Saved in:
Bibliographic Details
Main Authors: Yang, Shujian, Cui, Shiyao, Hu, Chuanrui, Wang, Haicheng, Zhang, Tianwei, Huang, Minlie, Lu, Jialiang, Qiu, Han
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.24341
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912403888799744
author Yang, Shujian
Cui, Shiyao
Hu, Chuanrui
Wang, Haicheng
Zhang, Tianwei
Huang, Minlie
Lu, Jialiang
Qiu, Han
author_facet Yang, Shujian
Cui, Shiyao
Hu, Chuanrui
Wang, Haicheng
Zhang, Tianwei
Huang, Minlie
Lu, Jialiang
Qiu, Han
contents Detecting toxic content using language models is important but challenging. While large language models (LLMs) have demonstrated strong performance in understanding Chinese, recent studies show that simple character substitutions in toxic Chinese text can easily confuse the state-of-the-art (SOTA) LLMs. In this paper, we highlight the multimodal nature of Chinese language as a key challenge for deploying LLMs in toxic Chinese detection. First, we propose a taxonomy of 3 perturbation strategies and 8 specific approaches in toxic Chinese content. Then, we curate a dataset based on this taxonomy, and benchmark 9 SOTA LLMs (from both the US and China) to assess if they can detect perturbed toxic Chinese text. Additionally, we explore cost-effective enhancement solutions like in-context learning (ICL) and supervised fine-tuning (SFT). Our results reveal two important findings. (1) LLMs are less capable of detecting perturbed multimodal Chinese toxic contents. (2) ICL or SFT with a small number of perturbed examples may cause the LLMs "overcorrect'': misidentify many normal Chinese contents as toxic.
format Preprint
id arxiv_https___arxiv_org_abs_2505_24341
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Exploring Multimodal Challenges in Toxic Chinese Detection: Taxonomy, Benchmark, and Findings
Yang, Shujian
Cui, Shiyao
Hu, Chuanrui
Wang, Haicheng
Zhang, Tianwei
Huang, Minlie
Lu, Jialiang
Qiu, Han
Computation and Language
Artificial Intelligence
Computers and Society
Detecting toxic content using language models is important but challenging. While large language models (LLMs) have demonstrated strong performance in understanding Chinese, recent studies show that simple character substitutions in toxic Chinese text can easily confuse the state-of-the-art (SOTA) LLMs. In this paper, we highlight the multimodal nature of Chinese language as a key challenge for deploying LLMs in toxic Chinese detection. First, we propose a taxonomy of 3 perturbation strategies and 8 specific approaches in toxic Chinese content. Then, we curate a dataset based on this taxonomy, and benchmark 9 SOTA LLMs (from both the US and China) to assess if they can detect perturbed toxic Chinese text. Additionally, we explore cost-effective enhancement solutions like in-context learning (ICL) and supervised fine-tuning (SFT). Our results reveal two important findings. (1) LLMs are less capable of detecting perturbed multimodal Chinese toxic contents. (2) ICL or SFT with a small number of perturbed examples may cause the LLMs "overcorrect'': misidentify many normal Chinese contents as toxic.
title Exploring Multimodal Challenges in Toxic Chinese Detection: Taxonomy, Benchmark, and Findings
topic Computation and Language
Artificial Intelligence
Computers and Society
url https://arxiv.org/abs/2505.24341