通过递归输入实现深度自我改进，突破传统元认知层级限制。
Meta$^n$: Recursive Self-Improvement through Emergent Depth
- 固定元操作Ω，反复作用于自身输出，逐层生成策略与工具库。
- 在8个基准上超越已有自进化模型，ARC-AGI-2唯一得分超零。
- 无需预设分层角色，深度由收敛决定，适合复杂推理任务研究者。
自进化大模型代理通常优化答案而非生成过程。现有系统将元层固定，或在自修改时保留部分编辑机制以维持稳定，导致元深度上限约两层。我们提出Meta$^n$，保持元操作Ω不变，转而对输入进行递归。该操作反复作用于自身产物，读取下层求解栈的痕迹与生成代码，并写入下一层次作为战略预处理和可调用工具库。由于Ω恒定，系统不致失稳；因输入持续增长，每层均从更高视角推理。深度由收敛决定，而非预先设定，演化存档搜索层链组合。在两种骨干模型下，Meta$^n$ 在全部八个基准家族中表现优于先前自进化代理。最显著案例为对抗技能记忆的ARC-AGI-2，仅其得分高于零。消融实验表明，递归收益主要来自各层间的条件传递，且随深度自然涌现不同角色，无显式提示指定。
原文摘要 · Abstract (English)
Self-improving LLM agents refine answers, not the process that produces those answers. Systems that add a meta-level hold that level fixed, and those that edit themselves must leave part of their own editing machinery untouched to stay stable, capping the meta-depth they realize at roughly two. We present Meta$^n$, which keeps the meta-operation fixed and recurses on its input instead. That operation, $Ω$, is applied repeatedly to its own products, reading the traces of the solver stack below together with the code that produced them, then writing the next layer as a strategic pre-process and a library of callable helpers. Because $Ω$ never changes, it cannot destabilize the system, and because its input strictly grows, each layer reasons from a higher vantage than the last. Depth is set by convergence rather than fixed in advance, and an evolutionary archive searches over layer chains. Across two backbones, Meta$^n$ outperforms prior self-improving agents on all eight benchmark families. The sharpest case is ARC-AGI-2, built to resist skill memorization, where it alone scores above zero. Ablations indicate that most of the gain from recursion comes from the conditioning each layer passes to the next, and distinct layer roles emerge with depth although no prompt prescribes them. Code available at https://github.com/minnesotanlp/meta-n
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。