RRSI: Regularizing Recursive Self-Improvement Boosts OOD Agent Benchmarks Up to 4.7 Points

burny_tech · x · 2026-09-28

Researchers from Google Cloud AI Research, UNC, Stanford and WUSTL introduce RRSI (Regularized Recursive Self-Improvement of Agent Harnesses), addressing overfitting in recursively self-improving agents.

Problem: An LLM agent's capability is largely magnified by its harness (prompts, control flow, tooling, memory, context management). Existing methods iteratively edit harness components for recursive self-improvement, but often memorize training tasks—in-distribution gains shrink or vanish on out-of-distribution benchmarks.

Method: Regularization principles constrain both proposal and selection:

Results: Across 8 benchmarks spanning coding, agentic workspace, and engineering design, RRSI gains up to 14.1 points on its evolved split and up to 4.7 points on 5 OOD benchmarks, while using 30% fewer policy tokens than unregularized evolution.

Original post →

More from coding & agent

coding & agent channel →