RSI-Anything：用世界模型量化递归自我改进
递归自我改进能涨多少？RSI-Anything 实测：实验规划得分翻倍，化学产率 96% 超过贝叶斯优化。
研究提出 RSI-Anything 框架，在代码修复、GUI 智能体、实验规划和化学四个领域量化递归自我改进（RSI）能力。框架用预测性世界模型决定哪些动作和训练数据值得消耗昂贵的真实反馈，反馈同时更新策略和世界模型。实验规划任务中，五个策略版本迭代后得分从 21.80% 升至 46.01%，翻了一倍多。固定编码器的代码修复中，世界模型自适应把成功率从 26.3% 提到 31.2%；化学反应优化里，同样 40 次实验的平均最佳产率达 96.15%，高于贝叶斯优化的 82.16%。
Recursive self-improvement is one of the most discussed ideas in AI right now.
RSI-Anything measures it in code repair, GUI agents, lab planning, and chemistry.
They find that the lab-planning score more than doubles from 21.80% to 46.01% across five policy versions.
The framework involves a predictive world model that decides which actions and which training data are worth spending costly real feedback on. The feedback then updates both the policy and the world model.
A policy update is kept only if the updated world model still fits it. That check stops the world model from drifting out of date as the policy changes.
In code repair with a fixed coder, success goes from 26.3% with a static verifier to 31.2% after world-model adaptation. In reaction optimization, it reaches a mean best yield of 96.15%, compared with 82.16% for Bayesian optimization with the same 40 experiments.
https://t.co/2ts0DkosP7
Chat with Paper: https://t.co/d7PawuqVcu
