| Challenge: | Existing adversarial examples generate invalid or ambiguous examples that fool the systems into wrong detection. |
| Approach: | They propose an annotation pipeline for quality control of generated toxic adversarial examples (TAE) they use model-based automated annotation and human-based quality verification to assess quality requirements of a TAE dataset. |
| Outcome: | The proposed pipeline can transfer-attack SOTA toxicity content moderation models and services with adversarial training. |
Similar Papers
Towards Building a Robust Toxicity Predictor (2023.acl-industry)
Copied to clipboard
| Challenge: | Recent studies have focused on robustness of toxicity language predictors, but this is problematic for real-world toxicity detection. |
| Approach: | They propose a novel adversarial attack that exploits greedy search strategies to fool toxic text classifiers. |
| Outcome: | The proposed attack can detect weaker toxicity language detectors even against unseen attacks. |
ToxiCraft: A Novel Framework for Synthetic Generation of Harmful Information (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing models for detecting harmful content lack diversity and quality of datasets. |
| Approach: | They propose a framework for synthesizing toxic information from social media datasets . their framework generates a wide variety of synthetic, yet remarkably realistic, examples of toxic information . |
| Outcome: | The proposed framework can generate a wide variety of synthetic, yet remarkably realistic, examples of toxic information. |
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. |
ToVo: Toxicity Taxonomy via Voting (2025.findings-naacl)
Copied to clipboard
Tinh Son Luong, Thanh-Thien Le, Thang Viet Doan, Linh Ngo Van, Thien Huu Nguyen, Nguyen Thi Ngoc Diep
| Challenge: | Existing toxic content detection models face limitations due to the closed-source nature of training data and the paucity of explanations for their evaluation mechanism. |
| Approach: | They propose a mechanism that integrates voting and chain-of-thought processes to produce a high-quality open-source dataset for toxic content detection. |
| Outcome: | The proposed model improves transparency and customizability while facilitating better fine-tuning for specific use cases. |
Large Language Models Are Better Adversaries: Exploring Generative Clean-Label Backdoor Attacks Against Text Classifiers (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Backdoor attacks manipulate model predictions by inserting malicious "poison" instances that contain a specific pattern or "trigger." |
| Approach: | They propose an attack that inserts style-based triggers into training and test data by using a poison selection technique to improve the effectiveness of both LLMBkd and existing backdoor attacks. |
| Outcome: | The proposed attack achieves high success rates across a wide range of styles with little effort and no model training. |
Attacking Misinformation Detection Using Adversarial Examples Generated by Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models can be used to attack content filtering algorithms in social media platforms. |
| Approach: | They propose to generate adversarial examples to test the robustness of social media content filtering algorithms. |
| Outcome: | The proposed model outperforms existing models in the case of propaganda, false claims, rumours and hyperpartisan news. |
ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection (2022.acl-long)
Copied to clipboard
| Challenge: | Toxic language detection systems often falsely flag text that contains minority group mentions as toxic . this over-reliance on spurious correlations also causes systems to struggle with detecting implicitly toxic language. |
| Approach: | They develop a machine-generated dataset of toxic and benign statements about 13 minority groups that generates subtly toxic and harmless text with a massive pretrained language model. |
| Outcome: | The proposed method can detect toxic and benign statements on a large scale . it can also detect hate speech on 94.5% of the toxic examples . |
A Closer Look into the Robustness of Neural Dependency Parsers Using Better Adversarial Examples (2021.findings-acl)
Copied to clipboard
| Challenge: | Neural network-based models have been successful in a wide range of NLP tasks, but their performance is undermined by adversarial examples that would pose no confusion for humans. |
| Approach: | They propose a method to generate high-quality adversarial examples with a higher number of candidate generators and stricter filters and then verify their quality using automatic and human evaluations. |
| Outcome: | The proposed method improves the robustness of English parsing models by relying on adversarial training and model ensembling. |
ReinforceBug: A Framework to Generate Adversarial Textual Examples (2021.naacl-main)
Copied to clipboard
| Challenge: | Recent studies have demonstrated that ML Models are sensitive to Adversarial Examples (AEs) AEs are generated by perturbingining examples that preserve the intrinsic utility of the ML solutions but influence the classifier's predictions between original and modified inputs. |
| Approach: | They propose a reinforcement learning framework that learns a policy that is transferable on unseen datasets and generates utility-preserving and transferable AEs. |
| Outcome: | The proposed framework is 10% more successful than the state-of-the-art attack TextFooler. |
ToxicChat: Unveiling Hidden Challenges of Toxicity Detection in Real-World User-AI Conversation (2023.findings-emnlp)
Copied to clipboard
| Challenge: | toxicity detection has been largely based on social media content, leaving the unique challenges inherent to real-world user-AI interactions insufficiently explored. |
| Approach: | They propose a benchmark to detect toxicity in real-world user-AI conversations . they compare existing models with social media content to find toxicity . |
| Outcome: | The proposed benchmark reveals that existing models fail to recognize toxicity in real-world user-AI conversations. |