| Challenge: | Detecting hateful content is a challenging and important problem, but it requires continuous training to adapt to the ever-changing landscape of social media. |
| Approach: | They evaluate eight open-source LLMs’ capability to detect antisemitic content . they use in-context definition as a policy guideline to explore different prompts . |
| Outcome: | The proposed model outperforms the fine-tuned model with Guided-CoT . |
Similar Papers
Probing LLMs for hate speech detection: strengths and vulnerabilities (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent efforts to detect hateful or toxic language using large language models have not used explanation, additional context and victim community information in the detection process. |
| Approach: | They use different prompt variations, input information and victim community information to evaluate large language models in zero shot setting without adding any in-context examples. |
| Outcome: | The proposed models perform significantly better when included in the pipeline than baseline models. |
Prompt-driven Detection of Offensive Urdu Language using Large Language Models (2026.eacl-long)
Copied to clipboard
| Challenge: | Offensive language detection systems require extensive tuning and careful model design . a resource gap exists for addressing offensive languages, especially those transcribed in non-native scripts . |
| Approach: | They evaluate pre-trained LLMs using different transcriptions of the Urdu language to assess their performance . they find that they can detect hateful and offensive content in diverse linguistic contexts . |
| Outcome: | The proposed methods perform comparable to fine-tuned benchmarks in diverse languages. |
Explain the Flag: Contextualizing Hate Speech Beyond Censorship (2026.findings-acl)
Copied to clipboard
Jason Liartis, Eirini Kaldeli, Lamprini Gyftokosta, Eleftherios Chelioudakis, Orfeas Menis Mastromichalakis
| Challenge: | a hybrid approach to detect and explain hate speech combines large language models with vocabularies to detect hate speech in three languages . authors: the spread of hate speech online has serious personal, social, and legal consequences . eu has launched initiatives to analyze, regulate, and counteract online hate speech, authors say . |
| Approach: | They propose a hybrid approach that combines Large Language Models with vocabularies to detect hate speech in English, French, and Greek. |
| Outcome: | The proposed approach outperforms baselines in English, French, and Greek . it uses large language models and vocabularies to detect and explain hate speech . human evaluation shows that the proposed approach is accurate and clear . |
HARE: Explainable Hate Speech Detection with Step-by-Step Reasoning (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent benchmarks have attempted to identify and explain hate speech but lack the reasoning to supervise detection models. |
| Approach: | They propose a framework that uses large language models to fill in the gaps in hate speech explanations by using existing annotations. |
| Outcome: | The proposed framework outperforms baselines on SBIC and Implicit Hate using model-generated data and improves generalization to unseen datasets. |
Model-Dependent Moderation: Inconsistencies in Hate Speech Detection Across LLM-based Systems (2025.findings-acl)
Copied to clipboard
| Challenge: | Content moderation systems powered by large language models are increasingly deployed to detect hate speech . if two systems produce different outcomes for the same content, it undermines consistency and predictability . |
| Approach: | They analyze 1.3+ million sentences from a factorial design to determine hate speech classification . they find identical content receives markedly different classification values across systems . |
| Outcome: | The proposed model finds that identical content receives markedly different classification values across systems. |
Decoding Hate: Exploring Language Models’ Reactions to Hate Speech (2025.naacl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are trained on vast amounts of unmoderated internet data, enabling them to generate text autonomously. |
| Approach: | They investigate the responses of seven state-of-the-art Large Language Models (LLMs) to hate speech by qualitative analysis. |
| Outcome: | The proposed models can handle hate speech inputs and mitigate it through fine-tuning and guideline guardrailing. |
Integrating Argumentation and Hate-Speech-based Techniques for Countering Misinformation (2024.emnlp-main)
Copied to clipboard
| Challenge: | scalable strategies to combat online misinformation are short-term and insufficient, authors say . current reactive approaches, like content flagging and banning, do little to change perception of misinformants . human evaluations show that our framework generates expert-like responses . |
| Approach: | They propose a framework that generates persuasive responses from hate-speech counter-responses . human evaluations show that the framework generates expert-like responses . |
| Outcome: | The proposed framework generates expert-like responses and is 14% more engaging, 21% more natural, and 18% more factual than the best available alternatives. |
MFTCXplain: A Multilingual Benchmark Dataset for Evaluating the Moral Reasoning of LLMs through Multi-hop Hate Speech Explanation (2025.findings-emnlp)
Copied to clipboard
Jackson Trager, Francielle Vargas, Diego Alves, Matteo Guida, Mikel K. Ngueajio, Ameeta Agrawal, Yalda Daryani, Farzan Karimi Malekabadi, Flor Miriam Plaza-del-Arco
| Challenge: | Existing evaluation benchmarks for large language models lack annotations that justify moral classifications and focus on English constrain moral reasoning across diverse cultural settings. |
| Approach: | They propose a multilingual benchmark dataset for evaluating moral reasoning of large language models . it includes 3,000 tweets annotated with binary hate speech labels, moral categories and rationales . |
| Outcome: | The proposed dataset shows a misalignment between LLM outputs and human annotations in moral reasoning tasks. |
LLM generated responses to mitigate the impact of hate speech (2024.findings-emnlp)
Copied to clipboard
Jakub Podolak, Szymon Łukasik, Paweł Balawender, Jan Ossowski, Jan Piotrowski, Katarzyna Bakowicz, Piotr Sankowski
| Challenge: | a study aims to determine the effectiveness of large language models to counteract hate speech . it is the first real-life A/B test evaluating the effectiveness . |
| Approach: | They conduct the first real-life A/B test assessing the effectiveness of LLM-generated counter-speech. |
| Outcome: | The proposed system reduces user engagement by over 20%, the study shows . the proposed metric is based on a simple metric and is scalable to other platforms . |
Exploiting contextual information to improve stance detection in informal political discourse with LLMs (2025.acl-srw)
Copied to clipboard
| Challenge: | Political stance detection is an increasingly relevant part of analyzing the flow of ideas in online environments where discourse is informal and implicitly expressed. |
| Approach: | They evaluate large language models for political stance detection in informal online discourse by analyzing user profiles derived from historical posts. |
| Outcome: | The proposed model improves accuracy by up to 74% on a political forum dataset. |