在回应前像人一样思考：对人格引导的大语言模型用于反击仇恨的多维度评估
Think Like a Person Before Responding: A Multi-Faceted Evaluation of Persona-Guided LLMs for Countering Hate
AI总结：
本文提出从人格框架、冗长性与可读性、情感基调和伦理稳健性四个维度评估LLM生成的反仇恨言论反叙事，并在多个模型和数据集上发现现有生成结果冗长且可及性受限，情感引导提示虽改善同理心与可读性，但安全性和有效性仍有待提升。
摘要：
自动化反叙事（CN）为缓解在线仇恨言论提供了一种有前景的策略，但对其情感基调、可及性和伦理风险的担忧依然存在。我们提出一个框架，从四个维度评估大语言模型（LLM）生成的反叙事：人格框架、冗长性与可读性、情感基调以及伦理稳健性。使用GPT-4o-Mini、Cohere的CommandR-7B和Meta的LLaMA 3.1-70B，我们在MT-Conan和HatEval数据集上评估了三种提示策略。我们的研究结果表明，LLM生成的反叙事往往冗长，并且适合具有大学水平读写能力的人群，这限制了其可及性。虽然情感引导的提示能产生更具同理心和可读性的回应，但在安全性和有效性方面仍存在担忧。
英文摘要：
Automated counter-narratives (CN) offer a promising strategy for mitigating online hate speech, yet concerns about their affective tone, accessibility, and ethical risks remain. We propose a framework for evaluating Large Language Model (LLM)-generated CNs across four dimensions: persona framing, verbosity and readability, affective tone, and ethical robustness. Using GPT-4o-Mini, Cohere's CommandR-7B, and Meta's LLaMA 3.1-70B, we assess three prompting strategies on the MT-Conan and HatEval datasets. Our findings reveal that LLM-generated CNs are often verbose and adapted for people with college-level literacy, limiting their accessibility. While emotionally guided prompts yield more empathetic and readable responses, there remain concerns surrounding safety and effectiveness.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
