{"id":"97d00dda-fe68-4df3-b1bb-75bb3229a1ed","arxiv_id":"2506.22387","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":3,"one_line_summary":"For Rydberg-atom VQE with global controls, a symmetry analysis of invariant subspaces predicts when the ground state is unreachable; the necessary condition matches simulations for Ising and Heisenberg targets up to ten qubits.","lead":"This paper shows that for a Rydberg-atom quantum simulator with only global controls, the symmetries of the control Hamiltonians can make it impossible for a variational quantum eigensolver to reach the target ground state, depending on the qubit number. It provides a symmetry-based test that predicts which VQE runs can succeed, and shows that being able to simulate a Hamiltonian does not mean you can prepare its ground state.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The invariant-subspace criterion is necessary but not sufficient; positive reachability claims for Ising and Heisenberg (4, 8, 10) rest on VQE simulations, not on the symmetry analysis, so the central predictive claim is overstated.","rationale":"The reader's verdict is CONDITIONAL with the weakest assumption being the a priori knowledge of the target ground state. I agree that is a practical limitation and it is explicitly acknowledged in the Conclusion. But I see a more load-bearing logical gap in the central claim: the symmetry test is a one-sided obstruction. All negative predictions (Heisenberg 3, 5, 6, 7, 9) are rigorous and valuable. All positive predictions are not implied by the test; they are only consistent with it, and the paper relies on VQE simulations for those cases. The two-qubit example in Sec. III C makes the insufficiency explicit, so the abstract's 'demonstrating the reliability of our approach in predicting ... could successfully reach' should be qualified: the approach predicts impossibility, not possibility. The suggested test would concretely demonstrate the false-positive behavior of the flowchart if interpreted as a positive predictor. This does not change the overall CONDITIONAL verdict, since the negative results and the empirical VQE demonstrations remain useful, but it sharpens why the paper's central claim should be stated more carefully.","tokens_in":23219,"tokens_out":10899,"duration_ms":129483,"concrete_test":"Apply the published flowchart (Fig. 5) to the two-qubit counterexample mentioned in Sec. III C: take resource Hamiltonians generating SU(2) ⊗ SU(2) acting irreducibly on C^4, initial state |00>, target Bell state. The commutant/isotypic calculation yields a single invariant subspace C^4, so the flowchart returns 'VQE could reach the target state.' But no SU(2) ⊗ SU(2) unitary maps |00> to a Bell state. If the flowchart indeed returns this false positive, it demonstrates that the invariant-subspace test is not sufficient for positive reachability predictions, so the paper's positive claims must be attributed to the VQE simulations rather than to the symmetry analysis.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central necessary condition (Sec. III C) is correct: if initial and target states have support in different invariant subspaces of the resource Lie algebra g_R, the target is unreachable. However, the abstract and Sec. IV present the framework as predicting whether the ground state 'could successfully be reached.' For the positive cases (Ising 3-10; Heisenberg 4, 8, 10), the condition is only that both states lie in the same irreducible subspace, which the authors themselves state in Sec. III C is not sufficient (their SU(2) ⊗ SU(2) example shows an irreducible action that cannot reach entangled states from product states). Thus Table II's positive 'reachability' entries are not symmetry predictions; they are post-hoc matches with VQE runs. The Conclusion acknowledges the need to know the target ground state a priori, but does not address this sufficiency gap. Consequently, the framework reliably rules out unreachable cases but cannot, by itself, certify reachability; positive predictions require either a controllability/sufficiency proof or an explicit empirical disclaimer.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies whether a variational quantum eigensolver (VQE) running on a Rydberg-atom analog simulator with global controls can reach the ground state of a target Hamiltonian. The authors consider the resource Hamiltonians G_R = {H_d, H_Ω, H_Δ} and analyze the invariant subspaces of the Lie algebra they generate, obtained through the commutant and isotypic/irreducible projectors. Their central result is a necessary condition: the initial state and the target ground state must have support in the same invariant subspace of G_R; otherwise VQE is provably unable to reach the target. Applying this condition to Heisenberg and Ising target Hamiltonians on rings of 3–10 qubits, they find that the Heisenberg ground state is unreachable for N = 3, 5, 6, 7, 9 and potentially reachable for N = 4, 8, 10, while the Ising ground state is in the same subspace as the initial state for all considered N. These symmetry-based predictions are compared with numerical VQE simulations, which agree. The paper also contrasts simulability with reachability, shows how changing the initial state can bypass symmetry obstructions in the six-qubit Heisenberg case, and draws connections to adiabatic state preparation. Appendices provide Lie-algebra and symmetry computations with Magma.","tokens_in":23493,"tokens_out":3553,"duration_ms":40540,"significance":"If taken as a framework for necessary conditions, the paper is a solid and useful contribution: the invariant-subspace criterion is mathematically airtight, and the negative predictions for Heisenberg N = 3, 5, 6, 7, 9 are rigorous given the model. The computational approach via the commutant scales better than full Lie-algebra closure and is clearly explained, with reproducible mathematical tools (Magma computations). The explicit acknowledgment that containment in the same irreducible subspace is not sufficient is honest and important. The main weakness is that the positive 'reachability' classifications in Table II are not derived from the symmetry analysis; they are only consistent with VQE simulations, so the abstract's claim of 'predicting whether a given quantum architecture could successfully reach the ground state' overstates what the method delivers. The paper also correctly notes in the Conclusion that the method requires prior knowledge of the target ground state, which limits its practical predictive power for large systems.","major_comments":[{"comment":"The paper explicitly states in Sec. III C that the condition that the initial and target states lie in the same irreducible invariant subspace is not sufficient for reachability, giving the SU(2) ⊗ SU(2) versus SU(4) example. Nevertheless, Table II and the abstract present same-subspace support as 'could reach' or 'could successfully reach' for the Ising target (all N) and for Heisenberg N = 4, 8, 10. These positive entries are not consequences of the symmetry analysis; they are, at best, statements that the symmetry does not forbid reaching the target, corroborated by VQE runs. Please reword the positive claims as 'not ruled out by symmetry' and restrict the predictive claim of the framework to necessary (obstruction) conditions, or supply an additional sufficiency/controllability argument within the irreducible subspace.","section":"Sec. III C and Table II; Abstract"},{"comment":"The method requires knowing the target ground state(s) or at least their invariant-subspace support, which in this paper is obtained by exact diagonalization. In a genuine VQE task the ground state is the quantity being sought, so the framework as presented cannot be used as an a priori predictor for large systems. The authors acknowledge this in the Conclusion, but the abstract's claim that the approach 'demonstrates the reliability of our approach in predicting whether a given quantum architecture could successfully reach the ground state' is too strong. Please qualify the abstract and title-level claims, e.g., by describing the result as necessary-condition diagnostics for small systems with empirical corroboration.","section":"Sec. VII (Conclusion) and Abstract"},{"comment":"The extrapolation beyond seven qubits is explicitly predicated on the observation that extending the resource Hamiltonians by the target Hamiltonian adds at most one center element. The authors state, 'our analysis beyond seven qubits is predicated by general correctness of this observation,' but no proof is provided for N = 8, 9, 10. While the central necessary-condition results for G_R alone do not depend on this assumption, the Appendix C conclusion that adding the target Hamiltonian 'does not help' for N = 8, 9, 10 does. Please either prove the observation for these sizes (the center dimension can be computed via the commutant, as the authors do) or explicitly label that conclusion as conjectural.","section":"Appendix C"}],"minor_comments":[{"comment":"The word 'isoyptic' appears twice in the text around the meataxe description; it should be 'isotypic'.","section":"Appendix B"},{"comment":"In the sentence 'Note that this assumption is not statisfied in all possible cases', 'statisfied' should be 'satisfied'.","section":"Sec. VI"},{"comment":"The phrase 'If the the system starts at τ = 0' contains a duplicated 'the'.","section":"Sec. VI"},{"comment":"The caption contains the typo 'inital' for 'initial'.","section":"Fig. 3 caption"},{"comment":"In the final sentence of the appendix, the enumeration 'C8 ⊞ C6 ⊞ C6 ⊞ C6 ⊞ C6' is described as 'three invariant subspaces', but it lists four components; this should be harmonized with Table IV, which groups them as C8 ⊞ (C6 ⊞ C6) ⊞ (C6 ⊞ C6).","section":"Appendix B"},{"comment":"The number of layers M used in the VQE simulations for Fig. 6 is not specified; only iteration counts are given. Please state M (and the optimizer details) for reproducibility.","section":"Sec. II B"}],"recommendation":"major_revision","confidential_remarks":"The core necessary-condition result is sound and the negative predictions are valuable. The main issue is framing: the abstract and Table II present positive reachability as if it were established by symmetry, whereas the paper itself correctly notes the condition is not sufficient. This is fixable by rewording and by making the status of positive cases explicit. The paper would also benefit from a reproducibility statement: the authors mention Magma computations but do not provide code or data files. I would lean toward major revision rather than rejection because the technical content is correct and the scope can be clarified without new physics."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take. The necessary-condition analysis is correct and the negative predictions are the real result: for the Rydberg ring with global controls, the Heisenberg ground state is unreachable from the trivial product state for N=3,5,6,7,9. That's rigorous, given the model. The companion observation that simulability of the target Hamiltonian does not imply ground-state reachability (three-qubit Heisenberg is simulable but unreachable; four-qubit is not fully simulable but reachable) is clean and worth stealing for a lecture. What's genuinely new is the application of known Lie-algebra/commutant tools to this specific Rydberg VQE setting, and the N-dependent obstruction pattern. The computational description is careful, with exact-field Magma computations; the VQE numerics for N<=10 are consistent with the negative predictions. The paper also states in Sec. III C that containment in the same irreducible subspace is necessary but not sufficient, and gives the SU(2) x SU(2) example. The soft spots are real but not fatal. The abstract and Table II use 'reachability' for the positive cases (Heisenberg 4,8,10; Ising 3-10) where the symmetry analysis only shows non-obstruction; the actual evidence for those entries is the VQE runs, not a sufficiency proof. So the phrase 'predicting whether ... could successfully reach' overstates what the framework alone certifies. The Conclusion does admit the need to know the target ground state a priori, which limits practical large-scale use. Minor: no code or data posted, and the extension beyond seven qubits depends on a plausible but unproven Lie-center assumption (Appendix C). These are all fixable. I'd send this to a serious referee. The core necessary condition is solid, the negative cases are useful, and the message that symmetry obstructions matter for VQE on analog simulators is worth publishing. I'd ask the authors to recalibrate the positive-claim language and consider adding code or a sufficiency discussion.","headline":"A rigorous necessary-condition analysis of VQE reachability for Rydberg global controls, with solid negative predictions; the positive 'reachability' labels outrun what the symmetry argument alone can certify.","tokens_in":23971,"tokens_out":2961,"would_cite":true,"duration_ms":34280,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A variational quantum eigensolver on a Rydberg-atom ring with global controls can reach a target ground state only when the initial state and the target state share an invariant subspace of the resource Hamiltonians; the paper shows this…","keywords":["variational quantum eigensolver","ground-state reachability","invariant subspaces","symmetry analysis","Rydberg atoms","Ising model","Heisenberg model","simulability"],"falsifier":"Compute the invariant-subspace decomposition for an eleven-qubit Rydberg ring with global controls and the Heisenberg target Hamiltonian, and run VQE from the trivial initial state; the symmetry test predicts failure when the initial and exact target ground states have disjoint support, so a single successful convergence to the ground-state energy would refute the claim that the dynamics cannot leave an invariant subspace.","tokens_in":23058,"feed_emoji":"⚛️","tokens_out":9200,"duration_ms":88418,"temperature":0.7,"pith_summary":"The paper asks a practical question: before running a variational quantum eigensolver (VQE) on a Rydberg-atom simulator with global controls, can one tell whether the optimizer has any chance of reaching the target ground state? It argues that the answer is controlled by the symmetries of the resource Hamiltonians: a VQE can reach the target only when the initial state and the target ground state have nonzero support in the same invariant subspace of those resources. Applying this test to ring geometries of three to ten qubits, it predicts that the Heisenberg ground state is unreachable for 3, 5, 6, 7, and 9 qubits and reachable for 4, 8, and 10, while the Ising ground state is reachable for all sizes considered; VQE simulations match these predictions. The paper also shows that Hamiltonian simulability does not imply ground-state reachability, and that choosing an initial state inside the target's invariant subspace can restore reachability.","feed_headline":"Symmetry blocks Rydberg VQE from most Heisenberg ground states","feed_subtitle":"Initial and target states must share an invariant subspace; predictions for 3–10 qubits match VQE runs.","key_machinery":"The central object is the decomposition of the $2^N$-dimensional Hilbert space into invariant subspaces of the resource Hamiltonians. The paper computes this decomposition through the commutant (the set of matrices commuting with all resource Hamiltonians) and then projects states onto isotypic and irreducible blocks using projection operators. The mechanism doing the work is support conservation: the variational dynamics is generated by the resources, so the amount of the wavefunction in each block cannot change, and a target ground state with zero overlap on the initial state's blocks is unreachable. This turns a symmetry calculation into a reachability certificate.","core_discovery":"The paper establishes a necessary condition for ground-state reachability in an analog Rydberg VQE. With resource set $G_R = \\{H_d, H_\\Omega, H_\\Delta\\}$ under global controls, every unitary produced by the variational circuit preserves the invariant subspaces of $G_R$, so any state reached from the trivial initial state stays inside the union of subspaces where that initial state has support. Consequently, if the target ground state has zero support on those subspaces, no amount of optimization can reach it. The paper computes the commutant and the isotypic and irreducible decompositions of $G_R$ for three to ten qubits, projects the initial state and the exactly known target ground states, and finds that the symmetry verdict matches VQE simulations: the Heisenberg ground state is blocked for $N = 3, 5, 6, 7, 9$ and allowed for $N = 4, 8, 10$, whereas the Ising ground state is allowed for all $N = 3, \\ldots, 10$. It further shows that adding the target Hamiltonian to the resources does not remove the obstruction, because the extra center elements act outside the subspace containing both states, and that an initial state chosen inside the target's invariant subspace restores reachability, as demonstrated for six qubits.","pith_inferences":["The paper's method requires the target ground state as input; an extension it leaves open is to predict the symmetry sector from the target Hamiltonian alone, which would turn the diagnostic into a large-scale predictive tool.","The same invariant-subspace test should transfer to other analog simulators and control sets: any global-control device with a nontrivial commutant will have forbidden ground states that could be identified before running VQE.","The six-qubit rescue suggests a practical workflow when VQE fails: compute the projector onto the target's invariant subspace and restart from one of its eigenvectors; but this assumes spectral information that would usually be unavailable, so it is a small-system remedy rather than a general solution.","The adiabatic connection implies that the symmetry test also predicts when a proposed adiabatic path between parent and target Hamiltonians must cross a sector not containing the desired ground state; checking gap closings per sector would be a concrete test of that link."],"forward_implications":["If the paper's symmetry criterion is right, then on the Rydberg ring with global controls a VQE aimed at the Heisenberg ground state from the trivial initial state is bound to fail for $N = 3, 5, 6, 7, 9$.","For $N = 4, 8, 10$, the control structure does not block the Heisenberg ground state, so any observed failure there must come from optimization, noise, or other practical limitations.","The Ising ground state is not symmetry-blocked for any considered size, so the device is in-principle capable of preparing it for $N = 3$ through $10$.","Extending the resource set by the target Hamiltonian itself does not lift the obstruction, since the additional symmetries have no effect on the relevant invariant subspace.","A simple workaround is available: starting from an initial state inside the target ground state's invariant subspace removes the symmetry obstruction, as the six-qubit Heisenberg example shows."],"supporting_citations":[{"why":"Introduces the variational quantum eigensolver whose ground-state reachability is the object of study.","marker":"[15]"},{"why":"Supplies the Lie-algebra and simulability criteria for spin-system control that the paper extends to reachability.","marker":"[63]"},{"why":"Provides the commutant-based symmetry analysis and the two-qubit example showing same-subspace support is necessary but not sufficient.","marker":"[72]"},{"why":"Gives the computational tools for Lie algebras, centers, and symmetries used to compute invariant subspaces.","marker":"[74]"},{"why":"Adapts the block-reduction and ansatz-symmetry techniques that the paper applies to the Rydberg VQE.","marker":"[29]"},{"why":"Defines adiabatic state preparation, the protocol the paper connects to the symmetry analysis.","marker":"[77]"},{"why":"Supplies the meataxe algorithm used to split isotypic subspaces into irreducible ones.","marker":"[96]"},{"why":"Provides the algorithm for computing isotypic projectors from the center of the commutant.","marker":"[101]"}],"fun_headline_variants":["Symmetry bars Rydberg VQE from Heisenberg states","Rydberg VQE blocked by symmetry for Heisenberg targets","Heisenberg ground states unreachable in Rydberg VQE","Symmetry wall: Rydberg VQE can't reach Heisenberg","Rydberg VQE symmetry forbids Heisenberg ground states"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that one already knows the target ground state, or at least which invariant subspace contains it; as the paper itself notes in its conclusion, a practical VQE lacks exactly that spectral information, so the criterion currently works as a diagnostic for small systems rather than a predictive tool for large ones.","fun_headline_variants_meta":{"raw":{"variants":["Symmetry bars Rydberg VQE from Heisenberg states","Rydberg VQE blocked by symmetry for Heisenberg targets","Heisenberg ground states unreachable in Rydberg VQE","Symmetry wall: Rydberg VQE can't reach Heisenberg","Rydberg VQE symmetry forbids Heisenberg ground states"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000177,"raw_usage":{"total_tokens":1310,"prompt_tokens":980,"completion_tokens":330,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":596,"completion_tokens_details":{"reasoning_tokens":242}},"tokens_in":596,"tokens_out":330,"duration_ms":3822,"temperature":1.0,"reasoning_tokens":242,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T22:05:36.087277+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the invariant-subspace decomposition for an eleven-qubit Rydberg ring with global controls and the Heisenberg target Hamiltonian, and run VQE from the trivial initial state; the symmetry test predicts failure when the initial and exact target ground states have disjoint support, so a single successful convergence to the ground-state energy would refute the claim that the dynamics cannot leave an invariant subspace.","supporting_citations":[{"cited_title":"Zimbor´ as, R","cited_arxiv_id":null,"evidence_quote":"Supplies the Lie-algebra and simulability criteria for spin-system control that the paper extends to reachability."},{"cited_title":"Zeier and T","cited_arxiv_id":null,"evidence_quote":"Provides the commutant-based symmetry analysis and the two-qubit example showing same-subspace support is necessary but not sufficient."},{"cited_title":"Zimbor´ as, R","cited_arxiv_id":null,"evidence_quote":"Gives the computational tools for Lie algebras, centers, and symmetries used to compute invariant subspaces."},{"cited_title":"Lux and H","cited_arxiv_id":null,"evidence_quote":"Supplies the meataxe algorithm used to split isotypic subspaces into irreducible ones."},{"cited_title":"Eberly and M","cited_arxiv_id":null,"evidence_quote":"Provides the algorithm for computing isotypic projectors from the center of the commutant."}],"review_version":1}