用持续状态更新替代模型参数，让智能体在长时任务中不断进化。
AutoFyn Technical Report: Non-Parametric Expert Iteration for Long-Horizon Agents
- 通过验证反馈更新持久状态，而非修改模型权重
- 在奥数、数据科学、网络安全任务中显著提升性能
- 适合需要长期迭代优化的复杂任务场景
我们提出AutoFyn，一种受专家迭代算法启发的智能体框架，通过在多轮中更新持久状态（如内存文件、报告、仓库状态）而非模型权重来适应冻结模型。每轮从新会话开始，协调器使用专业代理探索多种方案，任务验证器评估成果并提供客观奖励信号，该信号被提炼回持久状态以指导下一轮策略。本技术报告形式化了这一循环，描述其持久状态与验证接口，并在三个领域展示应用：在2026年国际数学奥林匹克六道新题上，所有可改进模型得分均高于原提供者代码代理；在Spider 2.0 dbt基准上构建出排名第一的智能体；并在Next.js、MetaMask、pnpm、Warp、LiteLLM、Langflow和Open WebUI中生成16份经维护者确认的漏洞通告。
原文摘要 · Abstract (English)
We introduce AutoFyn, an agent harness inspired by the Expert Iteration algorithm, adapting a frozen model across many rounds by updating persistent state from verified reward signals rather than model weights. Each round begins from a fresh model session, and durable information is reintroduced only through explicit interfaces such as persistent memory files, reports, and repository state. Within a round, an orchestrator explores, plans and builds many alternative approaches with specialized agents, while a task-grounded verifier verifies the work and supplies an objective reward for measuring progress. This reward is distilled back into the persistent state, which updates the effective policy for the next round. In this technical report, we formalize this loop and describe its persistent state and verification interfaces. We then demonstrate its use in three domains, namely olympiad mathematics, data science, and cybersecurity. On the six fresh problems of the 2026 International Mathematical Olympiad, every model with room to improve scores higher under AutoFyn than in its provider's own coding agent. AutoFyn also built the top-ranked agent on the Spider 2.0 dbt benchmark, and has produced $16$ maintainer-confirmed vulnerability advisories in Next.js, MetaMask, pnpm, Warp, LiteLLM, Langflow, and Open WebUI.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。