Authors Pengcheng Hou, Xiao-jun Tang, Youjin Deng, Xiaojun Tan, Sihan Hu, Ruisi Wang, Shuo Chen, Lei Wang, Kun Chen
Publication date 3 October 2026
Tldr This paper presents an environment for agent-driven evolution of differentiable scientific software that connects demand identification, development and quality evaluation. It constructs and validates automatic differentiation extensions across 20 packages spanning physical, chemical and biological modeling, with research workflows demonstrating reuse across tasks.
This page was built for publication: Recursive Improvement of a Differentiable Scientific Software Ecosystem
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7367469)
