SIA：具備框架與權重更新的自我改進人工智慧

摘要

人類是構建與改進AI的瓶頸。無論是模型本身，或是包裹模型的智能體，皆由人類編寫、調整與修正。能夠自行找出改進方法的AI，其長遠目標至今仍未達成。兩個大致互不相關的研究方向正試圖突破此瓶頸。「框架更新學派」讓元智能體改寫任務特定智能體的支架（其工具、提示詞、重試邏輯與搜尋程序），同時固定模型權重。「測試時訓練學派」則使用手寫的強化學習管線，根據任務回饋更新模型本身的權重，同時固定支架。這兩個領域各自獨立運作。我們提出SIA，一種自我改進循環，其中語言模型智能體（回饋智能體）同時更新任務特定智能體的支架與權重。我們在三個截然不同的領域進行評估：中國法律罪名分類、低階GPU核心最佳化，以及單細胞RNA去噪。在所有三項基準測試中，結合兩種槓桿的表現皆優於僅更新支架的迭代。在LawBench上提升56.6%，GPU核心執行時間減少91.9%，去噪效果相較初始基準提升502%。支架更新使模型具備智能體特性，塑造其搜尋與行動方式；而權重更新則建構出任何提示詞或支架都無法灌輸的領域直覺。

English

Humans are the bottleneck in building and improving AI. Both the models and the agents that wrap them are written, tuned, and corrected by people. The long-horizon goal of an AI that can figure out how to improve itself remains open. Two largely disjoint research lines attack this bottleneck. The harness-update school has a meta-agent rewrite the scaffold of a task-specific agent (its tools, prompts, retry logic, and search procedure) while the model weights are held fixed. The test-time training school uses hand-written RL pipelines to update the model's own weights on task feedback while the harness is held fixed. These two silos operate in isolation. We propose SIA, a self-improving loop in which a language-model agent (the Feedback-Agent) updates both the harness and the weights of a task-specific agent. We evaluate across three contrasting domains: Chinese legal charge classification, low-level GPU kernel optimisation, and single-cell RNA denoising. Combining both levers outperforms scaffold iteration alone on all three benchmarks. The gains are 56.6% on LawBench, 91.9% runtime reduction on GPU kernels, and 502% on denoising over the initial baseline. Harness updates make the model agentic, shaping how it searches and acts, while weight updates build the domain intuition that no prompt or scaffold can instil.