捧上踩下：探索分配视觉人设下多模态大语言模型的行为变化
Kiss up, Kick down: Exploring Behavioral Changes in Multi-modal Large Language Models with Assigned Visual Personas
AI总结：
本研究构建5000张虚拟头像的视觉人设数据集，首次探究多模态大语言模型的行为与视觉人设的对齐情况，发现模型对攻击性的判断与人类相似，且会根据自身与对手人设的攻击性程度调整谈判行为。
摘要：
本研究首次探索多模态大语言模型（LLMs）能否让自身行为与视觉人设对齐，填补了现有文献多聚焦文本人设的显著空白。我们构建了一个含5000张虚拟头像图像的新型数据集，用作分配给LLMs的视觉人设，并基于这些图像呈现的视觉特征分析其谈判行为，重点关注攻击性维度。结果表明，LLMs评估图像攻击性的方式与人类相似，当被赋予攻击性视觉人设提示时，会输出更具攻击性的谈判行为。有趣的是，当对手图像的攻击性看起来低于自身时，LLM会表现出更具攻击性的谈判行为；而当对手图像攻击性更高时，其行为攻击性则会降低。
英文摘要：
This study is the first to explore whether multi-modal large language models (LLMs) can align their behaviors with visual personas, addressing a significant gap in the literature that predominantly focuses on text-based personas. We developed a novel dataset of 5K fictional avatar images for assignment as visual personas to LLMs, and analyzed their negotiation behaviors based on the visual traits depicted in these images, with a particular focus on aggressiveness. The results indicate that LLMs assess the aggressiveness of images in a manner similar to humans and output more aggressive negotiation behaviors when prompted with an aggressive visual persona. Interestingly, the LLM exhibited more aggressive negotiation behaviors when the opponent's image appeared less aggressive than their own, and less aggressive behaviors when the opponents image appeared more aggressive.
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。
