{"id":"c8815a57-cf5a-45d4-9af1-e8c109ca4699","arxiv_id":"2411.15256","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"For the quantum Rabi model, the Levy-Lieb functional is almost fully explicit along the adiabatic connection, and all regular density pairs are uniquely v-representable.","lead":"This paper develops density-functional theory for the simplest quantum electrodynamics system: one two-level atom coupled to one photon mode. It derives exact explicit formulas for the universal functional and shows that every allowed polarization and photon displacement comes from a unique ground state.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The bijection, v-representability, convexity and differentiability claims all rest on companion-paper theorems [22] that are cited but not re-proven; a gap there would invalidate the central claim.","rationale":"The paper's central claim — unique v-representability for all regular (σ,ξ), with F_LL = F_L convex and differentiable, and an explicit adiabatic connection — hinges on the existence and Lagrange-multiplier theorems imported from [22]. These are not re-derived, and the reader correctly identified this as the weakest assumption. I checked the self-contained parts of the paper for internal consistency: the displacement rule (Thm. IV.2.2), the virial relation (Thm. IV.2.4), and the kinetic-hopping identities (Thm. IV.2.5-6) all follow from the stated transformations by direct calculation; the bounds on I^λ(σ) in Section V.C follow from the Gaussian trial states; and the exchange-energy zero result (Thm. V.3) is consistent with the derivative of I^λ at λ=0 vanishing. No internal contradiction surfaces. The critical unresolved point is whether the companion paper's proofs actually deliver what is claimed, especially the regularity of the constraint manifold and the conclusion that the optimizer is a ground state. This is a testable mathematical matter: a third party can inspect [22] and check the three theorems. Because the present paper is otherwise mathematically coherent and the external theorems are plausible, the conditional verdict stands unchanged. The available numerical code and the explicit bounds provide independent support, but they do not replace the missing proofs.","tokens_in":39690,"tokens_out":12579,"duration_ms":119136,"concrete_test":"Independently verify the proofs of Theorems 3.4, 3.7 and 3.18 in the companion paper [22] (with an expert referee or a written appendix), checking specifically: (i) the constrained-search infimum is attained for every (σ,ξ)∈[-1,1]×R within the form domain Q0; (ii) the Lagrange multiplier system for the three constraints (norm, σ, ξ) has a unique solution (E,v,j) for every σ∈(-1,1); and (iii) the second-order condition implies that the optimizer is the global ground state of H(v,j), not merely a stationary eigenstate. If any step fails, the central v-representability and differentiability results collapse; if all pass, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Corollary IV.8, which asserts unique pure-state v-representability, F_LL = F_L, and differentiability, depends on Theorems IV.3, IV.4 and IV.5. These are not proved in the present paper but are quoted verbatim from Bakkestuen et al. [22]: Theorem IV.3 cites [22, Thm. 3.4] for existence of optimizers; Theorem IV.4 cites [22, Thm. 3.7] for the Lagrange-multiplier characterization; and Theorem IV.5 cites [22, Thm. 3.18]. The subsequent proofs of the adiabatic connection (Theorem V.2), the bounds on I^λ(σ), and the photon-free effective potential all assume that for every regular (σ,ξ) there is a unique strictly positive optimizer satisfying the Schrödinger equation with some (v,j). If the companion-paper proofs have a gap — for instance, if the constraint set is not a regular C^1 manifold at some σ∈(-1,1), or if the second-order condition does not force the optimizer to be the global ground state — then the bijection between (v,j) and (-1,1)×R fails, and with it the explicit functional form. The paper's own derivations (Theorem IV.2 properties, displacement rule, virial relations, and the bounds in Section V) are self-consistent and check out, but they presuppose the existence and eigenvalue structure of the optimizers. Thus the external dependency is the single most load-bearing assumption of the paper.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops quantum-electrodynamical density-functional theory (QEDFT) for the quantum Rabi model, taking the polarization σ and the photon-field displacement ξ as the internal density variables, and discusses extensions to the Dicke model. It establishes a Hohenberg–Kohn-type injection for regular densities, proves structural properties of the Levy–Lieb functional (symmetry, displacement rule, virial relations, positivity of optimizers), and imports from the companion paper [22] the existence, Lagrange-multiplier, and ground-state characterizations that yield unique pure-state v-representability for all (σ,ξ)∈(−1,1)×R, equality of the Levy–Lieb and Lieb functionals, and differentiability. The paper then derives an explicit zero-coupling functional, an almost explicit adiabatic connection F^λ_LL(σ,ξ) = ω/2 − t√(1−σ²) + ω²ξ²/2 + λgσξ − λ²g²(1−σ²)/(2ω²) + I^λ(σ), rigorous bounds on the non-explicit correlation term I^λ(σ), and photon-free effective potentials. The analytical derivations in the paper are largely self-consistent and check out, but the central v-representability and differentiability results rest on external theorems from [22] rather than on proofs given in the present manuscript.","tokens_in":39996,"tokens_out":11865,"duration_ms":127418,"significance":"If the companion-paper theorems are accepted, this is a valuable and nearly unique model-system demonstration of QEDFT: it gives explicit v-representability, convexity, differentiability, and an almost closed-form adiabatic connection in a setting that still has nontrivial light–matter correlation. The constructive N-representability proof, the displacement rule, the zero-coupling optimizer, the virial relations, and the analytic bounds on I^λ(σ) are genuine strengths, as is the public availability of the numerical code. The main caveat is that the strongest claims—especially the bijection between (−1,1)×R and (v,j), FLL=FL, and differentiability—are not proved self-contained but are quoted from [22]; the paper would be substantially stronger if those theorems were stated and proved, or if the dependency were made unavoidable and explicit.","major_comments":[{"comment":"The central v-representability result of the paper is not self-contained. Theorem IV.3 (existence of optimizers), Theorem IV.4 (Lagrange-multiplier characterization), and Theorem IV.5 (critical boundary case) are imported from the companion paper [22, Theorems 3.4, 3.7, 3.18] and are not proved here. Corollary IV.8 then derives unique pure-state v-representability, FLL=FL, and differentiability from these imported statements. Since these are load-bearing for the paper's main claim, the manuscript should either reproduce the proofs, state the imported theorems verbatim with their exact hypotheses, or explicitly flag that the main theorem is conditional on [22]. If [22] is not yet published, an appendix with the missing proofs is needed.","section":"§IV.C, Theorems IV.3–IV.7 and Corollary IV.8"},{"comment":"The proof of Theorem IV.7 is incomplete as written. The text argues that the optimizer is the ground state because it 'has all the properties that we showed for ground states in Section IIC, in particular that it is unique.' However, Section IIC only proves that the unique ground state is strictly positive; it does not in itself exclude a strictly positive excited eigenstate. The missing step is an orthogonality/Perron–Frobenius argument showing that any strictly positive eigenfunction of H(v,j) must be the unique ground state. This step is directly load-bearing for the conclusion that every regular density pair is pure-state v-representable and should be supplied explicitly.","section":"§IV.C, proof of Theorem IV.7"},{"comment":"The statements lim_{λ→∞} a(λ,t)=1, lim_{λ→∞} b(λ,t)=t, and consequently I^∞(σ)=t√(1−σ²), are presented as established facts ('This shows, in particular...'), but they are inferred from numerical fits to the conjectured functional form of Conjecture V.5. The rigorous bounds in Eqs. (39) and (41) do not prove saturation of the upper bound. These claims should either be proved analytically or clearly labeled as numerical evidence for a conjecture rather than as derived limits.","section":"§V.D, strictly correlated regime"}],"minor_comments":[{"comment":"The notation would be clearer if ψ^λ and φ^ν were always explicitly tied to their density pairs: ψ^λ is the optimizer of F^λ_LL(σ,ξ) while φ^ν is the optimizer of F^ν_LL(σ,0). As written, the reader must infer this distinction from context.","section":"§V.A, Eq. (33) and Eq. (35)"},{"comment":"The parameter-fitting procedure behind Figs. 11 and 12 is not described in the text: the fit functional form, the fitted data range, and the error model are missing. The statement that the largest standard deviation is 0.05 refers only to b; reporting errors for a and the joint fit would be more informative.","section":"§V.D, Conjecture V.5"},{"comment":"The correlation factor η_c(g) is introduced after Eq. (46) without a derivation, with a citation to [34]. Since η_c is used in the main numerical comparison, the paper should state clearly whether η_c is derived, fitted, or taken from the cited work, and how it is computed for each g.","section":"§VI.A, Eq. (46) and Figs. 13–14"},{"comment":"The bounds on I^λ(σ) are rigorous and clearly stated, but it would help to note explicitly that the two upper bounds are complementary: the exponential bound is tighter for large λ and the λ²g²/(2ω²) bound is relevant for small λ. This is implicit in the text but could be made explicit.","section":"§V.C, Eqs. (39) and (41)"}],"recommendation":"major_revision","confidential_remarks":"The paper is well within the scope of the journal and is a serious, mostly careful contribution. The main risk is the external dependency on the companion paper [22] for the existence and Lagrange-multiplier theorems that underpin the central v-representability and differentiability claims. The proof of Theorem IV.7 also has a small but real gap that should be fixed. I would not reject on these grounds, as the issues appear fixable within the manuscript's scope, but the main theorem should not be presented as fully self-contained until the dependency is addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis is a solid, useful paper. It specializes the rigorous QEDFT framework the same group built for the Dicke model down to the quantum Rabi model, where the structure is simple enough to write the Levy–Lieb functional almost explicitly. That concreteness is the real contribution: the zero-coupling functional, the displacement rule, the virial relations, and the adiabatic-connection bounds are genuinely new and make the paper a strong pedagogical benchmark for QEDFT.\n\nWhat it does well: the derivation of Theorem IV.2 is clean, the bounds on I^λ are proven analytically and checked numerically, and the paper is transparent about what is approximate—the fitted a(λ,t), b(λ,t), and η_c are labeled as such, and the code is on GitHub. The Hohenberg–Kohn theorem for the Rabi model, giving a bijection between (v,j) and (-1,1)×R, is stated carefully, and the regular/critical polarization distinction is handled correctly.\n\nSoft spots: the central v-representability and differentiability results (Theorems IV.3–IV.5, Corollary IV.8) are imported from the companion paper [22] and not re-proven here. That is normal for a sequel, but it means the reader must trust [22] for the load-bearing existence and Lagrange-multiplier machinery. I do not see an error in the present paper's own derivations—the displacement rule, the adiabatic connection, and the bounds are self-consistent—but a referee should verify that the cited theorems in [22] actually deliver what this paper claims. The numerics in Section V are also described briefly; a few more details on how F_LL^λ was computed would strengthen reproducibility.\n\nThe circularity concern raised in the stress-test note is, I think, overblown. The functional expressions derived here do not assume the conclusion beyond what [22] establishes; the derived parts are internally consistent. So the external dependency is a legitimate caveat, not a flaw.\n\nWho this is for: anyone in QEDFT who wants a fully solved toy model, or anyone teaching DFT concepts via a nontrivial exact example. It deserves a serious referee. The main request to the authors should be to either reproduce the key theorems from [22] in an appendix or explicitly state which parts of the central claims are proven here versus imported.\n\nRecommendation: send to peer review, with a referee who knows the companion paper.","headline":"Explicit QEDFT universal functional for the Rabi model, with v-representability and a nearly closed adiabatic connection; the main caveat is that the load-bearing existence theorems are inherited from the companion paper.","tokens_in":40531,"tokens_out":2054,"would_cite":true,"duration_ms":39501,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81Q05","81Q10","81V80","46N50"],"pacs":["03.65.-w","31.15.Ew","42.50.Pq"],"model":"deepseek-v4-flash","headline":"For the quantum Rabi model, every regular density pair is realized by exactly one ground state.","keywords":["quantum Rabi model","quantum-electrodynamical density-functional theory","Levy-Lieb functional","v-representability","adiabatic connection","Dicke model","polarization and photon displacement","convex analysis"],"falsifier":"A dense numerical scan of ground states of $\\hat{H}(v,j)$ over a large box in $(v,j)$-space should hit every $(\\sigma,\\xi)\\in(-1,1)\\times\\mathbb{R}$ exactly once; finding a pair that is missed, or two different $(v,j)$ producing the same pair, would refute the bijection. Alternatively, testing the proposed ellipse form of $I^\\lambda(\\sigma)$ against the exact upper bound at large $\\lambda$, where $I^\\infty(\\sigma)=t\\sqrt{1-\\sigma^2}$, would settle the strong-coupling saturation claim.","tokens_in":39526,"feed_emoji":"⚛️","tokens_out":12080,"duration_ms":97038,"temperature":0.7,"pith_summary":"This paper proves that the quantum Rabi model—one two-level system coupled to a single photon mode—has a fully rigorous density-functional theory. Taking the polarization $\\sigma$ and the photon-field displacement $\\xi$ as the density variables, it shows that every pair $(\\sigma,\\xi)$ with $\\sigma \\in (-1,1)$ is the ground state of exactly one Hamiltonian with external pair $(v,j)$. The constrained-search Levy–Lieb functional $F_{\\mathrm{LL}}$ is therefore convex, differentiable, and equal to the mixed-state Lieb functional at these points. Along the adiabatic connection the functional is given almost explicitly, $F^\\lambda_{\\mathrm{LL}}(\\sigma,\\xi)=\\frac{\\omega}{2}-t\\sqrt{1-\\sigma^2}+\\frac{\\omega^2}{2}\\xi^2+\\lambda g\\sigma\\xi-\\frac{\\lambda^2 g^2}{2\\omega^2}(1-\\sigma^2)+I^\\lambda(\\sigma)$, with the correlation remainder $I^\\lambda(\\sigma)$ bounded between $0$ and $\\min\\{\\frac{\\lambda^2g^2}{2\\omega^2}(1-\\sigma^2),\\,t\\sqrt{1-\\sigma^2}(1-e^{-\\lambda^2g^2/\\omega^3})\\}$. Because the model is small enough for exact analysis, it gives the clearest available test bed for the mathematical machinery of QEDFT.","feed_headline":"Every Rabi density pair fixes a unique external pair","feed_subtitle":"A two-level system plus one photon mode admits an exact, almost explicit universal functional.","key_machinery":"The central object is the Levy–Lieb constrained-search functional $F_{\\mathrm{LL}}(\\sigma,\\xi)=\\inf_{\\psi\\in M_{\\sigma,\\xi}}\\langle\\psi|\\hat{H}_0|\\psi\\rangle$, minimized over normalized admissible states with fixed polarization and displacement. Its analysis uses the displacement rule $F_{\\mathrm{LL}}(\\sigma,\\xi+\\zeta)=F_{\\mathrm{LL}}(\\sigma,\\xi)+\\omega^2\\zeta(\\xi+\\zeta/2)+g\\sigma\\zeta$, virial and hypervirial identities, and explicit Gaussian trial states. For regular $\\sigma$, Lagrange-multiplier theorems from the companion paper turn optimizers into ground states of $\\hat{H}(v,j)$, which yields uniqueness, convexity, and differentiability. The only non-explicit piece in the adiabatic connection is $I^\\lambda(\\sigma)=-\\frac{4tg}{\\omega^2}\\int_0^\\lambda \\int (\\varphi^\\nu_+)' \\varphi^\\nu_- \\,dq\\,d\\nu$, which is bounded analytically by two trial-state comparisons and numerically fitted to an ellipse-like form.","core_discovery":"The paper's central claim is that the map $(v,j)\\mapsto(\\sigma,\\xi)$ from external potentials to ground-state density pairs is a bijection from $\\mathbb{R}^2$ onto $(-1,1)\\times\\mathbb{R}$. Thus every regular density pair is pure-state $v$-representable, and the representing pair is unique, not merely up to a constant. From this, $F_{\\mathrm{LL}}$ is differentiable and convex on regular densities and coincides with the Lieb functional $F_L$. The paper also derives the adiabatic connection for $F^\\lambda_{\\mathrm{LL}}$ with the coupling $\\lambda$ scaling the light-matter interaction, leaving only the bounded correlation term $I^\\lambda(\\sigma)$ non-explicit, and proves that the exchange energy vanishes exactly, so the whole beyond-direct-coupling contribution is correlation. For critical polarizations $\\sigma=\\pm1$, no ground state realizes these pairs and the functional has the explicit values $F_{\\mathrm{LL}}(\\pm1,\\xi)=\\omega/2\\pm g\\xi+\\omega^2\\xi^2/2$.","pith_inferences":["The same constrained-search analysis could be pushed to the multi-mode Dicke model; the paper leaves $v$-representability there open, and a failure would highlight what the harmonic confinement in the single-mode Rabi model provides.","The numerical ellipse ansatz for $I^\\lambda(\\sigma)$ suggests that exact or near-exact closed forms for correlation functionals may exist for other integrable light-matter models, and the fitted functions $a(\\lambda,t)$, $b(\\lambda,t)$ could be tested against the analytic bounds.","If the saturation $I^\\infty(\\sigma)=t\\sqrt{1-\\sigma^2}$ holds, then in the strictly correlated limit the kinetic contribution is fully quenched; this is a concrete prediction for strong-coupling QEDFT functional construction.","The photon-free approximation's site-coupling in the Dicke model implies an effective matter-matter interaction mediated by the eliminated photon mode—an effect that in macroscopic ensembles could be probed experimentally via polarization shifts at strong coupling."],"forward_implications":["Because the map $(v,j)\\mapsto(\\sigma,\\xi)$ is onto for all $\\sigma\\in(-1,1)$, QEDFT ground-state calculations for the Rabi model need not worry about missing $v$-representable densities or about mixed-state versus pure-state formulations.","The equality $F_{\\mathrm{LL}}=F_L$ turns the universal functional into a convex differentiable object, so subdifferential calculus and Legendre-transform methods apply without further approximation.","The almost-explicit adiabatic connection reduces the correlation energy of the model to a single bounded function $I^\\lambda(\\sigma)$, giving an exact benchmark for approximate QEDFT correlation functionals.","The vanishing exchange energy means that in this model all matter-photon interactions beyond the direct term are correlation, clarifying the target of approximate functionals in cavity QED.","The photon-free effective potential $v^{\\mathrm{pf},\\eta_c}_{\\mathrm{dc}}=g\\xi+\\eta_c g^2\\sigma/\\omega^2$ with $\\eta_c\\to1$ for strong coupling reproduces the exact potential except near $\\sigma=\\pm1$, offering a controlled approximation route."],"supporting_citations":[{"why":"Supplies the existence of Levy–Lieb optimizers, the Lagrange-multiplier characterization, and the Dicke-model Hohenberg–Kohn analysis that the paper extends.","marker":"[22]"},{"why":"Defines the Lieb functional and the convex-analysis framework used to conclude F_LL equals F_L and to discuss differentiability.","marker":"[20]"},{"why":"The original Hohenberg–Kohn theorem whose injectivity result the Rabi-model theorem sharpens to a bijection.","marker":"[18]"},{"why":"Levy's constrained-search construction that defines the functional F_LL used throughout.","marker":"[47]"},{"why":"Adiabatic connection integral representation that the paper adapts to the coupling-constant scaling of the Rabi model.","marker":"[54]"},{"why":"Provides the Newton–Leibniz trick and superdifferential formula used to derive the integral representation of F^lambda_LL.","marker":"[57]"},{"why":"Electron-photon exchange-correlation approximation that motivates the correlation factor eta_c in the photon-free effective potential.","marker":"[34]"},{"why":"Photon-free effective framework for Pauli–Fierz Hamiltonians that the Rabi-model photon-free Hamiltonian is compared to.","marker":"[75]"}],"fun_headline_variants":["Rabi model gives exact QEDFT density-potential map","Quantum Rabi model yields unique QEDFT density pairs","Exact bijection between potentials and densities in QED","Rabi model proves QEDFT functional is exact and explicit"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument inherits, without re-proving, the companion paper's theorems that constrained-search minimizers exist and satisfy the Lagrange-multiplier equations; if those theorems have gaps, the bijection and differentiability conclusions do not follow.","fun_headline_variants_meta":{"raw":{"variants":["Rabi model gives exact QEDFT density-potential map","Quantum Rabi model yields unique QEDFT density pairs","Exact bijection between potentials and densities in QED","Rabi model proves QEDFT functional is exact and explicit"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000436,"raw_usage":{"total_tokens":2202,"prompt_tokens":913,"completion_tokens":1289,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":529,"completion_tokens_details":{"reasoning_tokens":1220}},"tokens_in":529,"tokens_out":1289,"duration_ms":9918,"temperature":1.0,"reasoning_tokens":1220,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:51:57.341632+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A dense numerical scan of ground states of $\\hat{H}(v,j)$ over a large box in $(v,j)$-space should hit every $(\\sigma,\\xi)\\in(-1,1)\\times\\mathbb{R}$ exactly once; finding a pair that is missed, or two different $(v,j)$ producing the same pair, would refute the bijection. Alternatively, testing the proposed ellipse form of $I^\\lambda(\\sigma)$ against the exact upper bound at large $\\lambda$, where $I^\\infty(\\sigma)=t\\sqrt{1-\\sigma^2}$, would settle the strong-coupling saturation claim.","supporting_citations":[{"cited_title":"Density-functional theory for the Dicke Hamiltonian","cited_arxiv_id":"2409.13767","evidence_quote":"Supplies the existence of Levy–Lieb optimizers, the Lagrange-multiplier characterization, and the Dicke-model Hohenberg–Kohn analysis that the paper extends."},{"cited_title":"Density functionals for Coulomb-systems,","cited_arxiv_id":null,"evidence_quote":"Defines the Lieb functional and the convex-analysis framework used to conclude F_LL equals F_L and to discuss differentiability."},{"cited_title":"Universal variational functionals of electron densities, first- order density matrices, and natural spin-orbitals and solution of the v- representability problem,","cited_arxiv_id":null,"evidence_quote":"Levy's constrained-search construction that defines the functional F_LL used throughout."},{"cited_title":"Theexchange-correlationenergyofametal- lic surface,","cited_arxiv_id":null,"evidence_quote":"Adiabatic connection integral representation that the paper adapts to the coupling-constant scaling of the Rabi model."},{"cited_title":"Exchange-onlyvirialrelationfromthe adiabatic connection,","cited_arxiv_id":null,"evidence_quote":"Provides the Newton–Leibniz trick and superdifferential formula used to derive the integral representation of F^lambda_LL."},{"cited_title":"Electron-photon exchange-correlation approximation for quantum-electrodynamical density-functional theory,","cited_arxiv_id":null,"evidence_quote":"Electron-photon exchange-correlation approximation that motivates the correlation factor eta_c in the photon-free effective potential."},{"cited_title":"Mak- ing ab initio QED functional(s): Nonperturbative and photon-free effec- tive frameworks for strong light–matter coupling,","cited_arxiv_id":null,"evidence_quote":"Photon-free effective framework for Pauli–Fierz Hamiltonians that the Rabi-model photon-free Hamiltonian is compared to."}],"review_version":1}