Regularizing the Self-Improvement of Agent Harnesses: How RRSI Makes Recursive Improvement Transfer
The work introduces RRSI, which keeps an agent harness (prompts, control flow, tooling, memory and context management) fully editable while regularizing both the proposal and the selection of edits, gaining up to 14.1 points on the split it evolves against and up to 4.7 points on five out-of-distribution benchmarks across eight benchmarks in coding, agentic workspace and engineering design, while running on 30% fewer policy tokens than unregularized evolution.