research
An Agent-Written Functional Survives the Physics Check
A compact density-functional candidate retained its advantage after self-consistent and held-out evaluation.
Summary
A compact density-functional candidate retained its advantage after self-consistent and held-out evaluation.
Researchers connected language-model agents to a GPU evaluator for symbolic search over exchange-correlation functionals, with persistent memory and separate self-consistent and external checks. The surviving 13-parameter candidate adds a smooth density window that reduces same-spin correlation in selected regions. After refitting, it reduced heavy-atom total-energy error by 29% against a no-window control on one test collection and lowered a held-out molecular-chemistry error by 11.7% relative to the stated baseline. The evidence supports one compact candidate under the reported validation pipeline, not general autonomous discovery.
Why it matters
A compact density-functional candidate retained its advantage after self-consistent and held-out evaluation.
Limits and context
- The evidence supports one compact candidate under the reported validation pipeline, not general autonomous discovery.
Key claims
A compact density-functional candidate retained its advantage after self-consistent and held-out evaluation.
Qualification: The evidence supports one compact candidate under the reported validation pipeline, not general autonomous discovery.
Evidence: source-2026-11-12-007
Sources
- arXiv:2611.13329v1arXiv / Cornell University · primary research
Corrections
No corrections have been recorded for this story.