PreprintarXiv licencearXiv
Agentic AI for QCD perturbation theory or: how I learned to stop worrying and love the bot
Abstract
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.
Related
- How I Learned to Stop Worrying and Love the Redfield Equation: Completely-Positive Resummation of Non-Markovian Dynamics
- Agentic AI Reasoning and Learned Sampling for Task-Aware Projection Selection in Sparse-View CT: Predicting Hemodynamics
- A Note on a High-Energy Limit of Gravitational Perturbation Theory
- Threshold Corrections in Heterotic String Theory from Conformal Perturbation Theory
- Minimal superfluid vortices in chiral perturbation theory
BibTeX
@misc{lundstrum2026agentic,
title = {{Agentic AI for QCD perturbation theory or: how I learned to stop worrying and love the bot}},
author = {Erik Lundstrum},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.37960},
eprint = {2610.37960},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.37960},
}
RIS
TY - UNPB
TI - Agentic AI for QCD perturbation theory or: how I learned to stop worrying and love the bot
AU - Erik Lundstrum
PY - 2026
DA - 2026/10/27
PB - arXiv
DO - 10.48550/arxiv.2610.37960
UR - https://arxiv.org/abs/2610.37960
AB - 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.
ER -