LLM Reasoner and Automated Planner: A new NPC approach

TL;DR

This paper introduces a novel architecture combining Large Language Models with classical automated planners to enable intelligent agents to make human-like decisions and generate sound plans in unpredictable situations.

Contribution

It presents a new integrated approach leveraging LLMs and planners to improve decision-making and planning in NPCs for more realistic behavior.

Findings

Enhanced decision-making capabilities in NPCs.

Ability to generate sound plans for unanticipated situations.

Improved human-like behavior in simulated agents.

Abstract

In domains requiring intelligent agents to emulate plausible human-like behaviour, such as formative simulations, traditional techniques like behaviour trees encounter significant challenges. Large Language Models (LLMs), despite not always yielding optimal solutions, usually offer plausible and human-like responses to a given problem. In this paper, we exploit this capability and propose a novel architecture that integrates an LLM for decision-making with a classical automated planner that can generate sound plans for that decision. The combination aims to equip an agent with the ability to make decisions in various situations, even if they were not anticipated during the design phase.