大模型比有激励的人类更会说服，尤其在说真话时效果更好。
When Large Language Models are More PersuasiveThan Incentivized Humans, and Why
- 对比大模型与有奖励的人类，测试实时对话中的说服力。
- 大模型在说真话时提升答题准确率，说假话时反而降低准确率。
- 大模型表达更强信念感，可能是其说服力来源，适合研究人机交互者参考。
大型语言模型（LLMs）已被证明具有高度说服力，但其何时及为何优于人类仍不明确。本文通过实时互动对话场景，比较了Claude 3.5 Sonnet和DeepSeek v3两款大模型与受激励人类的说服表现。结果显示，大模型的说服优势具有情境依赖性：在追求正确答案（真实说服）时，其表现优于人类，并显著提升准确率；而在引导错误答案（欺骗性说服）时，虽更显说服力，却导致准确率下降。在首次大规模实验中，人类与大模型（Claude 3.5 Sonnet）分别劝说他人完成在线测验以获取奖励，结果表明前者在真伪两种情境下均更具说服力。第二次实验使用DeepSeek v3复现了准确率变化结果，但仅在欺骗性情境中表现出更强说服力。对说服文本的语言分析显示，大模型表达出更高程度的自信，可能正是其说服力的关键原因。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have been shown to be highly persuasive, but when and why they outperform humans is still an open question. We compare the persuasiveness of two LLMs (Claude 3.5 Sonnet and DeepSeek v3) against humans who had incentives to persuade, using an interactive, real-time conversational setting. We demonstrate that LLMs persuasive superiority is context-dependent: it depends on whether the persuasion attempt is truthful (towards the right answer) or deceptive (towards the wrong answer) and on the LLM model, and wanes over repeated interactions (unlike human persuasiveness). In our first large-scale experiment, humans vs LLMs (Claude 3.5 Sonnet) interacted with other humans who were completing an online quiz for a reward, attempting to persuade them toward a given (either correct or incorrect) answer. Claude was more persuasive than incentivized human persuaders both in truthful and deceptive contexts and it significantly increased accuracy if persuasion was truthful, but decreased it if persuasion was deceptive. In a follow-up experiment with Deepseek v3, we replicated the findings about accuracy but found greater LLM persuasiveness only if the persuasion was deceptive. Linguistic analyses of the persuaders texts suggest that these effects may be due to LLMs expressing higher conviction than humans.
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。