PaperPulse logo
FeedTopicsAI Researcher FeedBlogPodcastAccount

Stay Updated

Get the latest research delivered to your inbox

Platform

  • Home
  • About Us
  • Search Papers
  • Research Topics
  • Researcher Feed

Resources

  • Newsletter
  • Blog
  • Podcast
PaperPulse•

AI-powered research discovery platform

© 2024 PaperPulse. All rights reserved.

AdaMARP: An Adaptive Multi-Agent Interaction Framework for General Immersive Role-Playing

ArXivSource

Zhenhua Xu, Dongsheng Chen, Shuo Wang, Jian Li, Chengjie Wang, Meng Han, Yabiao Wang

cs.AI
cs.CL
|
Jan 16, 2026
1,975 views

One-line Summary

AdaMARP is an adaptive multi-agent role-playing framework that improves immersion and adaptability in interactive narratives by integrating dynamic scene management and character interactions.

Plain-language Overview

Role-playing with AI often feels limited because the systems don't adapt well to changes in the story or environment. AdaMARP is a new framework designed to make AI role-playing more immersive and flexible by allowing for dynamic scene changes and character interactions. It uses a special message format to combine thoughts, actions, environment, and speech, and includes a Scene Manager to handle scene transitions and character introductions. Tests show that AdaMARP improves the consistency and flow of stories, outperforming some commercial AI models.

Technical Details