優化器即智能體：跨提示、程式與機器學習工作流程的推理驅動搜尋

摘要

近期針對提示詞、程式與機器學習工作流進行最佳化的系統，通常依賴明確的外迴圈控制器，例如演化搜尋、強盜演算法或文本梯度方法。我們提出一個根本不同的問題：這種搜尋策略有多少部分可以被單一使用工具的智能體內部化？我們提出 ReASearch，一個以推理驅動最佳化的統一框架，其中智能體自主決定要評估什麼、如何診斷失敗、進行哪些修改，以及何時驗證或重新開始。ReASearch 不僅僅是藉由人工設計的啟發式方法所引導的提案生成器，它還會主動分析結果、分配預算，並透過持久記憶體在長時間跨度中不斷調整其策略。憑藉共享的智能體迴圈與特定領域工具，ReASearch 以完全相同的骨架來實例化最佳化提示詞、程式與機器學習工作流的流程。在 14 項多樣化任務中，它與專門的最佳化系統相比具有競爭力，且在多數情況下表現更佳，相較於強大的特定領域基線取得了 2% 至 40% 的提升，並在某些情況下發現了超越先前人類已知最佳結果的解決方案。關鍵的是，我們觀察到通常由明確控制器實作的複雜搜尋行為，會自然地從智能體的推理過程中湧現。

English

Recent systems for optimizing prompts, programs, and ML workflows typically rely on explicit outer-loop controllers such as evolutionary search, bandits, or textual-gradient methods. We ask a fundamentally different question: how much of this search policy can be internalized by a single tool-using agent? We present ReASearch, a unified framework for reasoning-driven optimization in which the agent autonomously decides what to evaluate, how to diagnose failures, which edits to make, and when to verify or restart. Rather than serving only as a proposal generator guided by hand-designed heuristics, the agent actively analyzes outcomes, allocates budget, and refines its strategy over long horizons through persistent memory. With a shared agent loop and domain-specific tools, ReASearch instantiates the exact same scaffold to optimize prompts, programs, and ML workflows. Across 14 diverse tasks, it is competitive with and mostly better than specialized optimization systems, achieving gains of 2% to 40% over strong domain-specific baselines, and in some cases discovering solutions that improve on prior human best-known results. Crucially, we observe that complex search behaviors, which are typically implemented by explicit controllers, emerge naturally from the agent's reasoning process.