AI and the World's Hard Problems · Quantum physics & quantum computing · arXiv · 2026-10-26
Quantum-aware Transformer model for state classification
Classifying entangled states, especially mixed states in higher dimensions, is a hard problem. The authors pretrained a transformer without labels by masking elements of vectorized Hermitian matrix representations of bipartite quantum states. After training, the model reached near-perfect accuracy in distinguishing separable from entangled states.