Google Research has developed Empirical Research Assistance (ERA), an AI tool designed to help scientists write and optimize code for computational experiments. Published in Nature, ERA utilizes Gemini and a tree search approach to explore thousands of coding options, achieving expert-level performance across diverse scientific benchmarks. This technology is being integrated into Google Labs' Computational Discovery prototype and is rolling out more broadly via Gemini for Science, aiming to accelerate scientific discovery and democratize access to advanced computational modeling. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Accelerates scientific discovery by automating complex coding tasks, potentially democratizing advanced computational modeling.
RANK_REASON The cluster describes a new AI tool for scientific coding published in a journal and integrated into a research prototype and product. [lever_c_demoted from research: ic=1 ai=1.0]
