Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmarks

Junyu Lu, Bo Xu, Xiaokun Zhang, Changrong Min, Liang Yang +1 more
5/8/2023
cs.CLcs.AI

No official code found. Best implementation: DUT-lujunyu/ToxiCN (127 stars, last commit 2025-05-25).

Abstract

The widespread dissemination of toxic online posts is increasingly damaging to society. However, research on detecting toxic language in Chinese has lagged significantly. Existing datasets lack fine-grained annotation of toxic types and expressions, and ignore the samples with indirect toxicity. In addition, it is crucial to introduce lexical knowledge to detect the toxicity of posts, which has been a challenge for researchers. In this paper, we facilitate the fine-grained detection of Chinese toxic language. First, we built Monitor Toxic Frame, a hierarchical taxonomy to analyze toxic types and expressions. Then, a fine-grained dataset ToxiCN is presented, including both direct and indirect toxic samples. We also build an insult lexicon containing implicit profanity and propose Toxic Knowledge Enhancement (TKE) as a benchmark, incorporating the lexical feature to detect toxic language. In the experimental stage, we demonstrate the effectiveness of TKE. After that, a systematic quantitative and qualitative analysis of the findings is given.

View on arXivView PDF

Code Implementations(1)

DUT-lujunyu/ToxiCNCommunity100%

The code and resource of "Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmark" (ACL2023).

1279PythonMay 3, 20231 years ago

Similar Papers

Detecting Toxic Language: Ontology and BERT-based Approaches for Bulgarian Text

87% match2026

Toxic language detection: a systematic review of Arabic datasets

86% match1 repo102023

Textual Toxicity in Social Media: Understanding the Bangla Toxic Language Expressed in Facebook Comment

84% match2023

Something Just Like TRuST : Toxicity Recognition of Span and Target

84% match2025

Defining, Understanding, and Detecting Online Toxicity: Challenges and Machine Learning Approaches

84% match2025

Concept-Based Interpretability for Toxicity Detection

83% match1 repo2025

Cite this paper

@article{lu2023facilitating,
  title  = {Facilitating Fine-grained Detection of Chinese Toxic Language: Hierarchical Taxonomy, Resources, and Benchmarks},
  author = {Junyu Lu and Bo Xu and Xiaokun Zhang and Changrong Min and Liang Yang and Hongfei Lin},
  year   = {2023},
  eprint = {2305.04446},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.04446v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.04446.svg)](https://codeofpaper.com/paper/2305.04446)

Discussion