Dynamic Personality Adaptation in Large Language Models via State Machines

TL;DR

This paper introduces a modular framework for dynamic personality adaptation in large language models using state machines and continuous scoring, enhancing interaction quality in complex dialogues.

Contribution

It presents a novel, architecture-agnostic approach employing state machines and a modular scoring pipeline for real-time personality adaptation in LLMs.

Findings

Successfully adapts personality states to user inputs

Influences user behavior to facilitate training scenarios

Maintains scoring precision with lightweight classifiers

Abstract

The inability of Large Language Models (LLMs) to modulate their personality expression in response to evolving dialogue dynamics hinders their performance in complex, interactive contexts. We propose a model-agnostic framework for dynamic personality simulation that employs state machines to represent latent personality states, where transition probabilities are dynamically adapted to the conversational context. Part of our architecture is a modular pipeline for continuous personality scoring that evaluates dialogues along latent axes while remaining agnostic to the specific personality models, their dimensions, transition mechanisms, or LLMs used. These scores function as dynamic state variables that systematically reconfigure the system prompt, steering behavioral alignment throughout the interaction.We evaluate this framework by operationalizing the Interpersonal Circumplex (IPC) in a medical education setting. Results demonstrate that the system successfully adapts its personality state to user inputs, but also influences user behavior, thereby facilitating de-escalation training. Notably, the scoring pipeline maintains comparable precision even when utilizing lightweight, fine-tuned classifiers instead of large-scale LLMs. This work demonstrates the feasibility of modular, personality-adaptive architectures for education, customer support, and broader human-computer interaction.