EvoMaster: A Foundational Agent Framework for Building Evolving Autonomous Scientific Agents at Scale

2026-04-19

AI Summary

Current agent frameworks are predominantly static and domain-specific, lacking the capacity for continual learning and evolution, thereby hindering their applicability to large-scale autonomous scientific discovery. This work proposes SciMaster—the first cross-domain, extensible foundational framework for self-evolving scientific agents—leveraging a large language model–driven mechanism for continuous self-improvement. SciMaster enables iterative hypothesis generation, self-critique, and cross-task knowledge accumulation, allowing high-capability research agents to be deployed with only approximately 100 lines of code. The framework achieves state-of-the-art performance on four established benchmarks (41.1%, 75.8%, 73.3%, and 53.3%), outperforming the general-purpose baseline OpenClaw by 159% to 316%, thereby strongly validating its effectiveness and generalization capability.

Technology Category

• Multiagent Systems: Multiagent Learning
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Evolutionary Learning

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The convergence of large language models and agents is catalyzing a new era of scientific discovery: Agentic Science. While the scientific method is inherently iterative, existing agent frameworks are predominantly static, narrowly scoped, and lack the capacity to learn from trial and error. To bridge this gap, we present EvoMaster, a foundational evolving agent framework engineered specifically for Agentic Science at Scale. Driven by the core principle of continuous self-evolution, EvoMaster empowers agents to iteratively refine hypotheses, self-critique, and progressively accumulate knowledge across experimental cycles, faithfully mirroring human scientific inquiry. Crucially, as a domain-agnostic base harness, EvoMaster is exceptionally easy to scale up -- enabling developers to build and deploy highly capable, self-evolving scientific agents for arbitrary disciplines in approximately 100 lines of code. Built upon EvoMaster, we incubated the SciMaster ecosystem across domains such as machine learning, physics, and general science. Evaluations on four authoritative benchmarks (Humanity's Last Exam, MLE-Bench Lite, BrowseComp, and FrontierScience) demonstrate that EvoMaster achieves state-of-the-art scores of 41.1%, 75.8%, 73.3%, and 53.3%, respectively. It comprehensively outperforms the general-purpose baseline OpenClaw with relative improvements ranging from +159% to +316%, robustly validating its efficacy and generality as the premier foundational framework for the next generation of autonomous scientific discovery. EvoMaster is available at https://github.com/sjtu-sai-agents/EvoMaster.

Problem

• Agentic Science
• autonomous scientific agents
• self-evolution
• iterative learning
• scientific discovery

Innovation

• evolving agents
• agentic science
• self-evolution
• domain-agnostic framework
• autonomous scientific discovery