awesome multi agent papers

Report Abuse

Basic Information

This repository is a curated, centralized compilation of significant research papers and references about multi-agent systems and multi-agent approaches built with large language models. Maintained by the Swarms team, it collects annotated entries and abstracts across topics such as collaboration and system design, frameworks and benchmarks, application-specific systems, evaluation and model improvement, social simulation, workflow and agent architecture, and many domain surveys. The README organizes papers by topical sections and typically notes abstracts and implementation pointers when available. The collection is intended as a literature hub for researchers, engineers, and practitioners interested in multi-agent research, toolchains, benchmarks, and emerging multi-agent design patterns. It also exposes contribution guidance and community touchpoints for keeping the list current.

Links

Categorization

App Details

Features
A topical, human-curated index of multi-agent papers organized into clear sections like collaboration, frameworks, applications, evaluation, social simulation, workflow, and surveys. Entries commonly include paper titles, author lists, short abstracts, and mentions of code or project repositories when provided. The repo supplies citation support via a bibtex file and documents format and contribution instructions for community additions. It highlights benchmarks, multi-agent frameworks, application domains such as healthcare, robotics, software engineering and finance, and lists surveys and evaluative studies. The README links to community and project resources, and the maintainers update the list to reflect recent publications and prominent implementations.
Use Cases
The collection reduces search overhead by aggregating important multi-agent literature, surveys, and benchmarks in one place so researchers and practitioners can quickly discover prior work and implementation resources. It helps newcomers learn the landscape via organized topical sections and annotated abstracts, assists literature reviews and citation gathering through the provided bibtex metadata, and surfaces benchmarks and open-source projects to reproduce or extend results. Community contribution guidance enables the list to grow with community-curated additions. Overall it supports faster onboarding, comparative study of methods, and identification of datasets, evaluation protocols, and code bases relevant to multi-agent research.

Please fill the required fields*