Thespian: Multi-Character Text Role-Playing Game Agents

TL;DR

This paper introduces Thespian, a multi-character text role-playing game agent capable of learning and emulating multiple characters using a soft prompt and attention mechanism, outperforming existing methods in multi-character and few-shot learning.

Contribution

We propose a novel Thespian framework with a soft prompt and attention mechanism enabling multi-character and few-shot learning in text role-playing agents.

Findings

Outperforms state-of-the-art in multi-character learning

Effective few-shot character learning demonstrated

Framework supports dynamic character emulation

Abstract

Text-adventure games and text role-playing games are grand challenges for reinforcement learning game playing agents. Text role-playing games are open-ended environments where an agent must faithfully play a particular character. We consider the distinction between characters and actors, where an actor agent has the ability to play multiple characters. We present a framework we call a thespian agent that can learn to emulate multiple characters along with a soft prompt that can be used to direct it as to which character to play at any time. We further describe an attention mechanism that allows the agent to learn new characters that are based on previously learned characters in a few-shot fashion. We show that our agent outperforms the state of the art agent framework in multi-character learning and few-shot learning.