taichengguo/LLM_MultiAgents_Survey_Papers

What it solves

This project addresses the challenge of keeping track of the rapidly growing body of research on Large Language Model (LLM)-based multi-agent systems. It provides a centralized, curated repository of academic papers to help researchers and developers navigate the field.

How it works

The repository categorizes research into five primary streams: frameworks for building agents, orchestration and efficiency, problem-solving applications, world simulations, and datasets/benchmarks. It is maintained by the authors of a comprehensive survey paper on the topic and is updated bi-weekly to include the latest developments.

Who it’s for

It is designed for researchers, AI engineers, and students who want to explore how multiple LLM-based agents can collaborate, debate, or simulate complex environments to solve tasks.

Highlights

• A curated list of papers covering multi-agent frameworks, orchestration, and efficiency.
• Extensive coverage of application domains including software development, robotics, scientific research, and medical decision-making.
• A dedicated section for world simulation research, including social, economic, and game-theory scenarios.
• A collection of datasets and benchmarks specifically for evaluating multi-agent systems.
• Regular updates to ensure the collection remains current with the latest academic literature.