Conversational agents in games

Abstract

Natural language processing(NLP) can potentially apply to various fields. As a sub-field of artificial intelligence, NLP focuses on the interaction between the computer and the user by generating and interpreting the natural language. Applying NLP to games is a challenge for both AI and game fields, potentially enhancing the gameplay experience. Conversational agents as one of the popular topics that is related to NLP have been more widely used in researching the natural language processing(NLP) field. Researching the dialogue agent work with the player on Minecraft via Malmo to perform interaction and collaboration abilities between the dialogue agent and the player during gameplay within Minecraft environment is an attractive research field with increasing potential efficiency in enhancing the interaction between the dialogue agent and the player during gameplay. Introducing the dialogue agent into Minecraft environment could help us explore the potential possibilities of NLP in AI fields. In this paper I will discuss how to make the dialogue agent collaborate with the player for construction including the execution within Minecraft environment, the exploration of the large language model's(LLM) interaction, and the collaboration utility in Minecraft via Malmo.