This survey explores the burgeoning field of role-playing with language models, focusing on their development from early persona-based models to advanced character-driven simulations facilitated by Large Language Models LLMs. Initially confined to simple persona consistency due to limited model capabilities, role-playing tasks have now expanded to embrace complex character portrayals involving character consistency, behavioral alignment, and overall attractiveness.
We provide a comprehensive taxonomy of the critical components in designing these systems, including data, models and alignment, agent architecture and evaluation. This survey not only outlines the current methodologies and challenges, such as managing dynamic personal profiles and achieving high-level persona consistency but also suggests avenues for future research in improving the depth and realism of role-playing applications.
The goal is to guide future research by offering a structured overview of current methodologies and identifying potential areas for improvement. Today, most large language models LLMs Brown et al. An increasing number of individuals have been instructing LLMs to take on roles they desire, such as movie stars, game characters, or even their own relatives. If the standard assistant role of LLMs meets the demand for increased productivity, then LLMs for role-playing aims to fulfill human needs at a psychological and entertainment level.
This lively trend underscores the versatility of LLMs and the limitless potential of human imagination in the realm of artificial intelligence. The requirements for role-playing with language models differ significantly from those of a generic assistant. This expectation is also evident in the benchmarks for such models: people often desire an assistant with extensive professional knowledge and strong logical reasoning abilities.
In other words, humans expect the LLMs to interact with them in a manner consistent with a specific role. This expectation introduces a fascinating dynamic that can sometimes contradict the traditional notion of helpfulness.
