研究人员正在探索AI自我改进的新范式，超越以代理为中心的优化，转向以知识为中心的方法。一种方法是让代理将见解贡献给一个共享的、持久的知识库，然后可以将其用于未来的任务，从而实现更具可检查性和可转移性的改进。另一种方法是递归约束自我改进（RHI），它侧重于优化用户构建的约束作为代理循环的提示级规范，通过迭代反馈对其进行改进，以提高性能和可追溯性，用于未来的模型训练。这些方法旨在通过使改进更有效和跨不同模型和任务可移植来加速AI研发和预测未来能力。
AI
arXiv:2607.19592v1 Announce Type: new Abstract: Self-improving AI systems typically treat the agent as the object that improves, by optimizing prompts, workflows, harnesses, or even the agent's own code. This agent-centric view can make improvements expensive to maintain and diff…
METR (Model Evaluation & Threat Research)
TIER_1English(EN)·
<p>We (Parker and Tom) recently coauthored a paper, <a href="https://elasticity.institute/rsi-paper.pdf">“The Economics of Recursive Self-Improvement”</a>, with 7 other economists. The paper walks through a series of simple models of how AI may accelerate AI R&D, and we thoug…
Self-improving AI systems typically treat the agent as the object that improves, by optimizing prompts, workflows, harnesses, or even the agent's own code. This agent-centric view can make improvements expensive to maintain and difficult to transfer, because gains become tied to …
arXiv:2607.15524v1 Announce Type: cross Abstract: Under model--harness co-evolution, harnesses are not merely inference-time scaffolds but data-generating components whose execution traces can shape future foundation models. This motivates harness-in-the-loop learning: optimizing…
Under model--harness co-evolution, harnesses are not merely inference-time scaffolds but data-generating components whose execution traces can shape future foundation models. This motivates harness-in-the-loop learning: optimizing harnesses for both immediate agent performance an…
