一作ACL 2026Findings· Findings of ACL
Penetrating Linguistic Disguises: A Slang-aware Label-Aligned Framework for Fine-Grained Toxicity Extraction in C hinese Hate Speech Detection
Wei Liu · Xiaoliang Chen · Duoqian Miao · Xu Gu · Xianyong Li · Yajun Du
摘要
Flexible word boundaries and linguistic obfuscation, particularly slang, challenge precise span-level hate speech detection in Chinese. While benchmarks such as STATE ToxiCN demand the exact extraction of Target-Argument-Hateful-Group quadruples, generative Large Language Models (LLMs) often fail...