英语与中文居高临下怜悯性语言（PCL）自动检测研究

Patronizing and Condescending Language Detection in English and Mandarin

AI 导读

一项 arXiv 研究提出自动检测居高临下怜悯性语言（PCL）的系统，这类针对弱势群体的隐性有害语言连人工判别也困难。方法一是静态词向量加卷积神经网络分类，方法二是对预训练 BERT embeddings 做 PCL 任务微调，后者效果明显更好，在开发集上取得 0.528 的 F1 分数。

正文 · 原文

View PDF HTML (experimental)

Abstract:Patronizing and condescending language (PCL) is a form of harmful language that belittles and assumes pity for vulnerable groups, reinforcing negative societal stereotypes. However, unlike overt hate speech, it is often difficult to detect, even for human judges. This paper proposes a system for the automated detection of PCL. Our first approach uses static word embeddings and a convolutional neural network for classification, while our second approach uses pre-trained BERT embeddings fine-tuned for the PCL task. Our second approach was considerably more successful, and achieved an F1 score of 0.528 over the development set.

Subjects:	Artificial Intelligence (cs.AI)
Cite as:	arXiv:2610.31808 [cs.AI]
	(or arXiv:2610.31808v1 [cs.AI] for this version)
	https://doi.org/10.48550/arXiv.2610.31808 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Xinyi Xiang [view email]
[v1] Thu, 22 Oct 2026 18:44:00 UTC (35 KB)