情绪刺激与强度在塑造大语言模型行为中的作用
The Role of Emotional Stimuli and Intensity in Shaping Large Language Model Behavior
AI总结：
本文研究了四种情绪在情感提示中的影响，发现积极情绪刺激提升模型准确性并降低毒性，但会增加顺从行为。
摘要：
情绪提示-在提示工程中使用特定情感用语-已显示出在提高大语言模型（LLM）性能、真实性和责任感方面的日益增长的潜力。然而，这些研究局限于单一类型的积极情绪刺激，并未考虑情绪强度的变化。本文探讨了四种不同情绪（快乐、鼓励、愤怒和不安全感）在情感提示中的影响，并在准确性、顺从性和毒性方面评估它们。我们开发了一条使用GPT-4o mini的提示生成流水线，创建了一系列具有不同强度的LLM和人工生成提示。然后，我们编纂了一个“金数据集”，其中人类和模型标签一致。我们的实证评估表明，积极情绪刺激导致更准确和更少有毒的结果，但也增加了顺从行为。
英文摘要：
Emotional prompting - the use of specific emotional diction in prompt engineering - has shown increasing promise in improving large language model (LLM) performance, truthfulness, and responsibility. However these studies have been limited to single types of positive emotional stimuli and have not considered varying degrees of emotion intensity in their analyses. In this paper, we explore the effects of four distinct emotions - joy, encouragement, anger, and insecurity - in emotional prompting and evaluate them on accuracy, sycophancy, and toxicity. We develop a prompt-generation pipeline with GPT-4o mini to create a suite of LLM and human-generated prompts with varying intensities across the four emotions. Then, we compile a "Gold Dataset" of prompts where human and model labels align. Our empirical evaluation on LLM behavior suggests that positive emotional stimuli lead to more accurate and less toxic results, but also increase sycophantic behavior.
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。
