lara-hpc: Validation-Driven Agentic Supercomputer Workflows for Atomistic Modeling
PRX Intelligence · 10.11.2026 · 1 (2)
DOI: 10.1103/cp6v-t93q ↗Kurz gefasst
Fachartikel in der Zeitschrift PRX Intelligence, online erschienen am 10.11.2026. Verfasst von William Dawson und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Foundation for Computational Science, Formation of a Research and Education Center, Field of Computational Science.
Abstract (Originalsprache)
Large language models and agentic systems have recently demonstrated potential for automating scientific workflows, including atomistic simulations. However, their deployment in high-performance computing (HPC) environments remains limited by the lack of mechanisms ensuring correctness, reproducibility, and safe interaction with computational resources. Generated workflows suffer from inconsistencies, incorrect application programming interface usage, or invalid physical configurations—leading to failed or unreliable simulations. In this work, we introduce (Lara Agentic Research Assistant for HPC), a validation-driven agentic framework to enable reliable workflow generation for atomistic modeling on HPC systems. Our approach is based on three key components: (1) a controlled execution layer that mediates all interactions with HPC resources; (2) simulation-native validation through dry-run capabilities, enabling execution-level verification without incurring resource cost; and (3) a multiphase agentic pipeline combining retrieval-augmented generation and iterative refinement. We demonstrate the effectiveness of this approach by performing an end-to-end atomistic simulation workflow on HPC by applying to density functional theory simulations. The results show that validation-driven generation significantly improves robustness and enables iterative correction of both syntactic and physical inconsistencies. More broadly, this work advocates for a shift from generation-first to validation-first paradigms in artificial-intelligence-assisted scientific computing. We argue that the future task of the computational physics community is to develop domain-specific agentic systems based on structured tooling to realize an HPC enabled copiloted research ecosystem.
Förderung
- Foundation for Computational Science
- Formation of a Research and Education Center
- Field of Computational Science
Zitieren
APA
Dawson, W., Beal, L., Curé, Y., Fisicaro, G., Rolland, D., & Genovese, L. (2026). lara-hpc: Validation-Driven Agentic Supercomputer Workflows for Atomistic Modeling. PRX Intelligence, 1(2). https://doi.org/10.1103/cp6v-t93q
Vancouver
Dawson W, Beal L, Curé Y, Fisicaro G, Rolland D, Genovese L. lara-hpc: Validation-Driven Agentic Supercomputer Workflows for Atomistic Modeling. PRX Intelligence. 2026;1(2). doi:10.1103/cp6v-t93q
BibTeX
@article{Dawson2026,
author = {Dawson, William and Beal, Louis and Curé, Yoann and Fisicaro, Giuseppe and Rolland, Dorian and Genovese, Luigi},
title = {lara-hpc: Validation-Driven Agentic Supercomputer Workflows for Atomistic Modeling},
journal = {PRX Intelligence},
year = {2026},
volume = {1},
number = {2},
doi = {10.1103/cp6v-t93q}
}
EUVIMED