Speak No Evil, Just Prompt: Low-resource Multilingual Toxic Speech Detection with Audio Language Model (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for toxic speech detection rely on high-resource languages and lack acoustic cues. |
| Approach: | They propose a prompt-based adaptation framework that performs end-to-end toxicity detection without ASR. |
| Outcome: | The proposed framework achieves a micro-averaged ROC-AUC of 98.07% on polySpeechTox . it is based on a frozen audio language model and can perform end-to-end toxicity detection without ASR . |
Similar Papers
MuTox: Universal MUltilingual Audio-based TOXicity Dataset and Zero-shot Detector (2024.findings-acl)
Copied to clipboard
Marta Costa-jussà, Mariano Meglioli, Pierre Andrews, David Dale, Prangthip Hansanti, Elahe Kalbassi, Alexandre Mourachko, Christophe Ropers, Carleigh Wood
| Challenge: | Existing studies on text-based toxicity detection for other languages are limited, especially for languages other than English. |
| Approach: | They propose a multilingual audio-based toxicity classifier which covers 14 different linguistic families and a dataset of 20,000 audio utterances for English and Spanish. |
| Outcome: | The new classifier improves F1-Score by an average of 100% when compared to existing wordlist-based classifiers. |
Realistic Evaluation of Toxicity in Large Language Models (2024.findings-acl)
Copied to clipboard
| Challenge: | a large amount of data exposes large language models to toxicity and bias . prompt engineering can be easily bypassed with minimal prompt engineering. |
| Approach: | They propose a dataset that uses manually crafted prompts to nullify protective layers of large language models. |
| Outcome: | The proposed dataset shows that prompts can nullify protective layers of large language models. |
A Survey of Toxicity Mitigation Strategies for Multilingual Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Large language models can reproduce and amplify toxic content, including hate speech, harassment, and bias. |
| Approach: | They propose a comprehensive survey of the many detoxification methods tailored to multilingual LLMs. |
| Outcome: | The proposed methods are based on data filtering, style transfer, expert-based logit steering, retrieval augmentation, and human feedback. |
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. |
No offence, Bert - I insult only humans! Multilingual sentence-level attack on toxicity detection networks (2023.findings-emnlp)
Copied to clipboard
| Challenge: | a new sentence-level attack on toxic detection models is shown to work on seven languages . toxicity detection systems are used to silence the voices of criticism, causing echo chambers . |
| Approach: | They propose a sentence-level attack that adds positive words to a hateful message . they show the attack works on seven languages from three different language families . |
| Outcome: | The proposed attack is shown to work on seven languages from three different language families. |
Data-Efficient Strategies for Expanding Hate Speech Detection into Under-Resourced Languages (2022.emnlp-main)
Copied to clipboard
| Challenge: | Hate speech datasets focus on English-language content, hindering effective models . annotating hateful content is expensive, time-consuming and potentially harmful to annotators. |
| Approach: | They propose to use ISO 639-1 codes to fine-tune models on one source language and apply them to another language. |
| Outcome: | The proposed approach performs well on some tasks, but fails on many others. |
UnityAI Guard: Pioneering Toxicity Detection Across Low-Resource Indian Languages (2025.emnlp-demos)
Copied to clipboard
Himanshu Beniwal, Reddybathuni Venkat, Rohit Kumar, Birudugadda Srivibhav, Daksh Jain, Pavan Deekshith Doddi, Eshwar Dhande, Adithya Ananth, null Kuldeep, Mayank Singh
| Challenge: | Existing systems target high-resource languages, but UnityAI-Guard addresses this gap by developing state-of-the-art models for binary toxicity classification targeting low-resourced Indian languages. |
| Approach: | They propose a framework for binary toxicity classification targeting low-resource Indian languages. |
| Outcome: | The proposed framework achieves an impressive average F1-score of 84.23% across seven languages, leveraging a dataset of 567k training instances and 30k manually verified test instances. |
♪ Something Just Like TRuST ♪ *: Toxicity Recognition of Span and Target (2026.findings-acl)
Copied to clipboard
| Challenge: | Toxic language is pervasive online, and because LLMs are trained on web data, it generates such content. |
| Approach: | They propose a large-scale dataset that synthesizes toxicity definitions and an annotation scheme . they use a rigorous human annotation process to evaluate the diversity of the annotations . |
| Outcome: | The proposed model outperforms existing models on three tasks and is not reliable. |
TOXIFRENCH: Benchmarking and Enhancing Language Models via CoT Fine-Tuning for French Toxicity Detection (2026.findings-acl)
Copied to clipboard
| Challenge: | toxicity detection in French remains underdeveloped due to the lack of culturally relevant, human-annotated, large-scale datasets. |
| Approach: | They propose a method that generalizes French online comments using a semi-automated annotation pipeline that reduces manual labeling to only 10% through high-confidence LLM-based pre-annotation and human verification. |
| Outcome: | The proposed model outperforms GPT-4o and DeepSeek-R1 on the benchmark while maintaining cross-lingual capabilities. |
Towards Cross-Lingual Audio Abuse Detection in Low-Resource Settings with Few-Shot Learning (2025.coling-main)
Copied to clipboard
| Challenge: | Online abusive content detection, particularly in low-resource settings, remains underexplored. |
| Approach: | They propose to use pre-trained audio representations to detect abusive language in Indian languages using Few Shot Learning (FSL) . |
| Outcome: | The proposed model can be used to classify abusive language in 10 languages using the ADIMA dataset with FSL. |