pipette
ENEnglish

Agentic AI for QCD perturbation theory or: how I learned to stop worrying and love the bot

Erik Lundstrum

PreprintAfirmaciones fuertes, leer con cuidado

En palabras de los autores

AI coding agents are extremely effective at performing calculations that can be broken down into modular steps, each with verifiable checks of correctness. Perturbative QCD calculations satisfy both of these properties and we find that they are well-suited to solution by agentic AI systems. We discuss our methods for instructing and verifying the work of agents in one- and two-loop perturbative calculations, with a focus on pre-empting common mistakes. We find that once an agent has successfully completed one perturbative calculation, its instructions transfer very well to other similar calculations which can then be completed and verified in a matter of hours with little effort from the user.

Resultado principalEl resumen no menciona limitaciones.

Apareció: viernes, 30 de octubre. arXiv. Preprint, todavía sin revisión por pares.

Comentario de los autores: 9 pages, 3 figures, presented at the 43rd International Symposium on Lattice Field Theory (Lattice 2026) July 26 to August 1, 2026 University of Maryland, College Park, USA