Two new research papers explore hybrid approaches to interactive narrative systems, aiming to balance authorial control with player agency using Large Language Models (LLMs). The first paper introduces a framework that dynamically aligns pre-authored content with real-time LLM generation and uses a feedback-driven optimization process to refine prompts iteratively. The second paper proposes a platform with interchangeable LLM-based managers that operate over a shared symbolic world model, allowing for analysis and evaluation of properties like player agency and author intent across different management strategies. AI
IMPACT These research papers propose novel frameworks for creating more dynamic and controllable interactive narrative experiences using LLMs, potentially advancing the field of AI-driven storytelling.
RANK_REASON Two academic papers published on arXiv detailing new methods for interactive narrative systems using LLMs.
- alphaXiv
- arXiv
- CatalyzeX
- DagsHub
- Experience Manager
- Gotit.pub
- Hugging Face
- Jules Clerc
- Little Red Riding Hood
- LLMs
- ScienceCast
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →
