2026Google Researchopen source
Regularized Recursive Self-Improvement of Agent Harnesses (RRSI)
AI Agents
Recursive Self-Improvement
RRSI is a framework for the recursive self-improvement of LLM agent harnesses, designed to optimize prompts, control flow, tools, and memory. It addresses overfitting in agent evolution by regularizing both the proposal and selection processes, ensuring robust performance across diverse task domains.
Regularized search trajectory to prevent overfitting during agent evolution
Open edit space allowing modifications to prompts, tools, memory, and control flow
Git-based candidate management for auditability and version control
pricingFree & Open source
regionUnited States