发表机构
University of Maryland, Baltimore County; Emergence AI（马里兰大学巴尔的摩县分校; Emergence AI）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对标准递归自我改进（RSI）仅做一阶表面修补的局限，本文提出SO-RSI框架，通过监控执行轨迹中的结构异常并执行诊断探针，进行二阶因果调查，从而在Lean 4证明和Verus代码生成任务上分别提升留出集通过率21.8和25.8个百分点，并抑制失败重复。
AI 中文摘要
递归自我改进（RSI）使智能体能够通过执行反馈迭代优化其工作流程。然而，标准RSI通常作为一阶优化器运行：它反复修补表面参数以应对即时失败症状，常常导致试错式折腾而未能解决底层机制。为解决此局限，我们引入SO-RSI，一个将工作流程优化提升至二阶诊断探究的框架，在承诺结构性干预之前调查失败发生的原因。SO-RSI被动监控执行轨迹中的三种结构异常（重复、对立编辑和期望不匹配），以触发针对性的机制调查。通过执行轻量级诊断探针并在RSI轮次间维持持久查询记忆，SO-RSI积累因果证据以指导系统性的工作流程编辑而非参数修补。在Lean 4证明生成和基于Verus的可验证代码生成中，在匹配的24小时搜索预算下，SO-RSI相较于朴素RSI将最终留出集通过率分别提高了21.8和25.8个百分点。行为分析进一步证实，SO-RSI显著抑制了失败重复并消除了无成效的零进展优化循环。
英文摘要
Recursive self-improvement (RSI) enables agents to iteratively optimize their workflows via execution feedback. However, standard RSI typically operates as a first-order optimizer: it repeatedly patches surface-level parameters in response to immediate failure symptoms, often leading to trial-and-error thrashing without resolving underlying mechanisms. To address this limitation, we introduce SO-RSI, a framework that elevates workflow optimization to a second-order diagnostic inquiry, investigating why failures occur before committing to structural interventions. SO-RSI passively monitors execution traces for three structural anomalies (recurrence, opposing edits, and expectation mismatch) to trigger targeted mechanism investigations. By executing lightweight diagnostic probes and maintaining persistent inquiry memory across RSI rounds, SO-RSI accumulates causal evidence to guide systematic workflow edits rather than parameter patches. Across Lean 4 proof generation and Verus-based verifiable code generation, SO-RSI improves final held-out pass rates over Naive RSI by 21.8 and 25.8 percentage points under matched 24-hour search budgets. Behavioral analyses further confirm that SO-RSI substantially suppresses failure recurrence and eliminates unproductive zero-progress optimization loops.
