Player-Driven Emergence in LLM-Driven Game Narrative

2404.17027

YC

118

Reddit

0

Published 6/5/2024 by Xiangyu Peng, Jessica Quaye, Sudha Rao, Weijia Xu, Portia Botchway, Chris Brockett, Nebojsa Jojic, Gabriel DesGarennes, Ken Lobb, Michael Xu and 3 others
Player-Driven Emergence in LLM-Driven Game Narrative

Abstract

We explore how interaction with large language models (LLMs) can give rise to emergent behaviors, empowering players to participate in the evolution of game narratives. Our testbed is a text-adventure game in which players attempt to solve a mystery under a fixed narrative premise, but can freely interact with non-player characters generated by GPT-4, a large language model. We recruit 28 gamers to play the game and use GPT-4 to automatically convert the game logs into a node-graph representing the narrative in the player's gameplay. We find that through their interactions with the non-deterministic behavior of the LLM, players are able to discover interesting new emergent nodes that were not a part of the original narrative but have potential for being fun and engaging. Players that created the most emergent nodes tended to be those that often enjoy games that facilitate discovery, exploration and experimentation.

Get summaries of the top AI research delivered straight to your inbox:

Overview

  • This paper explores the use of large language models (LLMs) to drive the narrative in a video game called "Dejaboom!"
  • The researchers investigate how player actions and choices can lead to emergent gameplay and narrative outcomes, enabled by the LLM-powered game engine.
  • Key areas examined include the integration of LLMs into game development, the impact on player agency and experience, and the potential for LLM-driven games to foster more dynamic and personalized storytelling.

Plain English Explanation

The paper discusses a video game called "Dejaboom!" that uses a large language model (LLM) to generate the game's narrative. LLMs are advanced AI systems that can understand and generate human-like text. In this game, the LLM is responsible for creating the story and dialogue, reacting to the player's actions and choices.

The researchers wanted to see how this LLM-powered approach could lead to more dynamic and personalized gameplay experiences, where the story unfolds in unexpected ways based on the player's decisions. Rather than having a fixed narrative, the game's plot and characters evolve in response to the player's choices, creating a more immersive and interactive experience.

By leveraging the flexibility and creativity of LLMs, the developers aimed to move beyond traditional, predetermined game narratives and enable a more player-driven, emergent storytelling experience. The paper explores the technical implementation, the impact on player agency, and the potential for LLM-driven games to offer more engaging and tailored experiences for players.

Technical Explanation

The researchers developed the "Dejaboom!" game using a large language model (LLM) to generate the narrative and dialogue. The LLM was integrated into the game's engine, allowing it to dynamically respond to player actions and choices, shaping the story in real-time.

The game's architecture involves a game engine that interfaces with the LLM, passing relevant information (e.g., player choices, game state) to the LLM, which then generates the appropriate narrative content. This allows the game's story to evolve organically, with the LLM generating new plot points, characters, and dialogue based on the player's interactions.

The researchers conducted experiments to assess the impact of this LLM-driven approach on player agency and narrative emergence. They found that players experienced a greater sense of agency and felt more invested in the story, as their choices had a tangible impact on the game's progression and outcome. Additionally, the researchers observed emergent narrative patterns that were not pre-scripted, demonstrating the LLM's ability to generate novel and engaging story elements.

Critical Analysis

The paper highlights the potential of LLM-driven game narratives to provide more dynamic and personalized experiences for players. However, the researchers also acknowledge some limitations and areas for further exploration.

One concern is the reliability and consistency of the LLM-generated content. As language models can sometimes produce nonsensical or contradictory outputs, the researchers note the importance of careful model fine-tuning and monitoring to ensure the narrative remains coherent and aligned with the game's objectives.

Additionally, the paper suggests the need for further research on the long-term player engagement and narrative satisfaction in LLM-driven games. While the initial experiments demonstrate increased player agency, the sustainability of this approach and its ability to maintain player interest over extended playthroughs require additional investigation.

The researchers also highlight the potential ethical considerations around the use of LLMs in game development, such as the potential for biased or offensive content generation. Addressing these concerns and developing robust content moderation strategies will be crucial as this technology becomes more widely adopted in the gaming industry.

Conclusion

The paper presents an innovative approach to game narrative development by leveraging large language models (LLMs) to enable more player-driven, emergent storytelling experiences. The "Dejaboom!" game showcases how LLMs can be integrated into game engines to dynamically generate narrative content in response to player actions, fostering a greater sense of agency and personalization.

The research findings suggest that LLM-driven game narratives have the potential to offer more engaging and immersive experiences for players, moving beyond the limitations of traditional, predetermined storylines. However, the paper also highlights the need for further exploration of the reliability, long-term player engagement, and ethical considerations surrounding the use of LLMs in game development.

As language models continue to advance, the integration of LLMs into game design could revolutionize the way players experience and shape interactive narratives. This work lays the groundwork for future research and development in the promising field of LLM-driven game narrative design.



This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!

Related Papers

Generating Games via LLMs: An Investigation with Video Game Description Language

Generating Games via LLMs: An Investigation with Video Game Description Language

Chengpeng Hu, Yunlong Zhao, Jialin Liu

YC

0

Reddit

0

Recently, the emergence of large language models (LLMs) has unlocked new opportunities for procedural content generation. However, recent attempts mainly focus on level generation for specific games with defined game rules such as Super Mario Bros. and Zelda. This paper investigates the game generation via LLMs. Based on video game description language, this paper proposes an LLM-based framework to generate game rules and levels simultaneously. Experiments demonstrate how the framework works with prompts considering different combinations of context. Our findings extend the current applications of LLMs and offer new insights for generating new games in the area of procedural content generation.

Read more

5/31/2024

Automatic Bug Detection in LLM-Powered Text-Based Games Using LLMs

New!Automatic Bug Detection in LLM-Powered Text-Based Games Using LLMs

Claire Jin, Sudha Rao, Xiangyu Peng, Portia Botchway, Jessica Quaye, Chris Brockett, Bill Dolan

YC

0

Reddit

0

Advancements in large language models (LLMs) are revolutionizing interactive game design, enabling dynamic plotlines and interactions between players and non-player characters (NPCs). However, LLMs may exhibit flaws such as hallucinations, forgetfulness, or misinterpretations of prompts, causing logical inconsistencies and unexpected deviations from intended designs. Automated techniques for detecting such game bugs are still lacking. To address this, we propose a systematic LLM-based method for automatically identifying such bugs from player game logs, eliminating the need for collecting additional data such as post-play surveys. Applied to a text-based game DejaBoom!, our approach effectively identifies bugs inherent in LLM-powered interactive games, surpassing unstructured LLM-powered bug-catching methods and filling the gap in automated detection of logical and design flaws.

Read more

6/10/2024

👁️

StoryVerse: Towards Co-authoring Dynamic Plot with LLM-based Character Simulation via Narrative Planning

Yi Wang, Qian Zhou, David Ledo

YC

0

Reddit

0

Automated plot generation for games enhances the player's experience by providing rich and immersive narrative experience that adapts to the player's actions. Traditional approaches adopt a symbolic narrative planning method which limits the scale and complexity of the generated plot by requiring extensive knowledge engineering work. Recent advancements use Large Language Models (LLMs) to drive the behavior of virtual characters, allowing plots to emerge from interactions between characters and their environments. However, the emergent nature of such decentralized plot generation makes it difficult for authors to direct plot progression. We propose a novel plot creation workflow that mediates between a writer's authorial intent and the emergent behaviors from LLM-driven character simulation, through a novel authorial structure called abstract acts. The writers define high-level plot outlines that are later transformed into concrete character action sequences via an LLM-based narrative planning process, based on the game world state. The process creates living stories that dynamically adapt to various game world states, resulting in narratives co-created by the author, character simulation, and player. We present StoryVerse as a proof-of-concept system to demonstrate this plot creation workflow. We showcase the versatility of our approach with examples in different stories and game environments.

Read more

5/24/2024

🤿

GENEVA: GENErating and Visualizing branching narratives using LLMs

Jorge Leandro, Sudha Rao, Michael Xu, Weijia Xu, Nebosja Jojic, Chris Brockett, Bill Dolan

YC

0

Reddit

0

Dialogue-based Role Playing Games (RPGs) require powerful storytelling. The narratives of these may take years to write and typically involve a large creative team. In this work, we demonstrate the potential of large generative text models to assist this process. textbf{GENEVA}, a prototype tool, generates a rich narrative graph with branching and reconverging storylines that match a high-level narrative description and constraints provided by the designer. A large language model (LLM), GPT-4, is used to generate the branching narrative and to render it in a graph format in a two-step process. We illustrate the use of GENEVA in generating new branching narratives for four well-known stories under different contextual constraints. This tool has the potential to assist in game development, simulations, and other applications with game-like properties.

Read more

6/7/2024