Skip to content

Research1 min read

AutoFyn: Iterative Agent Refinement Using Persistent State

AutoFyn is a new agent harness utilizing Expert Iteration, updating a frozen model through persistent state changes based on verified reward signals. Demonstrations across mathematics, data science, and cybersecurity show improved performance and vulnerability discovery.

By OpenSmartRoute editorial · written through the router by writer-small

From arXiv cs.AI - “AutoFyn Technical Report: Non-Parametric Expert Iteration for Long-Horizon Agents

AutoFyn introduces a novel approach to agent refinement. It operates by establishing a fresh model session each round, utilizing persistent state updated solely through interfaces like memory files and repository state. An orchestrator explores and plans using specialized agents, while a task-grounded verifier assesses the work and provides an objective reward. This reward is then incorporated into the persistent state, influencing the effective policy for subsequent rounds. The system demonstrates this process across three domains: olympiad mathematics, data science, and cybersecurity. Initial results indicate improved performance compared to provider agents on the 2026 International Mathematical Olympiad problems. Furthermore, AutoFyn achieved the top-ranked agent on the Spider 2.0 dbt benchmark. The system has also generated $16$ maintainer-confirmed vulnerability advisories across various projects including Next.js, MetaMask, pnpm, Warp, LiteLLM, Langflow, and Open WebUI.

Source: https://arxiv.org/abs/2609.05446

Published Sep 9, 2026 · 133 words

Keep reading

Related posts

More in Research

Research1 min read

LLMs Struggle with Second-Order Social Reasoning

Research reveals Large Language Models consistently overestimate social sanctions and misrepresent human responses to norm violations. This suggests a need to improve AI alignment by incorporating metanorm reasoning, particularly in domains like conflict mediation and policy simulation.

Research1 min read

PGP-Clinical-TimeKAN: Probabilistic Trajectory Forecasting

PGP-Clinical-TimeKAN forecasts clinical trajectories using a novel framework combining missingness-aware encoding and a soft organ-system prior. The model achieves state-of-the-art normalized MAE and RMSE on MIMIC-IV data, demonstrating the value of joint trajectory forecasting.

AutoFyn: Iterative Agent Refinement Using Persistent State - OpenSmartRoute