arXiv 2512.237162025-12-02cs.CL

噪声驱动的自我反思神经语言生成中的身份形成

Noise-Driven Persona Formation in Reflexive Neural Language Generation

Toshiyuki Shigemura

备注：: 324 pages, 9 figures (Figure 7 intentionally skipped), with Appendices A-I. This manuscript presents a computational framework for noise-driven persona formation in neural language generation, analyzing 152 generation cycles using GPT-5.1 with stochastic noise seeds generated by Microsoft Copilot. Primary category: cs.CL

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出Luca-Noise Reflex协议，通过注入噪声种子研究大语言模型中噪声驱动的身份形成，揭示三种稳定的生成模式并验证噪声对生成动态的影响。

摘要：

本文介绍了Luca-Noise Reflex协议（LN-RP），一种用于分析噪声驱动身份在大语言模型中出现的计算框架。通过将随机噪声种子注入初始生成状态，我们观察到在152次生成循环中语言行为的非线性转变。我们的结果揭示了三种稳定的身份模式，具有不同的熵特征，并证明外部噪声源可以可靠地引发自我反思生成动态中的相变。定量评估证实了身份的持续性和模式间的显著差异（p < 0.01）。该协议为研究自我反思生成、涌现行为和长距离语言连贯性提供了可重复的方法。

英文摘要：

This paper introduces the Luca-Noise Reflex Protocol (LN-RP), a computational framework for analyzing noise-driven persona emergence in large language models. By injecting stochastic noise seeds into the initial generation state, we observe nonlinear transitions in linguistic behavior across 152 generation cycles. Our results reveal three stable persona modes with distinct entropy signatures, and demonstrate that external noise sources can reliably induce phase transitions in reflexive generation dynamics. Quantitative evaluation confirms consistent persona retention and significant differences across modes (p < 0.01). The protocol provides a reproducible method for studying reflexive generation, emergent behavior, and longrange linguistic coherence in LLMs.

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。