REVIEW 3 major objections 7 minor 28 references
Lower bound on the energy-momentum relation of the polaron
T0 review · 3 major / 7 minor · reviewed 2026-07-07 · glm-5.2
Pith's one-line read Lower bound links polaron energy to vacuum overlap and spectral gap
desk verdict Solid paper extending polaron energy-momentum bounds and effective mass formula beyond the Fröhlich model; the Feynman–Kac formula for v not in L² is the real technical contribution. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The renewal transform of a probability measure (from a companion framework by Hinrichs and Polzer), applied to the spectral measure of H(P) with respect to the Fock vacuum. The finite-volume renewal transforms are identified with a Poisson point process representation derived from the Dyson expansion of Desio and Seiringer. A rank-one perturbation comparison via the Sherman-Morrison formula then yields the explicit bound.
What would settle it
Exhibit a polaron model satisfying Assumption 1 for which the bound of Theorem 1 is violated, i.e., find P, P0 with |P| >= |P0| such that E(P) < E(P0) + (1/2)(Delta(P0) + delta) - sqrt((1/4)(Delta(P0) + delta)^2 - Delta(P0) rho(P0) delta). Alternatively, find a model where the renewal transform identification (Proposition 5) or the dormant-time bound (equation 29) fails.
Extended reading notes
Core claim
The central mechanism is the identification of the renewal transform of the spectral measure of H(P) with the law of a point process derived from the Dyson expansion. This allows the energy E(P) to be compared to the ground-state energy of a rank-one perturbation A(P) = H(P0) + delta<Omega, .>Omega, whose spectral properties are explicitly computable. The resulting lower bound on E(P) - E(P0) depends only on rho(P0), Delta(P0), and the kinetic energy difference delta = (1/2)(|P|^2 - |P0|^2), making it the first bound of its kind that is expressed purely in terms of spectral data at the reference momentum.
Load-bearing premise
The proof depends on the inequality that the dormant time of the renewal process is bounded above by the Gaussian variance (equation 29), which links the stochastic representation to the spectral comparison. If this bound or the identification of the renewal transform with the path measure fails for models outside the stated regularity class, the main lower bound would not hold.
Editorial extensions
If this is right
- For the Fröhlich polaron, the bound provides a concrete lower bound on E(P) - E(0) for all P, expressed in terms of rho(0) and Delta(0), which could be combined with numerical or variational estimates of these quantities to constrain the energy-momentum relation.
- The equality m_eff^{-1} = lim sigma_hat^2_T(P) resolves a question raised in the literature about whether sub-diffusivity of the path measure implies infinite effective mass for general polaron-type models, not just the Fröhlich case.
- Corollary 1 gives a sharp criterion for boundedness of I0 for the Fröhlich polaron: I0 is bounded if and only if lim inf sigma_hat^2_T(P) = 0 for some P, connecting a spectral question to a purely probabilistic one about the path measure.
- The Feynman-Kac formula (Theorem 2) is extended to models where the coupling v is not square-integrable, removing the need for ultraviolet regularization in the path integral representation.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper establishes a lower bound on the energy-momentum relation E(P) for a class of polaron-type models satisfying Assumption 1. The bound (Theorem 1) is expressed in terms of the vacuum overlap rho(P0) and the spectral gap Delta(P0) of H(P0), and holds for all P with |P| >= |P0|. The proof proceeds via a renewal transform framework (building on [HP25]) applied to a rank-one perturbation A(P) = H(P0) + delta<Omega, .>Omega, combined with a point process representation of the partition function derived from the Dyson expansion of [DS25]. Additionally, Theorem 2 establishes a Feynman-Kac formula valid for v not in L^2, Theorem 3 proves convergence of the rescaled mean square displacement to the inverse effective mass under right-continuity of rho, and Corollary 1 gives a necessary and sufficient criterion for boundedness of I_0 for the Frohlich polaron.
Significance. The paper makes several contributions of genuine interest to the polaron and mathematical physics community. The lower bound in Theorem 1 is parameter-free, expressed purely in spectral quantities (rho and Delta), and applies to a broad class of models including the Frohlich polaron and the Nelson model. The extension of the Feynman-Kac formula (Theorem 2) to the v not in L^2 case via the Dyson expansion approach is a useful technical contribution. Theorem 3 answers a question raised in [BSS25b] by establishing m_eff^{-1} = lim sigma_hat^2_T for general polaron models under right-continuity of rho, and Corollary 1 provides a clean probabilistic criterion for the boundedness of I_0. The proof chain is carefully executed: the renewal transform identification (Proposition 2), the point process representation (Proposition 4), the Gaussian rewriting (Proposition 7), and the final rank-one perturbation argument via Sherman-Morrison are all clearly laid out. The v not in L^2 case is handled via the v_epsilon approximation and monotone convergence in Proposition 7, and inequality (29) is a purely Gaussian statement independent of v.
major comments (3)
- Theorem 1, proof (Section 4): The key inequality chain is E(P) >= inf supp mu_P, where mu_P is the spectral measure of A(P) = H(P0) + delta<Omega, .>Omega. The bound on s(lambda) uses that the spectral measure of H(P0) is supported on {E(P0)} union [E(P0)+Delta(P0), infinity), which follows from the definitions. The upper bound s(lambda) <= rho(P0)/(E(P0)-lambda) + (1-rho(P0))/(E(P0)+Delta(P0)-lambda) then yields the explicit lower bound by solving equation (34). This argument is sound. However, the step from E(P) >= inf supp mu_P to the explicit bound deserves one clarification: the claim that f(P) is an atom of mu_P when f(P) < E(P0) + Delta(P0) follows from the Sherman-Morrison formula and the fact that finite-rank perturbations leave the essential spectrum invariant, but the argument that f(P) cannot lie in the continuous spectrum of mu_P below E(P0) + Delta(P0) is only implicitly sk
- Theorem 3, part (2): The right-continuity assumption on rho is a genuine limitation, but it is clearly stated and does not affect Theorem 1. The author notes that it holds for the Frohlich polaron where rho is analytic on I_0. The proof uses the right-continuity to show lim sup f(T) >= 1, where f(T) is the ratio of partition functions at P_T and P. The argument that lim sup f(T) >= lim sup rho(P_T)/rho(P) = 1 under right-continuity is correct, but the intermediate step (the '+o(T)' term in the displayed equation following the definition of f(T)) is not explicitly justified. A brief comment on why the o(T) term vanishes in the limit would strengthen the proof.
- Proposition 5: The identification of the finite-volume renewal transform P_{P,T} with the distribution of (Y_hat_t) under Xi_hat_{P,T} relies on the product structure of F_T under disjoint clusters of intervals. The proof verifies this on an intersection-stable generator (finite sets of time points where Y_hat = 0), which is sufficient by the pi-lambda theorem. This is correct. One minor concern: the proof uses the independence of eta_{r_i, r_{i+1}} for disjoint intervals, which follows from the Poisson property, but the conditional independence argument on the event A (where no intervals cross the partition points) should perhaps note that A is independent of the restrictions eta_{r_i, r_{i+1}} precisely because of the Poisson structure. This is implicit but could be stated more explicitly.
minor comments (7)
- In the proof of Theorem 1, the notation delta(P, P0) is used for both the scalar 1/2(|P|^2 - |P0|^2) and as a parameter in the rank-one perturbation. While context disambiguates, a brief remark would improve readability.
- The notation for the measures (e.g., Xi_hat_{P,T}, Theta_hat_{P,T}, bP_{alpha,t}) uses hats and tildes in a way that is sometimes hard to distinguish. A summary table of notation would help the reader.
- In equation (34) and the surrounding text, the variable lambda ranges over (E(P0), E(P0)+Delta(P0)), but the spectral measure of H(P0) may have support below E(P0)+Delta(P0) if there are excited eigenvalues. The bound uses that any such eigenvalues would only decrease s(lambda), which is correct, but a brief remark clarifying this point would align better with the spectral setup.
- The paper states that the lower bound is suboptimal at large P (the limit as P -> infinity is Delta(P0)*rho(P0), which is strictly less than E_ess(P0) - E(P0)). This is an honest acknowledgment, but a brief discussion of whether the bound could be improved by a different choice of rank-one perturbation or a multi-step argument would add context.
- In Proposition 7, the v not in L^2 case is handled by approximating v with v_epsilon defined via mu_epsilon. The monotone convergence argument is correct, but the statement that |v_epsilon|^2 increases to |v|^2 should perhaps note that this follows from the monotone convergence of the measures mu_epsilon to mu.
- Reference [DS25] is cited as a preprint. If it has been accepted for publication, the reference should be updated.
- In the definition of sigma^2_T(xi, u) following equation (27), it would be helpful to clarify that this is well-defined because P_{xi,u} is a Gaussian measure on path space and X_{0,T} is a linear functional.
Simulated Author's Rebuttal
The authors thank the referee for a careful and constructive report. All three major comments request clarifications that can be incorporated in a minor revision. The standing objections list is empty.
read point-by-point responses
-
Referee: Theorem 1, proof (Section 4): The claim that f(P) is an atom of mu_P when f(P) < E(P0) + Delta(P0) follows from the Sherman-Morrison formula and the fact that finite-rank perturbations leave the essential spectrum invariant, but the argument that f(P) cannot lie in the continuous spectrum of mu_P below E(P0) + Delta(P0) is only implicitly sketched.
Authors: We thank the referee for this observation. The argument is indeed correct but should be stated more explicitly. The key point is as follows. By Weyl's theorem, the essential spectrum is invariant under finite-rank perturbations, so the essential spectrum of A(P) = H(P0) + delta<Omega, .>Omega coincides with that of H(P0). Since the spectral measure of H(P0) is supported on {E(P0)} union [E(P0)+Delta(P0), infinity), the essential spectrum of A(P) is contained in [E(P0)+Delta(P0), infinity). Therefore, the spectrum of A(P) below E(P0)+Delta(P0) consists entirely of isolated eigenvalues of finite multiplicity, i.e., it is purely discrete. In particular, if f(P) = inf supp mu_P < E(P0)+Delta(P0), then f(P) lies in the discrete spectrum of A(P), hence is an eigenvalue, and therefore f(P) is an atom of mu_P. We will add a sentence spelling out this Weyl theorem argument explicitly in the revision. revision: yes
-
Referee: Theorem 3, part (2): The intermediate step (the '+o(T)' term in the displayed equation following the definition of f(T)) is not explicitly justified. A brief comment on why the o(T) term vanishes in the limit would strengthen the proof.
Authors: The referee is correct that this step deserves justification. The notation '+o(T)' is imprecise; what is meant is that the contributions from the continuous spectrum decay exponentially in T. More precisely, we have the spectral decomposition: <Omega, e^{-T(H(P)-E(P))} Omega> = rho(P) + integral over [E(P)+Delta(P), infinity) of e^{-T(x-E(P))} mu_P(dx), where mu_P is the spectral measure of H(P) with respect to Omega. The integral term is bounded by e^{-T Delta(P)} (since the integrand is at most e^{-T Delta(P)} times the total mass, which is at most 1), and hence is o(1) as T -> infinity. The same applies to the numerator with P replaced by P_T. Therefore f(T) = [rho(P_T) + o(1)] / [rho(P) + o(1)], and since rho(P) > 0 by assumption, we obtain lim sup f(T) >= lim sup rho(P_T)/rho(P) = 1 by right-continuity of rho. We will replace the imprecise '+o(T)' notation with the explicit exponential decay bound and add a brief justification in the revised proof. revision: yes
-
Referee: Proposition 5: The conditional independence argument on the event A (where no intervals cross the partition points) should perhaps note that A is independent of the restrictions eta_{r_i, r_{i+1}} precisely because of the Poisson structure. This is implicit but could be stated more explicitly.
Authors: We agree that this point should be stated more explicitly. The independence of the event A = {N(tilde{eta}_{0,r_1}) = ... = N(tilde{eta}_{r_{k-1},r_k}) = 0} from the restrictions (eta_{0,r_1}, eta_{r_1,r_2}, ..., eta_{r_k,T}) follows directly from the Poisson property: the restrictions of a Poisson point process to disjoint measurable subsets are independent random variables. Since A is defined entirely in terms of the restrictions to the 'off-diagonal' regions (r_i, r_{i+1}] x (r_{i+1}, T], which are disjoint from the 'on-diagonal' regions (r_i, r_{i+1}] x (r_i, r_{i+1}] that determine eta_{r_i, r_{i+1}}, the independence follows. We will add a sentence making this Poisson independence explicit in the revision. revision: yes
Circularity Check
No significant circularity found; the derivation chain is self-contained given cited tools
full rationale
I traced the full proof chain for Theorem 1, the paper's main result. The chain is: (1) the Dyson expansion from [DS25] (Desio–Seiringer, independent authors) yields the point process representation (Proposition 4); (2) the product structure of F_T under disjoint interval clusters yields the renewal transform identification (Proposition 5), which is proved in the paper from scratch using Proposition 4; (3) Corollary 8 expresses the partition function ratio as an expectation of e^{-δ σ²_T} under Θ̂_{P₀,T}, proved directly from Propositions 4 and 7; (4) inequality (29) — D̂_T ≤ σ²_T — is a purely Gaussian statement (equation 28, a Pythagorean identity for L² projections) independent of the polaron; (5) Proposition 2 identifies the resulting expectation with the rank-one perturbation ratio, proved in the paper via Dyson series and a uniqueness lemma (Lemma 3); (6) the Sherman–Morrison formula and the support constraint on the spectral measure of H(P₀) yield the explicit bound. The bound on s(λ) uses only that the spectral measure is supported on {E(P₀)} ∪ [E(P₀)+Δ(P₀), ∞), which follows from the definitions of ρ and Δ. The self-citation to [HP25] (Hinrichs–Polzer) for the renewal transform (Theorem 4) is load-bearing but not circular: it is a general theorem about Laplace transforms of probability measures on ℝ, making no assumptions about polaron models or energy-momentum relations. The self-citation to [Pol23] provides context but is not load-bearing for the proofs here. No step reduces to its inputs by construction; no parameter is fitted and renamed as a prediction. The score of 1 reflects the presence of a load-bearing self-citation ([HP25]) that is a general, independently checkable mathematical tool rather than a circular dependency.
Assumptions & free parameters
assumptions (5)
- domain assumption Assumption 1: ω and v are radially symmetric; |k|² ↦ |v(k)|² is completely monotone with v ≢ 0 and lim_{k→∞} v(k) = 0; |k|² ↦ ω(k) is a strictly positive Bernstein function; integrability condition (2) holds.
- standard math The renewal transform of a probability measure μ exists and is unique (Theorem 4, citing [HP25, Theorem 4.1]).
- standard math The Dyson-type expansion (24) from [DS25, Theorem 1] holds for ⟨Ω, e^{-TH(P)}Ω⟩ under Assumption 1.
- domain assumption The semigroup e^{-tH(P)} is positivity improving, so E(P) = -lim_{T→∞} (1/T) log⟨Ω, e^{-TH(P)}Ω⟩ and ground states are unique with ⟨Ω, ψ⟩ > 0.
- domain assumption Right-continuity of ρ at P (needed for Theorem 3(2) to give equality rather than inequality).
Cite this review
Pith. "Pith review of Lower bound on the energy-momentum relation of the polaron." pith.science (2026). https://pith.science/paper/NRZ2IBSZ
@misc{pith2026260705286,
author = {Pith},
title = {Pith review of: Lower bound on the energy-momentum relation of the polaron},
year = {2026},
howpublished = {\url{https://pith.science/paper/NRZ2IBSZ}},
note = {Machine review of arXiv:2607.05286}
}
read the original abstract
For a class of polaron-type models, we establish a lower bound on the energy-momentum relation in terms of the vacuum overlap and the spectral gap of the total momentum zero Hamiltonian. We show convergence of the rescaled mean square displacement of the associated path measure to the inverse of the effective mass. We derive a probabilistic criterion for the absence of ground states at large total momentum.
Reference graph
Works this paper leans on
-
[1]
Effective Mass of the Fr\"ohlich Polaron and the Landau-Pekar-Spohn Conjecture
R. Bazaes, C. Mukherjee, M. Sellke, and S. R. S. Varadhan. Effective mass of the Fr\"ohlich Polaron and the Landau-Pekar-Spohn conjecture. Preprint, 2023. arXiv:2307.13058 http://arxiv.org/abs/2307.13058
work page Pith review arXiv 2023
-
[2]
V. Betz and S. Polzer. A Functional Central Limit Theorem for Polaron Path Measures. Commun. Pure Appl. Math. , 75(11):2345--2392, 2022. doi:10.1002/cpa.22080
-
[3]
V. Betz and S. Polzer. Effective Mass of the Polaron: A Lower Bound. Commun. Math. Phys. , 399(1):173--188, 2023. doi:10.1007/s00220-022-04553-0
-
[4]
V. Betz, T. Schmidt and M. Sellke. Mean square displacement of Brownian paths perturbed by bounded pair potentials. Electron. J. Probab. , 30: 1-17, 2025. doi:10.1214/24-EJP1263
- [5]
-
[6]
V. Betz and H. Spohn. A central limit theorem for Gibbs measures relative to Brownian motion. Probab. Theory Relat. Fields , 131(3):459–478, 2005. doi:10.1007/s00440-004-0381-8
-
[7]
J. Dahlb k. Spectral Analysis of Large Particle Systems . PhD thesis, Aarhus University, 2017
work page 2017
-
[8]
D. Desio and R. Seiringer. Dyson expansion for form-bounded perturbations, and applications to the polaron problem. Preprint, 2025. arXiv:2512.13443 http://arxiv.org/abs/2512.13443
work page Pith review arXiv 2025
Show all 28 references
-
[9]
W. F. Donoghue. Monotone Matrix Functions and Analytic Continuation. Springer, 1974
1974
-
[10]
Dybalski and H
W. Dybalski and H. Spohn. Effective mass of the polaron revisited. Ann. Henri Poincar\'e , 21(5):1573--1594, 2020. doi:10.1007/s00023-020-00892-7
2020 doi
-
[11]
R. Feynman. Statistical Mechanics. A Set of Lectures (Frontiers in Physics). The Benjamin/Cummings publishing company, Inc., Reading, Massachusetts, 1972
1972
-
[12]
Gubinelli
M. Gubinelli. Gibbs Measures for Self-Interacting Wiener Paths. Markov Process. Relat. Fields , 12:747–766, 2006
2006
-
[13]
Hinrichs and S
B. Hinrichs and S. Polzer. Wiener-Type Theorems for the Laplace Transform. With Applications to Ground State Problems. Preprint, 2025. arXiv:2511.02867 http://arxiv.org/abs/2511.02867
2025
-
[14]
T. Kato. Perturbation Theory for Linear Operators , volume 132 of Grundlehren der mathematischen Wissenschaften . Springer-Verlag, Berlin, Heidelberg, 2nd edition, 1980
1980
-
[15]
Keller, D
M. Keller, D. Lenz, H. Vogt, and R. K. Wojciechowski. Note on basic features of large time behaviour of heat kernels. J. Reine Angew. Math. , 708:73--95, 2015. doi:10.1515/crelle-2013-0070
2015 doi
-
[16]
Lampart, D
J. Lampart, D. Mitrouskas and K. My\' s liwy. On the Global Minimum of the Energy--Momentum Relation for the Polaron. J. Math. Phys. Anal. Geom. , 26(17), 2023. doi:10.1007/s11040-023-09460-x
2023 doi
-
[17]
Mitrouskas and R
D. Mitrouskas and R. Seiringer. Ubiquity of bound states for the strongly coupled polaron. Pure Appl. Anal. , 5(4):973--1008, 2023. doi:10.2140/paa.2023.5.973
2023 doi
-
[18]
T. Miyao. Nondegeneracy of ground states in nonrelativistic quantum field theory. J. Operator Theory , 64(1):207--241, 2010
2010
-
[19]
T. Miyao. Monotonicity of the polaron energy II: General theory of operator monotonicity. J. Stat. Phys. , 153(1):70--92, 2013. doi:10.1007/s10955-013-0812-y
2013 doi
-
[20]
J. S. Møller. The Polaron revisited . Rev. Math. Phys. , 18(5):485--517, 2006. doi:10.1142/S0129055X0600267X
2006 doi
-
[21]
Mukherjee
C. Mukherjee. Central limit theorem for Gibbs measures on path spaces including long range and singular interactions and homogenization of the stochastic heat equation. Ann. Appl. Probab. , 32(3): 2028--2062, 2022. doi:10.1214/21-AAP1727
-
[22]
Mukherjee and S
C. Mukherjee and S. R. S. Varadhan. Identification of the Polaron Measure I : Fixed Coupling Regime and the Central Limit Theorem for Large Times. Commun. Pure Appl. Math. , 73(2):350--383, 2020. doi:10.1002/cpa.21858
2020 doi
-
[23]
Mukherjee and S
C. Mukherjee and S. R. S. Varadhan. Corrigendum and Addendum: Identification of the Polaron Measure I: Fixed Coupling Regime and the Central Limit Theorem for Large Times. Commun. Pure Appl. Math. , 75(7):1642-1653, 2022. doi:10.1002/cpa.22052
2022 doi
-
[24]
S. Polzer. Renewal approach for the energy--momentum relation of the F r\" o hlich polaron. Lett. Math. Phys. , 113(4):90, 2023, doi:10.1007/s11005-023-01711-w
2023 doi
-
[25]
Seiringer
R. Seiringer. Absence of excited eigenvalues for Fr\"ohlich type polaron models at weak coupling. J. Spectr. Theory , 13(3):1045--1055, 2023. doi:10.4171/JST/469
2023 doi
-
[26]
M. Sellke. Almost quartic lower bound for the Fr\"ohlich polaron’s effective mass via Gaussian domination. Duke Math. J. 173(13): 2687--2727, 2024. doi:10.1215/00127094-2023-0064
2024 doi
-
[27]
H. Spohn. Effective mass of the polaron: A functional integral approach. Ann. Phys. , 175(2):278--318, May 1987. doi:10.1016/0003-4916(87)90211-9
1987 doi
-
[28]
H. Spohn. The polaron at large total momentum. J. Phys. A: Math. Gen. , 21(5):1199--1211, Mar. 1988. doi:10.1088/0305-4470/21/5/016
1988 doi
Reviewed July 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.