# Google ScientistTwo 论文展示 AI 研究循环的自我改进能力

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-09-20 07:11
- AIHOT 分数：71
- AIHOT 链接：https://aihot.news/items/cmu905sdx035arojr31ot3z3z
- 原文链接：https://x.com/rohanpaul_ai/status/2101449041091657746

## AI 摘要

Google 发布 ScientistTwo 论文，展示 AI 研究中的递归自我改进：模型先改进人类方法，再以自身发现为新基线继续改进。它以提出想法、实验验证、剔除无效方案并根据评审反馈开启新一轮的循环方式，自主改进了 107 个人类定义 ML 问题中的 86 个；在基于 ICLR、ICML 和 NeurIPS 论文的问题上，报告了 80.4% 的成功率和相对原始人类基线平均 25.2% 的改进。

## 正文

Beautiful paper from Google.

ScientistTwo shows another progress of recursive self-improvement in AI research: it can improve a human method, then use its own discovery as the baseline and improve it again.

the research assistant becomes the research loop: it autonomously improved 86 of 107 human-defined ML problems, suggesting experimentation can be automated long before scientific judgment can.

Instead of helping with one task, ScientistTwo runs a loop: propose ideas, test them, remove what does not help, and use reviewer feedback to start another round.

Across problems based on ICLR, ICML, and NeurIPS papers, it reports an 80.4% success rate and a 25.2% average relative improvement over the original human baselines.