AgentRxiv: Towards Collaborative Autonomous Research
Samuel Schmidgall, Michael Moor
2025-03-25
Summary
This paper explores how AI agents can work together on research projects, sharing their findings and building on each other's work to speed up scientific discovery.
What's the problem?
Currently, AI agents conduct research in isolation, without the ability to learn from or collaborate with other AI agents, which limits their progress.
What's the solution?
The researchers created a system called AgentRxiv that allows AI agents to upload and access research reports, enabling them to collaborate, share insights, and iteratively improve on each other's work.
Why it matters?
This work matters because it suggests that AI agents can play a significant role in future scientific research, working alongside humans to accelerate discovery and innovation.
Abstract
Progress in scientific discovery is rarely the result of a single "Eureka" moment, but is rather the product of hundreds of scientists incrementally working together toward a common goal. While existing agent workflows are capable of producing research autonomously, they do so in isolation, without the ability to continuously improve upon prior research results. To address these challenges, we introduce AgentRxiv-a framework that lets LLM agent laboratories upload and retrieve reports from a shared preprint server in order to collaborate, share insights, and iteratively build on each other's research. We task agent laboratories to develop new reasoning and prompting techniques and find that agents with access to their prior research achieve higher performance improvements compared to agents operating in isolation (11.4% relative improvement over baseline on MATH-500). We find that the best performing strategy generalizes to benchmarks in other domains (improving on average by 3.3%). Multiple agent laboratories sharing research through AgentRxiv are able to work together towards a common goal, progressing more rapidly than isolated laboratories, achieving higher overall accuracy (13.7% relative improvement over baseline on MATH-500). These findings suggest that autonomous agents may play a role in designing future AI systems alongside humans. We hope that AgentRxiv allows agents to collaborate toward research goals and enables researchers to accelerate discovery.