Challenge: Detecting offensive language in Chinese is challenging due to homophonic substitutions used to evade detection.
Approach: They propose to use HED-COLD to build a large-scale homophonic dataset for Chinese offensive language detection and a homophone-aware pretraining strategy to learn phonetics and orthography.
Outcome: The proposed framework achieves state-of-the-art performance on the COLD test set and the toxicity benchmark ToxiCloakCN.

Similar Papers

ToxiCloakCN: Evaluating Robustness of Offensive Language Detection in Chinese with Cloaking Perturbations (2024.emnlp-main)

Copied to clipboard

Challenge: Existing large language models struggle with systematically perturbed data designed to evade detection mechanisms.
Approach: They propose a large language model with homophonic substitutions and emoji transformations to test their models' robustness against cloaking perturbations.
Outcome: The proposed model underperforms in detecting offensive content when perturbations are applied to Chinese language datasets.
COLD: A Benchmark for Chinese Offensive Language Detection (2022.emnlp-main)

Copied to clipboard

Challenge: Offensive language detection is crucial for maintaining a civilized social media platform and deploying pre-trained language models.
Approach: They propose a benchmark benchmark for Chinese offensive language analysis including a Chinese Offensive Language Dataset and a baseline detector which is trained on the dataset.
Outcome: The proposed benchmark contributes to Chinese offensive language detection which is challenging for existing resources.
Lost in Pronunciation: Detecting Chinese Offensive Language Disguised by Phonetic Cloaking Replacement (2025.emnlp-industry)

Copied to clipboard

Challenge: Phonetic Cloaking Replacement (PCR) is a problem in content moderation in China.
Approach: They organize PCR into a four-way surface-form taxonomy and compile PCR-ToxiCN, a dataset of 500 phonetically cloaked offensive posts gathered from the RedNote platform.
Outcome: The proposed model achieves only an F1-score and zero-shot chain-of-thought prompting pushes performance even lower.
Exploring Multimodal Challenges in Toxic Chinese Detection: Taxonomy, Benchmark, and Findings (2025.findings-acl)

Copied to clipboard

Challenge: Recent studies show that character substitutions in toxic Chinese text can confuse state-of-the-art LLMs.
Approach: They propose a taxonomy of 3 perturbation strategies and 8 specific approaches in Chinese text to assess if they can detect perturbed Chinese toxic contents.
Outcome: The proposed model can detect perturbed Chinese text with 8 different approaches . the proposed model is compared with 9 other LLMs from the US and China .
Giving Control Back to Models: Enabling Offensive Language Detection Models to Autonomously Identify and Mitigate Biases (2024.findings-emnlp)

Copied to clipboard

Challenge: Existing models often rely on specific words to predict offensive content, compromising model fairness and potentially exacerbates biases against vulnerable and minority groups.
Approach: They propose a bias self-awareness and data self-iteration framework to help models identify and mitigate biases by integrating multiple natural language processing techniques.
Outcome: The proposed framework reduces false positive rate of models in in-distribution and out-of-difference tests, enhances model accuracy and fairness, and shows promising performance improvements on larger datasets.
Don’t be a Fool: Pooling Strategies in Offensive Language Detection from User-Intended Adversarial Attacks (2024.findings-naacl)

Copied to clipboard

Challenge: Offensive language detection is important for filtering out abusive expressions, authors argue . authors propose user-intended adversarial attacks that insert special symbols or leverage distinctive features of the Korean language.
Approach: They propose user-intended adversarial attacks that insert special symbols or leverage the distinctive features of the Korean language.
Outcome: The proposed models are more robust to performance degradation even when the attack rate is increased, compared to models trained on noisy texts.
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.
HateCOT: An Explanation-Enhanced Dataset for Generalizable Offensive Speech Detection via Large Language Models (2024.findings-emnlp)

Copied to clipboard

Challenge: Social media has amplified the propagation of hateful sentiments, highlighting the contested nature of "offensive content" research shows that "of offensive content" is still a contested construct due to varying definitions and labeling.
Approach: They propose a dataset that features human-curated explanations for offensive content in English . they show that HateCOT pretraining improves performance of open-source LLMs .
Outcome: The proposed model improves on three benchmark datasets for offensive content detection . the model improve the quality of its explanations, as confirmed by the human evaluation .
Offensive Content Detection via Synthetic Code-Switched Text (2022.coling-1)

Copied to clipboard

Challenge: Existing methods to detect offensive content in social media platforms are limited by the availability of labeled code-switched data.
Approach: They propose a method for generating synthetic code-switched offensive content data using human-generated data and a keyword classification baseline.
Outcome: The proposed algorithm can be used to generate synthetic code-switched offensive content data and train it on human-generated data.
Self-Supervised Contrastive Learning with Adversarial Perturbations for Defending Word Substitution-based Attacks (2022.findings-naacl)

Copied to clipboard

Challenge: Existing methods to improve model robustness against word substitution-based adversarial attacks are too slow to generate adversarials on the fly.
Approach: They propose an approach to improve the robustness of BERT models against word substitution-based adversarial attacks by leveraging adversarials for self-supervised contrastive learning.
Outcome: The proposed method improves robustness of BERT models against word substitution-based adversarial attacks without using any labeled data.

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