Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 52 min ago
PreprintarXiv licencearXiv

Topology Obstructs Pure Foundation Neural Quantum States

Abstract

Foundation models for ground states in spin-1/2 systems are a promising method for problems ranging from quantum chemistry to identifying new phase diagrams. Nearly all such models are currently pure-states that condition on the Hamiltonian's parameters, whose Monte Carlo samples give energy estimates according to the variational principle. In this contribution, we show that this representation is topologically obstructed. For any gapped Hamiltonian family whose ground-state bundle is non-trivial, every continuous normalized state-vector model has zero fidelity with the ground state at some parameter value in the Hamiltonian family. For that value, the energy is at least one spectral gap, Δ, with an O(Δ) gap in an open-neighbourhood of that point. We show that this is a sufficient no-go also in the case of degenerate ground-state manifolds, time dynamics, and periodic systems with mixed space-time topology, demonstrating these obstructions on one- and two-qubit systems. We discuss how this causes a spike in the fidelity susceptibility, giving a numerical signature of a phase-transition where there is none. We then show that operator-valued models canonically avoid these obstructions and preserve topological information, implying a structural necessity in representation for foundation neural quantum states.

Related

  1. Minimal representations of topology-preserving quantum-like states
    William Horvat, Debadrita Saha, Nicolás Gigena +2
    arXivPreprintarXiv licenceFull textCombinatorics & discrete mathematicsQuantum physics & quantum informationarXiv
  2. The ℤ2-Index of a Pair of Pure States and the Topology of Interacting 1D Superconductors
    Anna Mazhar, Jacob Shapiro
    arXivPreprintCC BY 4.0Full textMathematical physicsQuantum physics & quantum informationarXiv
  3. KATOsuper: Surrogate-accelerated neural topology optimization with sensitivity-consistent Fourier neural operators
    Shengyu Yan, Jasmin Jelovica
    arXivPreprintCC BY-NC-ND 4.0Full textNumerical analysis & scientific computingarXiv
  4. Neural topology optimization of ship structures under propulsion machinery vibrations
    Shengyu Yan, Muhammad Muztahidul Hakim Zareer, Jasmin Jelovica
    arXivPreprintCC BY-NC-ND 4.0Full textNumerical analysis & scientific computingComputational physicsarXiv
  5. Odd branching obstructs multiplicity-one integral current structures
    Deguang Zhong
    arXivPreprintCC BY 4.0Full textGeometry & topologyarXiv
BibTeX
@misc{heightman2026topology,
  title = {{Topology Obstructs Pure Foundation Neural Quantum States}},
  author = {Timothy Heightman and Elena Orlova and Philip Mantrov and Aleksei Ustimenko},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.07591},
  eprint = {2609.07591},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.07591},
}
RIS
TY  - UNPB
TI  - Topology Obstructs Pure Foundation Neural Quantum States
AU  - Timothy Heightman
AU  - Elena Orlova
AU  - Philip Mantrov
AU  - Aleksei Ustimenko
PY  - 2026
DA  - 2026/09/07
PB  - arXiv
DO  - 10.48550/arxiv.2609.07591
UR  - https://arxiv.org/abs/2609.07591
AB  - Foundation models for ground states in spin-1/2 systems are a promising method for problems ranging from quantum chemistry to identifying new phase diagrams. Nearly all such models are currently pure-states that condition on the Hamiltonian's parameters, whose Monte Carlo samples give energy estimates according to the variational principle. In this contribution, we show that this representation is topologically obstructed. For any gapped Hamiltonian family whose ground-state bundle is non-trivial, every continuous normalized state-vector model has zero fidelity with the ground state at some parameter value in the Hamiltonian family. For that value, the energy is at least one spectral gap, Δ, with an O(Δ) gap in an open-neighbourhood of that point. We show that this is a sufficient no-go also in the case of degenerate ground-state manifolds, time dynamics, and periodic systems with mixed space-time topology, demonstrating these obstructions on one- and two-qubit systems. We discuss how this causes a spike in the fidelity susceptibility, giving a numerical signature of a phase-transition where there is none. We then show that operator-valued models canonically avoid these obstructions and preserve topological information, implying a structural necessity in representation for foundation neural quantum states.
ER  -