A comprehensive survey project collecting, categorizing, and organizing academic research on Large Language Model-based multi-agent systems, documenting progress and challenges in this rapidly evolving field.
One-Minute Overview#
This is a research paper collection project focused on Large Language Model (LLM)-based multi-agent systems. Maintained by researchers, it systematically collects and categorizes the latest research findings in this field into five main directions: multi-agent frameworks, orchestration and efficiency, problem-solving, world simulation, and datasets and benchmarks. This project is ideal for scholars, developers, and researchers interested in multi-agent systems and cutting-edge AI research.
Core Value: Provides researchers with a systematic literature review and collection of the latest research findings, helping them quickly understand the research progress and future directions in this field.
Getting Started#
Installation Difficulty: Low - This is a literature survey project requiring no installation,可直接访问和使用
# Clone the repository to access the complete paper list
git clone https://github.com/taichengguo/LLM_MultiAgents_Survey_Papers.git
Is this suitable for me?
- ✅ Academic Research: Researchers looking for literature on LLM-based multi-agent systems
- ✅ Technology Investigation: Developers and product managers wanting to understand the latest developments in multi-agent systems
- ✅ Course Design: Educators needing relevant literature to support teaching or curriculum design
- ❌ Production Deployment: This is not a deployable software project but a literature resource
- ❌ Code Development: While containing some research frameworks, it primarily provides literature rather than directly usable code
Core Capabilities#
1. Systematic Literature Collection - Comprehensive Coverage of Research Frontiers#
- The project continuously collects and organizes the latest research findings on LLM-based multi-agent systems
- Systematically categorized by research direction for easy location of relevant literature Actual Value: Researchers save significant time searching for scattered literature, accessing core research findings in one place
2. Five Research Direction Categories - Structured Knowledge Organization#
- Multi-Agent Frameworks: Research on basic architectures for building multi-agent systems
- Orchestration and Efficiency: Focus on system coordination and performance optimization
- Problem Solving: Solutions applied to specific problem domains
- World Simulation: Applications in various fields like society, games, psychology
- Datasets and Benchmarks: Resources and standards for evaluation and comparison Actual Value: Helps researchers understand the application scenarios and research focus of LLM-based multi-agent systems from different perspectives
3. Regular Update Mechanism - Tracking Latest Research Progress#
- Planned updates every two weeks to the paper list
- All included papers will be incorporated into the next survey version
- Open to community contributions, encouraging submission of important missing literature 实际价值: Researchers can access the latest research findings in a timely manner, avoiding information lag
Maintenance Status#
- Development Activity: Actively maintained with regular updates to the paper list, planned every two weeks
- Recent Updates: Frequent updates include the latest research findings from March 2024
- Community Response: Encourages community contributions through pull requests and issues, fostering a collaborative ecosystem
Documentation and Learning Resources#
- Documentation Quality: Comprehensive and detailed, providing categorized indexes and links to survey papers
- Official Documentation: Project README and survey paper https://arxiv.org/abs/2402.01680
- Example Code: Not applicable (literature survey project)
- Learning Resources: Includes survey paper, categorized paper index, latest research trend visualization, and system architecture diagrams