UPLex: Fine-Grained Personality Control in Large Language Models via Unsupervised Lexical Modulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Tianlong, Liu, Wenhao, Wu, Muling, Dou, Shihan, Wang, Zhenghua, Lv, Changze, Wang, Xiaohua, Zheng, Xiaoqing, Huang, Xuanjing
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918137735151616
author Li, Tianlong
Liu, Wenhao
Wu, Muling
Dou, Shihan
Wang, Zhenghua
Lv, Changze
Wang, Xiaohua
Zheng, Xiaoqing
Huang, Xuanjing
author_facet Li, Tianlong
Liu, Wenhao
Wu, Muling
Dou, Shihan
Wang, Zhenghua
Lv, Changze
Wang, Xiaohua
Zheng, Xiaoqing
Huang, Xuanjing
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2310_16582
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle UPLex: Fine-Grained Personality Control in Large Language Models via Unsupervised Lexical Modulation
Li, Tianlong
Liu, Wenhao
Wu, Muling
Dou, Shihan
Wang, Zhenghua
Lv, Changze
Wang, Xiaohua
Zheng, Xiaoqing
Huang, Xuanjing
Computation and Language
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.
title UPLex: Fine-Grained Personality Control in Large Language Models via Unsupervised Lexical Modulation
topic Computation and Language
url https://arxiv.org/abs/2310.16582