LLM-Based Authoring of Agent-Based Narratives through Scene Descriptions

TL;DR

This paper introduces a system that uses large language models to procedurally generate agent-based narratives from scene descriptions, enabling rapid prototyping of virtual agent interactions.

Contribution

It presents a novel system that converts scene descriptions into structured agent behaviors using LLMs, supporting dynamic interactions and ease of use.

Findings

LLMs reliably translate scene descriptions into agent behaviors.

System supports diverse interaction types and dynamic object manipulation.

Performance varies across different lightweight LLMs.

Abstract

This paper presents a system for procedurally generating agent-based narratives using large language models (LLMs). Users could drag and drop multiple agents and objects into a scene, with each entity automatically assigned semantic metadata describing its identity, role, and potential interactions. The scene structure is then serialized into a natural language prompt and sent to an LLM, which returns a structured string describing a sequence of actions and interactions among agents and objects. The returned string encodes who performed which actions, when, and how. A custom parser interprets this string and triggers coordinated agent behaviors, animations, and interaction modules. The system supports agent-based scenes, dynamic object manipulation, and diverse interaction types. Designed for ease of use and rapid iteration, the system enables the generation of virtual agent activity suitable for prototyping agent narratives. The performance of the developed system was evaluated using four popular lightweight LLMs. Each model's process and response time were measured under multiple complexity scenarios. The collected data were analyzed to compare consistency across the examined scenarios and to highlight the relative efficiency and suitability of each model for procedural agent-based narratives generation. The results demonstrate that LLMs can reliably translate high-level scene descriptions into executable agent-based behaviors.