用大模型推理能力，让仇恨言论检测更懂不同政治立场。
Language, Culture, and Ideology: Personalizing Offensiveness Detection in Political Tweets with Reasoning LLMs
- 让大模型扮演不同政治角色判断推文是否冒犯
- 具备推理能力的模型在多语言下表现更稳定
- 适合做跨文化政治文本分析的研究者
我们研究大语言模型（LLMs）在提示其采用特定政治与文化视角时，如何评估政治话语中的冒犯性。基于聚焦2020年美国大选的多语言MD-Agreement数据子集，我们评估了DeepSeek-R1、o4-mini、GPT-4.1-mini、Qwen3、Gemma和Mistral等近期模型，在英语、波兰语和俄语语境中，从极右翼、保守派、中间派、进步派等不同政治身份视角判断推文是否具有冒犯性。结果显示，具备显式推理能力的大模型（如DeepSeek-R1、o4-mini）在意识形态与文化差异上更具一致性与敏感性，而小模型往往无法捕捉细微差别。推理能力显著提升了判断的个性化与可解释性，表明此类机制是实现跨语言、跨意识形态社会政治文本分类的关键。
原文摘要 · Abstract (English)
We explore how large language models (LLMs) assess offensiveness in political discourse when prompted to adopt specific political and cultural perspectives. Using a multilingual subset of the MD-Agreement dataset centered on tweets from the 2020 US elections, we evaluate several recent LLMs - including DeepSeek-R1, o4-mini, GPT-4.1-mini, Qwen3, Gemma, and Mistral - tasked with judging tweets as offensive or non-offensive from the viewpoints of varied political personas (far-right, conservative, centrist, progressive) across English, Polish, and Russian contexts. Our results show that larger models with explicit reasoning abilities (e.g., DeepSeek-R1, o4-mini) are more consistent and sensitive to ideological and cultural variation, while smaller models often fail to capture subtle distinctions. We find that reasoning capabilities significantly improve both the personalization and interpretability of offensiveness judgments, suggesting that such mechanisms are key to adapting LLMs for nuanced sociopolitical text classification across languages and ideologies.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。