{"ID":2898016,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.03853","arxiv_id":"2507.03853","title":"OrbitAll: A Unified Quantum Mechanical Representation Deep Learning Framework for All Molecular Systems","abstract":"Despite the success of deep learning methods in quantum chemistry, their representational capacity is most often confined to neutral, closed-shell molecules. However, real-world chemical systems often exhibit complex characteristics, including varying charges, spins, and environments. We introduce OrbitAll, a geometry- and physics-informed deep learning framework that can represent all molecular systems with electronic structure information. OrbitAll utilizes spin-polarized orbital features from the underlying quantum mechanical method, and combines it with graph neural networks satisfying SE(3)-equivariance. The resulting framework can represent and process any molecular system with arbitrary charges, spins, and environmental effects. OrbitAll demonstrates superior performance and generalization on predicting charged, open-shell, and solvated molecules, while also robustly extrapolating to molecules significantly larger than the training data by leveraging a physics-informed architecture. OrbitAll achieves chemical accuracy using 10 times fewer training data than competing AI models, with a speedup of approximately $10^3$ - $10^4$ compared to density functional theory.","short_abstract":"Despite the success of deep learning methods in quantum chemistry, their representational capacity is most often confined to neutral, closed-shell molecules. However, real-world chemical systems often exhibit complex characteristics, including varying charges, spins, and environments. We introduce OrbitAll, a geometry-...","url_abs":"https://arxiv.org/abs/2507.03853","url_pdf":"https://arxiv.org/pdf/2507.03853v1","authors":"[\"Beom Seok Kang\",\"Vignesh C. Bhethanabotla\",\"Amin Tavakoli\",\"Maurice D. Hanisch\",\"William A. Goddard\",\"Anima Anandkumar\"]","published":"2025-07-05T01:21:56Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"physics.chem-ph\"]","methods":"[\"Graph Neural Network\"]","has_code":false}