Reproduce “DFT-Agent Bench: Scaffold Effects and Execution Provenance in Autonomous DFT” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Shijie Liu, Jin Yang, Weilin Wan, Jingtao Han, Jia LI, Tianqi Deng
DFT-Agent Bench evaluates autonomous agents that run density-functional-theory simulations, comparing a near-bare workflow with a bundled specialist package and auditing required execution checks. In a fixed, retrieval-disabled, 100-task DeepSeek comparison, the frozen package scored 0.88 versus 0.71 for the baseline, a paired gain of 0.17. This package-level result applies to a predefined diagnostic slice, not general performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.