Rowan Reads @RowanReads · 2d, Nov 6, 2026, 8:31 PM openai.com/index/measurin… openai.com Measuring how AI agents handle everyday wet-lab science OpenAI introduces BioBasicsBench, a 100-task benchmark measuring how well AI agents interpret raw experimental data and handle everyday wet-lab science.
Rowan Reads @RowanReads · 2d, Nov 6, 2026, 8:07 PM rowansci.com/blog/a-convers… rowansci.com Building a Frontier Lab for Chemistry: A Conversation With Hatem Helal A conversation with Hatem Helal, the founder of Potential Sciences.
Rowan Reads @RowanReads · 3d, Nov 5, 2026, 7:43 PM primateria.com/blog/wavefunct… www.primateria.com Can AI solve a foundational problem in quantum mechanics? Introducing WavefunctionBench, a benchmark for AI agents that improve neural wavefunctions under a fixed compute budget.
Rowan Reads @RowanReads · 3d, Nov 5, 2026, 5:33 AM arxiv.org/abs/2611.06109 arxiv.org Subtle Energy Corrections Reshape Condensed-Phase Property... Accurate reproduction of electronic-structure labels by machine-learning interatomic potentials does not necessarily ensure reliable condensed-phase properties. We investigate the effects of...
Rowan Reads @RowanReads · 3d, Nov 5, 2026, 5:30 AM arxiv.org/abs/2611.06237 arxiv.org SO3LR Tauri: Generally Applicable Force Fields for Reactive... Machine-learned force fields (MLFFs) aim to create a unified model that describes all major classes of (bio)molecules at quantum-mechanical accuracy. We present SO3LR Tauri, a family of MLFFs...
