Mimosa Framework: Toward Evolving Multi-Agent Systems for Scientific Research

Current Autonomous Scientific Research (ASR) systems, despite leveraging large language models (LLMs) and agentic architectures, remain constrained by fixed workflows and toolsets that prevent adaptation to evolving tasks and environments. We introduce Mimosa, an evolving multi-agent framework that automatically synthesizes task-specific multi-agent workflows and iteratively refines them through experimental feedback. Mimosa leverages the Model Context Protocol (MCP) for dynamic tool discovery, generates workflow topologies via a meta-orchestrator, executes subtasks through code-generating agents that invoke available tools and scientific software libraries, and scores executions with an LLM-based judge whose feedback drives workflow refinement.

On ScienceAgentBench, Mimosa achieves a success rate of 43.1% with DeepSeek-V3.2, surpassing both single-agent baselines and static multi-agent configurations. Our results further reveal that models respond heterogeneously to multi-agent decomposition and iterative learning, indicating that the benefits of workflow evolution depend on the capabilities of the underlying execution model. Beyond these benchmarks, Mimosa's modular architecture and tool-agnostic design make it readily extensible, and its fully logged execution traces and archived workflows support auditability by preserving every analytical step for inspection and potential replication.

Combined with domain-expert guidance, the framework has the potential to automate a broad range of computationally accessible scientific tasks across disciplines. Released as a fully open-source platform, Mimosa aims to provide an open foundation for community-driven ASR. The framework is described in a 48-page document, including 4 figures, 1 table, and is available as a clean arXiv version prepared with main manuscript plus appendix/supplementary-style implementation details and prompt listings, dated 30 March 2026. The document is classified under the subjects of Artificial Intelligence, Machine Learning, and Multiagent Systems, with specific MSC and ACM classes. It can be cited as arXiv:2603.28986 [cs.AI] and has a DOI of https://doi.org/10.48550/arXiv.2603.28986.

The submission history of the document shows that it was submitted by Louis-Felix Nothias on Mon, 30 Mar 2026 20:35:57 UTC, with a file size of 1,548 KB.