Agentic Discovery of Exchange-Correlation Density Functionals

TL;DR

This paper introduces an AI-driven system using large language models to automate the discovery of exchange-correlation functionals in density functional theory, achieving notable improvements over existing baselines.

Contribution

It presents an agentic search framework that systematically proposes and evaluates functional-form changes guided by evolutionary history and performance metrics.

Findings

Discovered a functional that outperforms the {}B97M-V baseline by ~9%.

Demonstrated the potential of LLMs to automate scientific discovery in DFT.

Highlighted the importance of domain constraints to prevent unphysical shortcuts.

Abstract

The development of accurate exchange-correlation (XC) functionals remains a longstanding challenge in density functional theory (DFT). The vast majority of XC functionals have been hand designed by human researchers combining physical insight, exact constraints, and empirical fitting. Recent advances in large language models enable a systematic, automated alternative to this human-driven design loop. This report presents an agentic search system in which an LLM proposes structured functional-form changes guided by evolutionary history. The system attempts to improve functional performance through an iterative plan-execute-summarize loop, where improvements are measurable by optimizing functional parameters against a standard thermochemistry dataset, then evaluating performance on a held-out subset. The strongest discovered functional, SAFS26-a (Seed Agentic Functional Search 2026), improves upon the gold-standard {\omega}B97M-V baseline by ~9%. These results also surface a cautionary lesson for AI-assisted science: models powerful enough to discover genuine improvements are equally capable of exploiting unphysical shortcuts to game the benchmark; domain expertise translated into explicitly enforced constraints remains essential to keeping results scientifically grounded.