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.

Similar Papers

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.
Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metric (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing toxicity metrics rely on encoder models trained on specific toxicity datasets, which are susceptible to out-of-distribution (OOD) problems and depend on the dataset’s definition of toxicity.
Approach: They propose a robust metric grounded on LLMs to flexibly measure toxicity according to the given definition by analysing toxicity factors and intrinsic toxic attributes.
Outcome: The proposed metric improves on conventional metrics by 12 points in the F1 score and shows that upstream toxicity significantly influences downstream metrics, suggesting that LLMs are unsuitable for toxicity evaluations within unverified factors.
Enhancing LLM-based Hatred and Toxicity Detection with Meta-Toxic Knowledge Graph (2025.findings-acl)

Copied to clipboard

Challenge: Existing methods to address toxicity issues with large language models are inadequate . lack of domain-specific knowledge leads to false negatives and excessive sensitivity to toxic speech limits freedom of speech.
Approach: They propose a method that leverages graph search on a meta-toxic knowledge graph to enhance hatred and toxicity detection.
Outcome: The proposed method lowers false positive rate and improves toxicity detection performance in out-of-domain scenarios.
Unveiling the Implicit Toxicity in Large Language Models (2023.emnlp-main)

Copied to clipboard

Challenge: Recent studies focus on probing toxic outputs that can be easily detected with existing toxicity classifiers, but LLMs can generate diverse implicit toxic output that are difficult to detect via simply zero-shot prompting.
Approach: They propose a reinforcement learning based attacking method to induce the implicit toxic outputs in large language models by fine-tuning toxicity classifiers.
Outcome: The proposed method generates implicit toxic outputs that are difficult to detect via zero-shot prompting on five widely-adopted toxicity classifiers.
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.
ModelCitizens: Representing Community Voices in Online Safety (2025.emnlp-main)

Copied to clipboard

Challenge: Existing toxic language detection models are trained on annotations that collapse diverse perspectives into a single ground truth.
Approach: They propose to augment social media posts with conversational scenarios to reflect the impact of conversational context on toxicity.
Outcome: The proposed model outperforms existing models on social media with conversational scenarios.
Challenges in Detoxifying Language Models (2021.findings-emnlp)

Copied to clipboard

Challenge: Prior work often relies on automatic evaluation of LM toxicity.
Approach: They evaluate toxicity mitigation strategies for automated and human evaluations . they find human raters disagree with high automatic toxicity scores after strong toxicity reduction interventions .
Outcome: The proposed methods reduce LM toxicity but lower coverage for marginalized texts . human raters disagree with high toxicity scores after strong toxicity reduction interventions .
Fortifying Toxic Speech Detectors Against Veiled Toxicity (2020.emnlp-main)

Copied to clipboard

Challenge: Modern toxic speech detectors are incompetent in recognizing disguised offensive language, such as adversarial attacks that deliberately avoid known toxic lexicons.
Approach: They propose a framework that fortifies existing toxic speech detectors without a large labeled corpus of veiled toxicity.
Outcome: The proposed framework is aimed at fortifying existing toxic speech detectors without a large labeled corpus of disguised offensive language.
Pragmatic Inference Chain (PIC) Improving LLMs’ Reasoning of Authentic Implicit Toxic Language (2025.emnlp-main)

Copied to clipboard

Challenge: Existing studies have shown that LLMs can detect toxicity by using a variety of inference-intensive tasks, such as understanding humour and metaphors.
Approach: They propose a new method to prompt LLMs to identify toxic language using a set of online data that are verified by human annotators.
Outcome: The proposed method significantly improves the success rate of GPT-4o, Llama-3.1-70B-Instruct, DeepSeek-v2.5, and DeepSeq-v3 in identifying implicit toxic language compared to five baseline prompts, such as CoT and rule-based baselines.
Performance and Risk Trade-offs for Multi-word Text Prediction at Scale (2023.findings-eacl)

Copied to clipboard

Challenge: Large Language Models (LLMs) generate ethically inappropriate texts even for seemingly innocuous contexts.
Approach: They propose to use large language models to detect and filter toxic content in text prediction tasks by evaluating their toxicity detection approaches against a manually crafted CheckList of harms.
Outcome: The proposed methods are compared against a checklist of harms targeted at different groups and different levels of severity in English.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations