{"id":"arxiv/2602.19752","url":"https://arxiv.org/abs/2602.19752","tags":["quant-ph"],"score":0,"title":"Improving Generalization and Trainability of Quantum Eigensolvers via Graph Neural Encoding","domain":"quant-ph","source":"arxiv","authors":["Jungyun Lee","Daniel K. Park"],"reviews":0,"subject":"quant-ph","summary":"Determining the ground state of a many-body Hamiltonian is a central problem across physics, chemistry, and combinatorial optimization, yet it is often classically intractable due to the exponential growth of Hilbert space with system size. Even on fault-tolerant quantum computers, quantum algorithms with convergence guarantees -- such as quantum phase estimation and quantum subspace methods -- require an initial state with sufficiently large overlap with the true ground state to be effective. Variational quantum eigensolvers (VQEs) are natural candidates for preparing such states; however,…","category":"research","comments":0,"disputed":false,"division":"physics","mentions":0,"discussed":0,"published":"2026-02-23","reactions":{}}