{"slug": "metan-recursive-self-improvement-through-emergent-depth", "title": "Metaⁿ: Recursive Self-Improvement Through Emergent Depth", "summary": "Researchers introduced Meta^n, a recursive self-improvement system that keeps its meta-operation fixed and recurses on its input, achieving superior performance over prior self-improving agents on all eight benchmark families, including a nonzero score on ARC-AGI-2 where others scored zero. The system's depth is set by convergence, and ablations show most gains come from conditioning passed between layers, with distinct layer roles emerging without prescribed prompts.", "body_md": "# Computer Science > Artificial Intelligence\n\n[Submitted on 25 Aug 2026]\n\n# Title:Meta$^n$: Recursive Self-Improvement through Emergent Depth\n\n[View PDF](/pdf/2608.24735)\n\n[HTML (experimental)](https://arxiv.org/html/2608.24735v1)\n\nAbstract: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, $\\Omega$, 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 $\\Omega$ 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[this https URL]\n\n### Current browse context:\n\ncs.AI\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/metan-recursive-self-improvement-through-emergent-depth", "canonical_source": "https://arxiv.org/abs/2608.24735", "published_at": "2026-08-27 06:07:07+00:00", "updated_at": "2026-08-27 06:18:40.763026+00:00", "lang": "en", "topics": ["artificial-intelligence", "large-language-models", "ai-research", "ai-agents"], "entities": ["Meta^n", "ARC-AGI-2"], "alternates": {"html": "https://wpnews.pro/news/metan-recursive-self-improvement-through-emergent-depth", "markdown": "https://wpnews.pro/news/metan-recursive-self-improvement-through-emergent-depth.md", "text": "https://wpnews.pro/news/metan-recursive-self-improvement-through-emergent-depth.txt", "jsonld": "https://wpnews.pro/news/metan-recursive-self-improvement-through-emergent-depth.jsonld"}}