aiqbench analysis
This article describes FUSION, a specialized AI agent designed to safely execute and verify nuclear-physics codes by integrating code-specific skills and rigorous benchmark checks. The system addresses the risk of general-purpose coding agents introducing physical errors by requiring reproduction of known benchmarks before reporting results. This represents a significant advancement in applying machine learning to high-stakes scientific computing where accuracy is non-negotiable.
FUSION: a skill-based research agent for publicly obtainable nuclear-physics codes
arXiv:2609.04742v1 Announce Type: cross Abstract: Running an unfamiliar nuclear-physics code is rarely difficult because of the physics alone. One must find and build the program, learn its input conventions, and decide whether a plausible output is actually correct. A general-purpose coding agent helps with the first two tasks but may make the last one harder: it can write an input file that runs with the wrong physical convention. FUSION addresses this problem with code-specific skills. A skill obtains the code from its public source, starts from a verified input, runs and parses the calculation, records known failure modes, and must reproduce a stated benchmark to a stated tolerance before reporting a result. The current release covers twenty codes, spanning optical models and reactions, nuclear structure, fission and statistical models, astrophysics and R-matrix analysis, and heavy-ion transport. It also includes an offline, searchable collection of 61 167 pages derived from the nucl-th literature. User notes and credentials remain outside the public repository. FUSION is available under the MIT license at https://github.com/jinleiphys/FUSION; documentation is at https://vibeinscience.com. Here I describe the design, the checks behind the current release, and one complete calculation from input to comparison with measured data.
Abstract:Running an unfamiliar nuclear-physics code is rarely difficult because of the physics alone. One must find and build the program, learn its input conventions, and decide whether a plausible output is actually correct. A general-purpose coding agent helps with the first two tasks but may make the last one harder: it can write an input file that runs with the wrong physical convention. FUSION addresses this problem with code-specific skills. A skill obtains the code from its public source, starts from a verified input, runs and parses the calculation, records known failure modes, and must reproduce a stated benchmark to a stated tolerance before reporting a result. The current release covers twenty codes, spanning optical models and reactions, nuclear structure, fission and statistical models, astrophysics and R-matrix analysis, and heavy-ion transport. It also includes an offline, searchable collection of 61 167 pages derived from the nucl-th literature. User notes and credentials remain outside the public repository. FUSION is available under the MIT license at this https URL documentation is at this https URL. Here I describe the design, the checks behind the current release, and one complete calculation from input to comparison with measured data.
11 pages, 10 figures. Platform at this https URL (MIT), documentation at this https URL
Nuclear Theory (nucl-th); Machine Learning (cs.LG); Computational Physics (physics.comp-ph)
arXiv:2609.04742 [nucl-th]
(or arXiv:2609.04742v1 [nucl-th] for this version)
https://doi.org/10.48550/arXiv.2609.04742 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 308 words
