Search papers, labs, and topics across Lattice.
ElectrolyteMD-Bench evaluates autonomous AI agents conducting end-to-end molecular dynamics (MD) research across dilute, high-concentration, and localized high-concentration electrolyte solvation regimes. Assessing agent capabilities across the full loop of simulation construction, property analysis, and interpretation is essential to determine whether LLMs can perform genuine computational materials discovery rather than just automated scripting. Across eight model-harness setups, all 24 runs successfully generated valid MD trajectories, yet zero complete studies satisfied all scientific validity requirements due to flawed physical definitions, implementation errors, and uncalibrated completion claims.
AI agents achieve a 100% success rate at executing molecular dynamics trajectories, yet fail completely when audited for scientifically sound property analysis and interpretation.
AI agents are advancing the automation of molecular dynamics (MD) research, with the potential to accelerate discovery and design of lithium-battery electrolytes. However, whether they can autonomously complete scientifically reliable electrolyte MD studies remains insufficiently evaluated. We introduce ElectrolyteMD-Bench, which unifies dilute, high-concentration, and localized high-concentration electrolytes in a comparative study across distinct solvation regimes. The benchmark connects local coordination, ionic association, single-particle dynamics, and collective transport within one continuous research process. Agents autonomously perform system construction, equilibration, production sampling, property analysis, and interpretation under a frozen scientific model, while deciding whether to correct errors, continue sampling, or stop. We audit each study along four dimensions: simulation validity, analysis validity, evidence integrity, and outcome calibration. Across eight model--harness configurations, all 24 system runs produced valid production trajectories, yet none of the eight studies satisfied all scientific requirements. Failures included missing analyses, incorrect physical definitions or numerical implementations, insufficient statistical support, and completion claims inconsistent with retained evidence. These results expose a gap between successful MD workflow execution and reliable scientific study completion. ElectrolyteMD-Bench therefore provides a foundational test and defines a key capability threshold that autonomous electrolyte simulation must cross to advance toward computation-first materials exploration.
