):1–25. find
  76. Zayne Sprague, Fangcong Yin, Juan Diego Rodriguez, Dongwei Jiang, Manya Wadhwa, Prasann Singhal, Xinyu Zhao, Xi Ye, Kyle Mahowald, and Greg Durrett. 2024. To CoT or not to CoT? Chain-of-thought helps mainly on math and symbolic reasoning. ArXiv:2409.12183 [cs]. our card
  77. Rongwu Xu, Zi’an Zhou, Tianwei Zhang, Zehan Qi, Su Yao, Ke Xu, Wei Xu, and Han Qiu. 2024. Walking in others’ shoes: How perspective-taking guides large language models in reducing toxicity and bias. find
  78. Yingjie Zhang, Tong Liu, Zhe Zhao, Guozhu Meng, and Kai Chen. 2025. Align in depth: Defending jailbreak attacks via progressive answer detoxification. find
  79. Settaluri Lakshmi Sravanthi, Meet Doshi, Tankala Pavan Kalyan, Rudra Murthy, Pushpak Bhattacharyya, and Raj Dabre. 2024. PUB: A Pragmatics Understanding Benchmark for Assessing LLMs’ Pragmatics Capabilities. ArXiv:2401.07078 [cs]. our card
  80. Cagri Toraman, Furkan Şahinuç, and Eyup Yilmaz. 2022. Large-Scale Hate Speech Detection with Cross-Domain Transfer. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 2215–2225, Marseille, France. European Language Resources Association. find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/chen-2025-pragmatic-inference-chain-pic-improving/