QAMA: Scalable Quantum Annealing Multi-Head Attention Operator for Deep Learning

Peng Du, Jinjing Shi, Wenxuan Wang +3

Attention mechanisms underpin modern deep learning, while the quadratic time and space complexity limit scalability for long sequences. To address this, Quantum Annealing Multi-Hea…