发表机构
Hefei National Laboratory; Institute for Advanced Algorithms Research; University of Science and Technology of China; Institute of Physics, Chinese Academy of Sciences; Institute of Theoretical Physics, Chinese Academy of Sciences（合肥国家实验室; 先进算法研究所; 中国科学技术大学; 中国科学院物理研究所; 中国科学院理论物理研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出一个智能体驱动的环境，通过统一微分接口连接需求识别、开发与质量评估，在20个科学软件包上实现自动微分扩展，显著节省计算并支持跨任务复用，从而递归改进可微分科学软件生态系统。
AI 中文摘要
可微分编程将科学计算与基于梯度的推理、学习和设计联系起来。在异构软件生态系统中扩展这些能力需要专门的工作来实现导数、集成接口和评估质量。AI 编码智能体可以加速这一转变，但将其能力转化为有用的科学软件需要识别研究需求并评估实现满足这些需求的程度。我们提出了一个用于智能体驱动的可微分科学软件进化的环境，该环境连接了需求识别、开发和质量评估。统一的微分接口在现有数值例程旁边暴露了可复用的导数规则，允许研究任务共享这些能力。研究需求指导开发，实现通过独立的导数检查、工作流测试和性能评估进行评估。经过验证的软件、研究程序和测试成为后续研究的共享资源。我们在涵盖物理、化学和生物建模的 20 个软件包中构建并验证了自动微分扩展，研究工作流展示了跨任务的复用。基准测试表明，在梯度评估和完整参数估计中，与有限差分相比，计算节省显著。研究驱动的修订使以前不支持的工作流变得可微分，纠正科学可观测量的导数，并消除冗余计算。量子控制和热设计研究根据物理评估修订目标，在复用现有导数的同时改进设计。这项工作为在既有科学软件中扩展可微分编程提供了一种实用方法，并围绕共享计算生态系统的递归改进组织 AI 智能体。
英文摘要
Differentiable programming connects scientific computation with gradient-based inference, learning and design. Extending these capabilities across a heterogeneous software ecosystem requires specialized effort to implement derivatives, integrate interfaces and evaluate quality. AI coding agents can accelerate this transformation, but translating their capabilities into useful scientific software requires identifying research needs and evaluating how well implementations meet them. We present an environment for agent-driven evolution of differentiable scientific software that connects demand identification, development and quality evaluation. A unified differentiation interface exposes reusable derivative rules alongside existing numerical routines, allowing research tasks to share these capabilities. Research requirements guide development, with implementations assessed through independent derivative checks, workflow tests and performance evaluation. Validated software, research programs and tests become shared resources for subsequent studies. We construct and validate automatic differentiation extensions across 20 packages spanning physical, chemical and biological modeling, with research workflows demonstrating reuse across tasks. Benchmarks demonstrate computational savings over finite differences in gradient evaluation and complete parameter estimation. Research-driven revisions make previously unsupported workflows differentiable, correct derivatives of scientific observables and eliminate redundant computation. Quantum-control and thermal-design studies revise objectives in response to physical evaluation, improving designs while reusing existing derivatives. This work provides a practical approach to expanding differentiable programming across established scientific software and organizing AI agents around the recursive improvement of a shared computational ecosystem.
