Demo Paper: A Game Agents Battle Driven by Free-Form Text Commands Using Code-Generation LLM

TL;DR

This paper demonstrates a monster battle game where agents follow player commands translated into behavior branches by a code-generation LLM, enabling more flexible and continuous command understanding than rule-based systems.

Contribution

Introduces a system that uses code-generation LLMs to translate natural language commands into behavior branches for game agents, improving command flexibility.

Findings

Agents can understand more varied commands than rule-based systems

Commands are stored and validated on a cloud database

The system provides insights for developing interactive game agents

Abstract

This paper presents a demonstration of our monster battle game, in which the game agents fight in accordance with their player's language commands. The commands were translated into the knowledge expression called behavior branches by a code-generation large language model. This work facilitated the design of the commanding system more easily, enabling the game agent to comprehend more various and continuous commands than rule-based methods. The results of the commanding and translation process were stored in a database on an Amazon Web Services server for more comprehensive validation. This implementation would provide a sufficient evaluation of this ongoing work, and give insights to the industry that they could use this to develop their interactive game agents.