RRSI: Regularized Recursive Self-Improvement of Agent Harnesses

arXiv:2609.24972 · cs.LG, cs.AI, cs.CL · Submitted 2026-09-21 · Read on arXiv

cs.LG, cs.AI, cs.CL

Submitted: 2026-09-21

Updated: 2026-10-04

Code: https://github.com/google-research/rrsi

Project page: https://regularized-rsi.com

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Benchmark — 39× in this paper · explained in 2 episodes
• Regularized Recursive Self-Improvement of Agent Harnesses — 25× in this paper
• Pruning — 16× in this paper · explained in 4 episodes
• Regularization — 14× in this paper · explained in 3 episodes
• APEX — 9× in this paper · explained in 1 episode
• Agent Harness — 7× in this paper · explained in 1 episode
• Generalization — 6× in this paper · explained in 13 episodes
• Sparsity — 5× in this paper · explained in 4 episodes
• Adaptation — 4× in this paper · explained in 1 episode
• Benchmarking — 4× in this paper · explained in 3 episodes
• Credit assignment — 4× in this paper · explained in 3 episodes
• Verifier — 4× in this paper · explained in 2 episodes
• Test-Time Harness Evolution — 3× in this paper
• Ablation — 2× in this paper · explained in 1 episode
• Accuracy — 2× in this paper · explained in 1 episode
• Actor — 2× in this paper · explained in 3 episodes
• Admissibility — 2× in this paper · explained in 2 episodes
• LLM judge — 2× in this paper · explained in 2 episodes
• Reinforcement Learning — 2× in this paper · explained in 4 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Verification — 2× in this paper · explained in 2 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Agent Architectures — 1× in this paper · explained in 1 episode
• Agent Skills — 1× in this paper · explained in 1 episode
• Agentic Harness Engineering — 1× in this paper
• Agentic Problem — 1× in this paper · explained in 1 episode
• Agentic System — 1× in this paper · explained in 2 episodes
• Cross-Model Transfer — 1× in this paper · explained in 2 episodes
• LLM-as-a-Judge — 1× in this paper · explained in 3 episodes
• Online Adaptation — 1× in this paper · explained in 3 episodes
• Physical Constraints — 1× in this paper · explained in 7 episodes
• Provenance — 1× in this paper · explained in 3 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Self-Evolving Agents — 1× in this paper · explained in 2 episodes
• Steering — 1× in this paper · explained in 2 episodes

Sources

• HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry
• Frontier-Eng: Benchmarking Self-Evolving Agents on Real-World Engineering Tasks with Generative Optimization
• G-Zero: Self-Play for Open-Ended Generation from Zero Data
• Evo-Bench: Can Language Models Improve Agent Harness?
• Continual Harness: Online Adaptation for Self-Improving Foundation Agents
• EvoHarnessBench: Can Your Agents Keep Pace with an Evolving Harness?
• Recursive Harness Self-Improvement
• JobBench: Aligning Agent Work With Human Will
• Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
• Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents
• EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents
• Adaptive Auto-Harness: Sustained Self-Improvement for Agentic System Deployment on Open-Ended Task Streams
• AutoHarness: improving LLM agents by automatically synthesizing a code harness
• HarnessBank: Semantic Gene-Bank Search with Gated Verification for Agent-Harness Self-Evolution
• TTHE: Test-Time Harness Evolution
• Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
• Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving
• NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
• APEX-Agents

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks