CREA: A Collaborative Multi-Agent Framework for Creative Content Generation with Diffusion Models

2025-04-07

AI Summary

To address the limited creative editing capability in AI-based image generation, this paper introduces the novel task of “creative editing,” transcending conventional prompt-driven paradigms. Methodologically, we propose a diffusion-based multi-agent collaborative framework comprising four specialized agents—Curator, Generator, Critic, and Optimizer—that emulate the human creative loop via dynamic collaboration protocols and iterative feedback. Our core contribution lies in modeling creativity as an interpretable, controllable multi-agent process, thereby unifying originality, semantic consistency, and artistic intent. Extensive experiments demonstrate that our approach significantly outperforms state-of-the-art methods in diversity, semantic alignment, and quality of creative transformations. Both quantitative metrics and qualitative analysis confirm its superior expressiveness and interpretability.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Computer Vision: Diffusion Models for Vision
• Multiagent Systems: Adversarial Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Creativity in AI imagery remains a fundamental challenge, requiring not only the generation of visually compelling content but also the capacity to add novel, expressive, and artistically rich transformations to images. Unlike conventional editing tasks that rely on direct prompt-based modifications, creative image editing demands an autonomous, iterative approach that balances originality, coherence, and artistic intent. To address this, we introduce CREA, a novel multi-agent collaborative framework that mimics the human creative process. Our framework leverages a team of specialized AI agents who dynamically collaborate to conceptualize, generate, critique, and enhance images. Through extensive qualitative and quantitative evaluations, we demonstrate that CREA significantly outperforms state-of-the-art methods in diversity, semantic alignment, and creative transformation. By structuring creativity as a dynamic, agentic process, CREA redefines the intersection of AI and art, paving the way for autonomous AI-driven artistic exploration, generative design, and human-AI co-creation. To the best of our knowledge, this is the first work to introduce the task of creative editing.

Problem

• Enhancing AI creativity in generating novel, expressive image transformations
• Developing autonomous iterative approach for artistic image editing
• Introducing multi-agent collaboration to mimic human creative process

Innovation

• Multi-agent framework for creative content generation
• Dynamic collaboration among specialized AI agents
• Autonomous iterative approach balancing originality and coherence