打破助手模式：在基于大语言模型的程序化角色生成中建模行为变异
Breaking the Assistant Mold: Modeling Behavioral Variation in LLM Based Procedural Character Generation
AI总结：
本文提出PersonaWeaver框架，通过解耦世界观构建与行为构建，生成更具多样性和戏剧张力的角色。
摘要：
程序化内容生成通过层级、地图和任务构建了广阔虚拟世界，但大规模角色生成仍被忽视。我们发现现有方法存在两种对齐诱导偏差：积极道德偏差（角色普遍采用可接受立场）和助手法偏差（角色总是直接回答问题）。此类倾向适合指令遵循系统，但导致戏剧张力缺失和可预测角色，源于最大似然训练和助手微调。为解决此问题，我们引入PersonaWeaver框架，将世界观构建（角色、人口统计）与行为构建（道德立场、互动风格）解耦，生成更具多样反应和道德立场的角色，以及在风格标记如长度、语气和标点上的二次多样性。代码：https://github.com/mqraitem/Persona-Weaver
英文摘要：
Procedural content generation has enabled vast virtual worlds through levels, maps, and quests, but large-scale character generation remains underexplored. We identify two alignment-induced biases in existing methods: a positive moral bias, where characters uniformly adopt agreeable stances (e.g. always saying lying is bad), and a helpful assistant bias, where characters invariably answer questions directly (e.g. never refusing or deflecting). While such tendencies suit instruction-following systems, they suppress dramatic tension and yield predictable characters, stemming from maximum likelihood training and assistant fine-tuning. To address this, we introduce PersonaWeaver, a framework that disentangles world-building (roles, demographics) from behavioral-building (moral stances, interactional styles), yielding characters with more diverse reactions and moral stances, as well as second-order diversity in stylistic markers like length, tone, and punctuation. Code: https://github.com/mqraitem/Persona-Weaver
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。
