pith. sign in

arxiv: 2606.05968 · v1 · pith:U3GDZTCSnew · submitted 2026-06-04 · 🪐 quant-ph

Shattering the Symmetry Trap in Fixed-Ansatz VQE: An Accelerated ADAPT-VQE Study of Three Pillar Molecules under Bravyi-Kitaev Mapping

Hermawan Kresno Dipojono This is my paper

Pith reviewed 2026-06-28 00:49 UTC · model grok-4.3

classification 🪐 quant-ph
keywords ADAPT-VQEBravyi-Kitaev mappingvariational quantum eigensolverfull configuration interactionmolecular simulationsymmetry breakingquantum chemistryoperator pool selection
0
0 comments X

The pith

ADAPT-VQE isolates symmetry-breaking operators to reach exact FCI energies in the first macro-cycle for LiH, HF and H2O under Bravyi-Kitaev mapping.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Fixed-ansatz VQE methods such as UCCSD encounter zero-gradient traps and initialization paralysis with the Bravyi-Kitaev mapping on molecules in stretched or asymmetric geometries. The paper shows that the adaptive ADAPT-VQE loop overcomes these traps by using analytical commutator gradients to select the dominant symmetry-breaking operators from the pool. A vector-based Taylor series expansion replaces dense matrix exponentiation to keep computations feasible. This combination produces exact full configuration interaction energies within the first macro-cycle for lithium hydride, hydrogen fluoride and water, with numerical stability through 12 qubits. A sympathetic reader would care because the result offers a concrete route to stable variational simulation of strongly correlated molecules on near-term hardware.

Core claim

The accelerated ADAPT-VQE framework achieves instant, exact Full Configuration Interaction (FCI) convergence within the very first macro-cycle across LiH, HF, and H2O under Bravyi-Kitaev mapping by isolating the dominant symmetry-breaking operators using analytical commutator gradients, while the vector-based Taylor series state-evolution engine maintains absolute numerical stability up to a 12-qubit register space.

What carries the argument

The ADAPT-VQE operator-selection loop driven by analytical commutator gradients together with the vector-based Taylor series state-evolution engine.

If this is right

  • Conventional UCCSD-VQE reaches a zero energy shift due to global phase cancellations in the BK tree structures.
  • The dynamic ADAPT-VQE loop isolates dominant symmetry-breaking operators via analytical commutator gradients.
  • Exact FCI convergence occurs within the first macro-cycle for all three molecular systems.
  • Numerical stability holds up to a 12-qubit register space.
  • The vector-based engine bypasses the O(N^3) cost of dense matrix exponentiation and SVD.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same operator-selection strategy may extend to other fermion-to-qubit mappings that induce symmetry traps.
  • Hardware resource estimates for simulating polarized triatomic systems could be lowered by avoiding extra optimization cycles.
  • Testing the method on larger registers or additional molecules would clarify whether the first-cycle convergence pattern persists.
  • Analytical commutator gradients may reduce the total number of circuit evaluations required in related variational algorithms.

Load-bearing premise

The vector-based Taylor series expansion state-evolution engine produces numerically exact state updates equivalent to full matrix exponentiation for the 12-qubit registers and operator pools used.

What would settle it

A direct comparison in which the ADAPT-VQE energy after the first macro-cycle deviates from the known FCI value for any of the three molecules would falsify the claim of instant exact convergence.

read the original abstract

Fixed-ansatz Variational Quantum Eigensolvers (VQE), such as the Unitary Coupled Cluster with Singles and Doubles (UCCSD) framework, frequently suffer from severe initialization paralyzation and zero-gradient traps when evaluated using the non-local Bravyi-Kitaev (BK) fermion-to-qubit mapping. In this work, we systematically demonstrate how the Adaptive Derivative-Assembled Pseudo-Trotter (ADAPT-VQE) framework shatters these structural limitations across three distinct electronic and geometric molecular pillars: Lithium Hydride ($\text{LiH}$), Hydrogen Fluoride ($\text{HF}$), and Water ($\text{H}_2\text{O}$), under heavily stretched or asymmetric multi-reference configurations. While conventional UCCSD-VQE flatlines completely at a zero energy shift ($0.000000$~Ha) due to global phase cancellations inherent to the BK tree structures, our dynamic ADAPT-VQE loop successfully isolates the dominant symmetry-breaking operators using analytical commutator gradients. To bypass the severe $\mathcal{O}(N^3)$ computational bottlenecks of dense matrix exponentiation and Singular Value Decomposition on larger registers, we implement a highly optimized, vector-based Taylor series expansion state-evolution engine. Our numerical results show that the accelerated ADAPT-VQE framework achieves instant, exact Full Configuration Interaction (FCI) convergence within the very first macro-cycle across all three molecular systems, maintaining absolute numerical stability up to a 12-qubit register space. This study establishes a robust, hardware-efficient path for simulating strongly correlated, highly polarized triatomic chemical environments on near-term local architectures.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The manuscript claims that an accelerated ADAPT-VQE algorithm, employing analytical commutator gradients for operator selection and a vector-based Taylor series expansion for state evolution, overcomes zero-gradient traps in fixed-ansatz VQE (such as UCCSD) under Bravyi-Kitaev mapping. It reports achieving instant, exact Full Configuration Interaction (FCI) convergence within the first macro-cycle for LiH, HF, and H2O in stretched or asymmetric configurations, with absolute numerical stability up to 12 qubits, while bypassing O(N^3) costs of dense matrix exponentiation.

Significance. If the central numerical claims hold with rigorous verification, the work would demonstrate a practical route to adaptive VQE for multi-reference molecular systems on near-term hardware by dynamically isolating symmetry-breaking operators and replacing expensive linear-algebra steps with a cheaper state-update engine.

major comments (2)
  1. [Abstract] Abstract: the headline claim of 'instant, exact FCI convergence within the very first macro-cycle' and 'absolute numerical stability' up to 12 qubits is asserted without any tabulated energies, error metrics, operator-pool sizes, convergence plots, or direct comparison to full-matrix FCI or exact exponentiation.
  2. [Abstract] Abstract (methods description): the vector-based Taylor series state-evolution engine is stated to deliver 'numerically exact' updates equivalent to full matrix exponentiation, yet no truncation order, remainder bound, accumulation-error analysis, or side-by-side numerical verification against dense expm for the specific 12-qubit registers and operator pools is supplied.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading of the manuscript and for highlighting these points regarding the abstract. We address each major comment below and indicate the revisions made to strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the headline claim of 'instant, exact FCI convergence within the very first macro-cycle' and 'absolute numerical stability' up to 12 qubits is asserted without any tabulated energies, error metrics, operator-pool sizes, convergence plots, or direct comparison to full-matrix FCI or exact exponentiation.

    Authors: The abstract serves as a concise summary of the principal findings. All supporting numerical evidence—including tabulated energies, error metrics, operator-pool sizes, convergence plots, and direct comparisons to full-matrix FCI and exact exponentiation—is presented in the main text (Sections 3–4) together with the supplementary material. To address the concern, we have revised the abstract to include explicit cross-references to these sections and to the three molecular systems examined. revision: partial

  2. Referee: [Abstract] Abstract (methods description): the vector-based Taylor series state-evolution engine is stated to deliver 'numerically exact' updates equivalent to full matrix exponentiation, yet no truncation order, remainder bound, accumulation-error analysis, or side-by-side numerical verification against dense expm for the specific 12-qubit registers and operator pools is supplied.

    Authors: We agree that the abstract’s methods description would benefit from additional technical detail. In the revised manuscript we have expanded the Methods section with a new paragraph that specifies the Taylor-series truncation order employed, the Lagrange-form remainder bound, an accumulation-error analysis, and direct numerical comparisons against dense matrix exponentiation (scipy.linalg.expm) for the 12-qubit registers and operator pools used, confirming agreement to within 10^{-14} Ha. revision: yes

Circularity Check

0 steps flagged

No circularity detected in derivation or claims

full rationale

The paper presents numerical outcomes from an adaptive ADAPT-VQE loop that isolates operators via commutator gradients and reports exact FCI convergence as a result of that loop. The Taylor-series state-evolution engine is introduced as an implementation detail to avoid O(N^3) costs, with the exactness claim tied to the reported numerical stability rather than any self-definitional equation, fitted parameter renamed as prediction, or load-bearing self-citation. No equations or steps reduce by construction to their own inputs; the central results remain independent numerical demonstrations on the three molecular systems. This is the expected self-contained case with no flagged patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract invokes standard VQE/ADAPT-VQE machinery and the known properties of BK mapping without introducing new fitted parameters, axioms, or entities.

pith-pipeline@v0.9.1-grok · 5833 in / 1037 out tokens · 37265 ms · 2026-06-28T00:49:06.775654+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Alleviating the Sparse Matrix Scaling Bottleneck in Adaptive VQE via High-Order Taylor State Evolution

    quant-ph 2026-06 unverdicted novelty 4.0

    Fifth-order Taylor truncation enables O(Nz) state updates in adaptive VQE by chaining sparse matrix-vector products, preserving >0.999999 fidelity and subchemical accuracy on 12-14 qubit systems.

  2. Representation-Induced Symmetry Trapping in Adaptive Variational Quantum Simulations of Multi-Reference Topologies

    quant-ph 2026-06 unverdicted novelty 4.0

    Bravyi-Kitaev mapping induces representation-locked optimization trapping in asymmetric multi-reference adaptive VQE, protected by symmetry, with a proposed covariance-driven adaptive shot-allocation filter.

Reference graph

Works this paper leans on

24 extracted references · 1 canonical work pages · cited by 2 Pith papers

  1. [1]

    The core shell is frozen, isolating 2 active valence electrons across 5 active spatial orbitals, mapped to a 10-qubit register

    Lithium Hydride ( LiH):Evaluated at an equilibrium-stretched coordinate of R = 1 .696 ˚A. The core shell is frozen, isolating 2 active valence electrons across 5 active spatial orbitals, mapped to a 10-qubit register

  2. [2]

    The Fluorine 1s2 shell is frozen, leaving 8 active valence electrons across 5 active spatial orbitals, mapped to a 10-qubit register

    Hydrogen Fluoride ( HF):Evaluated at a heavily stretched bond distance of R = 3.0 ˚A. The Fluorine 1s2 shell is frozen, leaving 8 active valence electrons across 5 active spatial orbitals, mapped to a 10-qubit register

  3. [3]

    The Oxygen 1 s2 shell is frozen, leaving 8 active valence electrons distributed across 6 active spatial orbitals, requiring a full 12-qubit register

    Water (H2O):Configured under an asymmetric double stretch with bond vectors R1(O-H) = 1.95 ˚A and R2(O-H) = 1.50 ˚A to create an intricate multi- reference landscape. The Oxygen 1 s2 shell is frozen, leaving 8 active valence electrons distributed across 6 active spatial orbitals, requiring a full 12-qubit register. The operator pools are generated manuall...

  4. [4]

    Lithium Hydride (LiH) 10-Qubit Space Execution Log [INFO] Initialize ADAPT-VQE Engine | Register: 10 Qubits | Pool: 24 SUSD Operators [INFO] Geometry: R(Li-H) = 1.696 Angstrom | Reference State: |1111000000> [INFO] Computing Baseline Mapped UCCSD-VQE... [OPTIMIZER] SLSQP Iteration 001 | Energy Shift: -0.00000000 Ha | Gradient Norm: 0.00000000 [OPTIMIZER] ...

  5. [5]

    [OPTIMIZER] SLSQP Iteration 001 | Energy Shift: 0.00000000 Ha | Gradient Norm: 0.00000000 [WARN] Plain BK-UCCSD Optimization Terminated: Zero-Gradient Initialization Trap Detected

    Hydrogen Fluoride (HF) 10-Qubit Space Execution Log [INFO] Initialize ADAPT-VQE Engine | Register: 10 Qubits | Pool: 24 SUSD Operators [INFO] Geometry: R(H-F) = 3.000 Angstrom | Reference State: |1111111100> [INFO] Computing Baseline Mapped UCCSD-VQE... [OPTIMIZER] SLSQP Iteration 001 | Energy Shift: 0.00000000 Ha | Gradient Norm: 0.00000000 [WARN] Plain ...

  6. [6]

    [OPTIMIZER] SLSQP Iteration 001 | Energy Shift: 0.00000000 Ha | Gradient Norm: 0.00000000 [WARN] Plain BK-UCCSD Fixed-Ansatz Frozen at Hartree-Fock Level

    Water (H 2O) 12-Qubit Space Execution Log [INFO] Initialize ADAPT-VQE Engine | Register: 12 Qubits | Pool: 92 SUSD Operators [INFO] Geometry: Asymmetric Double Stretch | R1=1.95 A, R2=1.50 A | Angle=104.5 deg [INFO] Reference State Layout: |111111110000> [INFO] Computing Baseline Mapped UCCSD-VQE... [OPTIMIZER] SLSQP Iteration 001 | Energy Shift: 0.000000...

  7. [7]

    Peruzzoet al., A variational eigenvalue solver on a photonic quantum processor, Nat

    A. Peruzzoet al., A variational eigenvalue solver on a photonic quantum processor, Nat. Commun.5, 4213 (2014)

    2014

  8. [8]

    J. R. McClean, J. Romero, R. Babbush, and A. Aspuru-Guzik, The theory of variational hybrid quantum-classical algorithms, New J. Phys.18, 023023 (2016)

    2016

  9. [9]

    Tillyet al., The Variational Quantum Eigensolver: A review of methods and best practices, Phys

    J. Tillyet al., The Variational Quantum Eigensolver: A review of methods and best practices, Phys. Rep.985, 1-128 (2022)

    2022

  10. [10]

    Jordan and E

    P. Jordan and E. Wigner, ¨Uber das Paulische ¨Aquivalenzverbot, Z. Phys.47, 631 (1928)

    1928

  11. [11]

    Fradkin, Jordan-Wigner transformation for quantum-spin systems in two dimensions and fractional statistics, Phys

    E. Fradkin, Jordan-Wigner transformation for quantum-spin systems in two dimensions and fractional statistics, Phys. Rev. Lett.63, 322 (1989)

    1989

  12. [12]

    Bravyi and A

    S. Bravyi and A. Kitaev, Fermionic Quantum Computation, Ann. Phys.298, 210 (2002)

    2002

  13. [13]

    Kumar N., I

    E. Kumar N., I. S., V. Jani, Sethukkarasi, C., and D. Ethirajan, Molecular Ground-State Energy Estima- tion Using VQE, inProceedings of the 2025 IEEE Pune Section International Conference (PuneCon)DOI: 10.1109/PuneCon67554.2025.11378391

    work page doi:10.1109/punecon67554.2025.11378391 2025

  14. [14]

    H. R. Grimsley, S. E. Economou, E. Barnes, and N. J. Mayhall, An adaptive variational algorithm for exact molecular simulations on a quantum computer, Nat. Commun.10, 3007 (2019)

    2019

  15. [15]

    Feniou, M

    C. Feniou, M. Hassan, D. Traore, E. Giner, Y. Maday, J. P. Piquemal, and J.-P. Philip, Overlap-ADAPT-VQE: Practical Quantum Chemistry on Quantum Computers via Overlap-Guided Compact Ansatz, Commun. Phys.6, 192 (2023)

    2023

  16. [16]

    H. R. Grimsley and F. A. Evangelista, Challenging excited states from adaptive quantum eigensolver: Subspace expansions vs. state-averaged strategies, Quantum Sci. Technol.10, 025003 (2025)

    2025

  17. [17]

    Zhang and D

    J. Zhang and D. Lacroix, Excited states from ADAPT-VQE convergence path in many-body problems: Application to nuclear pairing problem and H4 molecule dissociation, Phys. Lett. B869, 139841 (2025)

    2025

  18. [18]

    J. S. Van Dyke, G. S. Barron, E. Barnes, S. E. Economou, and N. J. Mayhall, Scaling adaptive quantum simulation algorithms via operator pool tiling, Phys. Rev. Research6, L012030 (2024)

    2024

  19. [19]

    H. L. Tang, Y. Chen, P. Biswas, A. B. Magann, C. Arenz, and S. E. Economou, Nonvariational ADAPT algorithm for quantum simulations, Phys. Rev. Research7, 023275 (2025)

    2025

  20. [20]

    J. T. Seeley, M. J. Richard, and P. J. Love, The Bravyi-Kitaev transformation for quantum computation of molecular electronic wave functions, J. Chem. Phys.137, 224109 (2012)

    2012

  21. [21]

    J. R. McClean, S. Boixo, V. N. Smelyanskiy, R. Babbush, and H. Neven, Barren plateaus in quantum neural network training landscapes, Nat. Commun.9, 4812 (2018)

    2018

  22. [22]

    Kraft, A Software Package for Sequential Quadratic Programming, Technical Report DFVLR-FB 88-28, Institut f¨ ur Dynamik der Flugsysteme, Oberpfaffenhofen (1988)

    D. Kraft, A Software Package for Sequential Quadratic Programming, Technical Report DFVLR-FB 88-28, Institut f¨ ur Dynamik der Flugsysteme, Oberpfaffenhofen (1988)

    1988

  23. [23]

    C. N. Selfet al., Measurement optimization in adaptive variational quantum eigensolvers, Phys. Rev. Research3, 043164 (2021)

    2021

  24. [24]

    J. R. McCleanet al., Theory of noisy variational quantum eigensolvers: A statistical approach, New J. Phys.18, 023023 (2016)

    2016