对比大模型与人在冲突对话中的行为一致性，发现不同模型各有优劣。
Evaluating Behavioral Alignment in Conflict Dialogue: A Multi-Dimensional Comparison of LLM Agents and Humans
- 用五大人格特质控制大模型，模拟真实冲突对话
- GPT-4.1在语言和情绪上最接近人类，Claude-3.7-Sonnet策略最像人
- 尽管有进步，但大模型仍与人类存在显著行为差距
大型语言模型（LLMs）在社交复杂、互动驱动的任务中应用日益广泛，但其在情感与策略复杂的场景中模仿人类行为的能力仍缺乏深入研究。本研究通过模拟多轮冲突对话并融入谈判机制，评估人格提示下的LLM在对抗性争议解决中的行为一致性。每个LLM均采用匹配的五大人格特质（Five-Factor personality profile）以控制个体差异并提升真实性。从语言风格、情绪表达（如愤怒动态）和战略行为三个维度进行评估。结果显示，GPT-4.1在语言风格与情绪动态上最接近人类，而Claude-3.7-Sonnet在战略行为上表现最佳。然而，仍存在显著的行为差距。研究建立了社会复杂交互中大模型与人类行为一致性的基准，凸显了人格引导在对话建模中的潜力与局限。
原文摘要 · Abstract (English)
Large Language Models (LLMs) are increasingly deployed in socially complex, interaction-driven tasks, yet their ability to mirror human behavior in emotionally and strategically complex contexts remains underexplored. This study assesses the behavioral alignment of personality-prompted LLMs in adversarial dispute resolution by simulating multi-turn conflict dialogues that incorporate negotiation. Each LLM is guided by a matched Five-Factor personality profile to control for individual variation and enhance realism. We evaluate alignment across three dimensions: linguistic style, emotional expression (e.g., anger dynamics), and strategic behavior. GPT-4.1 achieves the closest alignment with humans in linguistic style and emotional dynamics, while Claude-3.7-Sonnet best reflects strategic behavior. Nonetheless, substantial alignment gaps persist. Our findings establish a benchmark for alignment between LLMs and humans in socially complex interactions, underscoring both the promise and the limitations of personality conditioning in dialogue modeling.
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。