UPLex：通过无监督词汇调制实现大型语言模型中的细粒度人格控制
UPLex: Fine-Grained Personality Control in Large Language Models via Unsupervised Lexical Modulation
AI总结：
提出UPLex方法，利用无监督构建的个性化词汇表在解码阶段动态调整词概率，实现对大型语言模型人格的细粒度操控，实验验证其有效性和可插拔性。
摘要：
人格是塑造人类沟通模式的关键因素，因此调节大型语言模型（LLMs）的人格在提升用户体验方面具有显著潜力。以往的方法要么依赖于在特定语料上对LLMs进行微调，要么需要手工设计提示词以激发LLMs的特定人格。然而，前者效率低下且成本高昂，而后者无法在细粒度层面上精确操控人格特质。为应对这些挑战，我们提出了UPLex，一种在解码阶段使用无监督构建的个性化词汇表（UPL）来操控LLM人格特质的方法。UPL可以从新构建的情境判断测试数据集中以无监督方式构建，并通过动态改变模型对即将出现单词的预测概率，以可插拔的方式调节LLM的人格表达。大量实验证明了我们方法在细粒度操控LLM人格方面的显著有效性和可插拔性。
英文摘要：
Personality is a crucial factor that shapes human communication patterns, thereby regulating the personalities of large language models (LLMs) holds significant potential in enhancing their user experiences. Previous approaches either relied on fine-tuning LLMs on specific corpora or required manually crafted prompts to evoke specific personalities from LLMs. However, the former is inefficient and costly, while the latter cannot precisely manipulate personality traits at a fine-grained level. To address these challenges, we propose UPLex, a method that uses an Unsupervisedly-Built Personalized Lexicon (UPL) during the decoding phase to manipulate LLM's personality traits. UPL can be constructed from a newly built situational judgment test dataset in an unsupervised fashion, and used to modulate the personality expression of LLMs by dynamically altering their predicted probability of upcoming words in a pluggable fashion. Extensive experimentation demonstrates the remarkable effectiveness and pluggability of our method for fine-grained manipulation of LLMs' personalities.
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。
