arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2610.14843cs.AI
收藏

用于递归自我改进的预注册创世纪核算:物理发现运行时中首个经审计的算子创世纪事件

Pre-Registered Genesis Accounting for Recursive Self-Improvement: First Audited Operator-Genesis Events in a Physics-Discovery Runtime

HoomanBio AI Lab(HoomanBio AI实验室);Huaman Biotechnology (Beijing) Co., Ltd.(华曼生物科技(北京)有限公司)

Wenbin Mao, Pei Zou

AI总结:

本研究提出预注册两层核算方法,报告首个经审计的机器创建算子创世纪事件,在氢、氦等物理测量中生成标准库无法表达的算子,贡献为机器创建机制的测量与治理。

摘要:

递归自我改进的主张易于提出却难以证伪:这种改进是来自复用设计者提供的机制,还是来自系统自身创建的机制?我们提出一种预注册的两层核算方法:M1统计设计者提供词汇范围内的机器生成内容实例,M2统计机器创建的机制类型(注册到运行时自身词汇中的新算子,附带保留验证证书)。我们报告首个经审计的M2创世纪事件:在主要事件中,对基础算子进行深度受限的组合搜索,通过对真实NIST氢巴尔末测量值的保留R²评判,生成了算子sq(inv(n))=1/n²——即里德伯能级形式——该算子无法用任何标准库变换表达(最佳标准变换的保留R²为-0.546,而组合算子的保留R²为1.0000000)。审计中发现了一个协议缺陷(自变量由所测试定律的逆构建),并通过独立标签协议修复;对未见过的能级(n=8、9,训练于n≤7)的外推达到约10⁻⁶的相对误差,比最佳标准算子高出4至6个数量级。第二个事件从JPL星历向量中独立恢复了sqrt(inv(a))——开普勒速度定律。还纳入了:一个经审计的复合通道(首个领域级测量:真实保留领域上的准确率比为1.109倍)以及盲目的、去修饰的创世纪检查,其诚实计数为零。开普勒事件通过逐行审计;氢间隙在He II上重现;证书阈值稳定。与11个现有系统相比,本研究的贡献在于对机器创建机制的测量与治理,而非搜索算法或重新发现的物理规律。局限性:检查由知晓教科书答案、触发器手册、语法观察者注入内容的智能体进行;两个事件均展示了创建能力,尚未实现自主创世纪。

英文摘要:

Claims of recursive self-improvement are easy to make and hard to falsify: did improvement come from re-using designer-provided mechanisms, or from mechanisms the system authored itself? We introduce a pre-registered, two-tier accounting: M1 counts machine-generated content instances within a designer-provided vocabulary, while M2 counts machine-authored mechanism types (new operators registered into the runtime's own vocabulary, with a held-out-validation certificate). We report the first audited M2 genesis events. In the primary event, depth-limited compositional search over elementary operators, judged by held-out R^2 on real NIST hydrogen Balmer measurements, produced the operator sq(inv(n)) = 1/n^2 -- the Rydberg-level form -- which no standard-library transform could express (held-out R^2: -0.546 best standard transform vs. 1.0000000 composed). A protocol flaw (independent variable built by inverting the very law under test) was caught in audit and repaired via an independent-label protocol; extrapolation to unseen levels (n=8,9 trained on n<=7) attains relative error ~10^-6, four to six orders beyond the best standard operator. A second event independently recovered sqrt(inv(a)) -- Kepler's velocity law -- from JPL ephemeris vectors. Also incorporated: an audited compounding channel (first domain-level measurement: 1.109x accuracy ratio on a real held-out domain) and blind, decurated genesis exams whose honest tally is zero. The Kepler event passes line audit; the hydrogen gap reproduces on He II; the certificate is threshold-stable. Against eleven prior systems, the contribution is the measurement and governance of machine-authored mechanisms, not the search algorithm or rediscovered physics. Limitations: exams were curated by agents who knew the textbook answer, triggers manual, grammar observer-injected; both events demonstrate capability to author, not yet autonomous genesis.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑