Orbital stability of solitary waves for the generalized Choquard model

George Venkov; Mirko Tarulli; Vladimir Georgiev

REVIEW 2 major objections 4 minor 12 references

Orbital stability of solitary waves for the generalized Choquard model

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Choquard ground states are locally unique for 2 ≤ p < 7/3

desk verdict A clever analytic-continuation proof of local uniqueness for generalized Choquard minimizers, but the central decay estimate is under-proved and must be fixed before the theorem is solid. read the letter →

arxiv 1908.08106 v2 pith:YPY7B37D submitted 2019-08-21 math.AP

Vladimir Georgiev , Mirko Tarulli , George Venkov This is my paper · ORCID

classification math.AP MSC 37K4035Q5535Q51

keywords generalizedChoquardequationlocaluniquenessgroundstatesorbitalstabilityRieszpotentialGagliardo-Nirenberginequalityradialminimizersanalyticcontinuation

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper tries to prove that, for the generalized Choquard equation in three dimensions, ground states with fixed $L^2$ mass are locally unique for $2 \le p < 7/3$: any two radial positive minimizers that are close in $H^1_{\mathrm{rad}}$ must be identical. The point is that local uniqueness is the missing ingredient for orbital stability of the standing waves. The proof avoids the classical Sturm oscillation and shooting arguments, which break down because the nonlinearity is nonlocal, and instead studies an analytic continuation of the energy along directions in the kernel of the linearized operator. If Theorem 1 holds, orbital stability in the range $p \in (5/3, 7/3)$ follows from existing stability machinery, because the relevant ground-state branch cannot split.

What carries the argument

The load-bearing object is the analytic continuation $K(z) = E_p\!\left(\sqrt{\sigma}\, (Q + z h)/\|Q + z h\|_{L^2}\right)$, where $h$ is a nontrivial element of the kernel of $L_+ = -\Delta + \omega - p I(Q^{p-1}\,\cdot)Q^{p-1} - (p-1) I(Q^p)Q^{p-2}$. The analyticity domain is built with the wedge $\Lambda_\delta$ near the diagonal $\operatorname{Re} z = \operatorname{Im} z$, using Proposition 4.1: every nonzero kernel element $h$ has $|h(r)| \lesssim e^{-r}/r$, so the quotient $h/Q$ is bounded and the $p$-th power $(1 + z h/Q)^p$ can be complex-analytically continued without crossing the branch cut of the logarithm. Since the no-branching assumption makes all derivatives of $K$ vanish at $z = 0$, $K$ is constant; letting $z = R + iR$ and using dominated convergence identifies $E_p(\sqrt{\sigma} h) = E_p(Q)$, which contradicts the radial ODE lemma. The kernel-dimension bound $\dim(\ker L_+) \le 2$, proved by a Fuchs–Painlevé series expansion, turns the pointwise directional argument into a uniform $\varepsilon_0$.

What would settle it

Follow the unique radial positive ground state at $p = 2$ numerically and continue it in $p$ up to $7/3$: if a second radial positive normalized minimizer branches off at some $p \in [2, 7/3)$, Theorem 1 is false. A cheaper test is to solve the linearized equation $L_+ h = 0$ radially and check whether a nontrivial $h$ obeys $|h(r)| \lesssim e^{-r}/r$ with a nonzero first coefficient; a numerical counterexample to that decay would invalidate Proposition 4.1 and with it the proof.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central discovery is Theorem 1: under $2 \le p < 7/3$, there is $\varepsilon > 0$ such that any two radial positive minimizers $Q_1, Q_2 \in H^1_{\mathrm{rad}}$ of the constrained energy $E_\sigma$ with $\|Q_1 - Q_2\|_{H^1_{\mathrm{rad}}} \le \varepsilon$ are equal. The proof proceeds by contradiction: if two minimizers approached each other, their difference would converge to a nonzero element $h$ of the kernel of the linearized operator $L_+$; the authors then define $K(z)$, an analytic continuation of the energy along the normalized family $(Q + z h)/\|Q + z h\|_{L^2}$, show $K(z)$ is constant in a domain of analyticity, and take the limit along the diagonal $z = R + iR$ to reach $E_p(\sqrt{\sigma} h) = E_p(Q)$. Orthogonality of $h$ to $Q$ then gives a zero of $h$ at some radius, and a simple ODE lemma forces $h \equiv 0$, a contradiction. The paper also proves Theorem 2, an equivalence between minimizers of the Choquard energy and of the associated Gagliardo–Nirenberg functional.

Load-bearing premise

Proposition 4.1 claims that every nonzero kernel element $h$ decays like $e^{-r}/r$ and is nonzero for all large radii; the analytic continuation of $K(z)$ and the diagonal limit both lean on the bound $h/Q \le C$, so if that decay estimate fails the contradiction in Theorem 1 collapses.

Editorial extensions

If this is right

If Theorem 1 is correct, each radial positive ground state of the generalized Choquard equation in the range $2 \le p < 7/3$ is an isolated point among radial minimizers of fixed $L^2$ mass.
Combined with the linearized stability classification, this supplies the nonlinear orbital stability of the corresponding standing waves for $p \in (5/3, 7/3)$: no nearby distinct minimizer can act as a competing orbit.
Theorem 2 implies that, for $p \in (5/3, 7/3)$, the minimizers of the Choquard energy and the sharp Gagliardo–Nirenberg functional coincide, so the profile of the optimizer is characterized independently of which variational problem one solves.
The analytic-continuation argument only needs decay of kernel elements, not spectral nondegeneracy, so it tolerates a kernel of dimension up to 2; this is a new route to local uniqueness for nonlocal equations.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The same analytic-continuation device might transfer to other nonlocal equations with different kernels, as long as the kernel elements have enough decay to bound $h/Q$.
The theorem concerns radial positive minimizers with $2 \le p < 7/3$; a quantitative version of Proposition 4.1, with constants depending only on $p$ and $\sigma$, would turn the qualitative $\varepsilon$ into an explicit radius of uniqueness, but the paper does not do this.
Theorem 2 together with Corollary 3.1 gives a formula for the sharp Gagliardo–Nirenberg constant in terms of the ground-state energy; if accurate numerical values of $E_\sigma$ become available, they would provide an independent test of the uniqueness claim.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: the paper takes a genuinely open step—local uniqueness of radial ground states for the generalized Choquard equation with p>2—and the main idea is good. The analytic-continuation trick, extending K(z) to a domain where it must be constant and then pushing z to infinity along R+iR, is a real departure from the usual Sturm/shooting approach, and it does not require the kernel of L+ to be trivial. The equivalence between the two variational problems (Theorem 2) and the formula for the Gagliardo-Nirenberg constant are clean and useful. This deserves a serious referee.

The soft spots are concentrated in Section 4. Proposition 4.1 is load-bearing: the bound |h|/Q ≤ C is what justifies analyticity of D((Q+zh)^p) and the limit z=R+iR. The proof as written is not complete. In the c1=0 case, the maximum-principle argument assumes h has two zeros r10 only guarantee at least one zero; an eventually sign-definite h tending to zero with a single zero is compatible with the assumptions, and the stated contradiction does not apply. The subsequent conclusion c1=d1=0 also imports Lemma 4.1 from [4] without statement or proof, and the asymptotic expansions (4.6)–(4.8) are asserted rather than derived. If c1=0 and d1≠0 is possible, the ratio |h|/Q might still be bounded, but the proof as written does not establish that; if h decayed like e^{-(p-2)r}, the argument in Theorem 1 collapses. This is a gap, not a minor omission.

Smaller issues: equation (1.15) has an exponent typo—it should be β^{1-γ} on the right, not β^{γ-1}; the derivation in the proof uses the corrected version. The boundary with your own paper [4] is not drawn explicitly; the title of [4] suggests a uniqueness result that may overlap with Theorem 1, so the authors should state exactly what is new. Lemma 5.2 on dim(Ker L+) is plausible but only sketched; a few lines showing how Fuchs–Painleve local solvability yields a global dimension bound would help.

None of this changes my view that the result is probably true and the method is publishable. The right reader is someone working on uniqueness and stability for Choquard-type equations. Send it to a serious referee, but the referee should be told to demand a complete proof of Proposition 4.1 and a fixed (1.15).

Referee Report

2 major / 4 minor

Summary. The manuscript studies the generalized Choquard equation i∂tu + Δu + I(|u|p)|u|p−2u = 0 in R3 and, in particular, the local uniqueness of radial positive minimizers of the constrained energy Ep(u)=12‖∇u‖L22−12pD(|u|p,|u|p) under ‖u‖L22=σ. Theorem 1 asserts that, for 2≤p<7/3, any two such minimizers that are sufficiently close in H1rad must coincide. The proof strategy is to reduce local uniqueness to directional uniqueness along elements h of the kernel of the linearized operator L+; if a direction h gives a flat expansion of the normalized energy, the function K(z) obtained by replacing ε by a complex parameter z is shown, via analytic continuation on a domain Ωδ, to be constant, and taking z→∞ along the diagonal yields Ep(√σh)=Ep(Q), making √σh a minimizer. A zero of h coming from orthogonality h⊥Q then contradicts Lemma 5.1. Theorem 2 characterizes minimizers of Eσ as exactly those functions attaining equality in the Gagliardo–Nirenberg inequality with best constant C∗. The paper also contains auxiliary ODE lemmas and Fuchs–Painleve series arguments.

Significance. If the proof can be completed, this is a valuable contribution: it gives a local uniqueness statement for a nonlocal Choquard model in a range of p where classical Sturm comparison arguments are not available, and it does so without requiring non-degeneracy of the linearized operator L+. The analytic-continuation mechanism is quite elegant, and the link established in Theorem 2 between constrained minimizers and equality cases of the Gagliardo–Nirenberg inequality is useful in its own right. However, the central uniqueness theorem is currently not rigorously established, because the key decay estimate for elements of the kernel of L+ (Proposition 4.1) has a substantive gap in its proof. Since that estimate is used to justify the analytic continuation and the limit along z=R+iR, the main claim is not yet proven as written.

major comments (2)

[Section 4, Proposition 4.1] The proof of the case c1=0, d1≠0 is incomplete. The argument assumes that h(r)=0 has two roots r2>r1>r0 and then applies the maximum principle to the interval [r1,r2]. However, orthogonality h⊥Q with Q>0 only guarantees that h has at least one zero; h may have exactly one zero and, for example, be negative near the origin and positive on the tail. In that situation no interval with two boundary zeros exists, and the maximum principle applied to equation (4.9) gives no contradiction. This step is what forces c1=d1=0 and hence the decay estimate |h(r)|≲e−r/r in (4.5). That estimate is used in Section 2 to justify the bound |h|/Q≤C, the analytic continuation of K(z), and the limit along z=R+iR. If h instead decays like e−(p−1)r, then |h|/Q grows like e−(p−2)r and the whole argument in Theorem 1 collapses. A complete treatment of the single-zero and eventually-positive case is therefore required.
[Section 4, equations (4.6)–(4.11)] Even apart from the two-zero issue, the reduction from the system (4.1) to the integral inequalities used with Lemma 4.1 is only sketched. The asymptotic expansions (4.6)–(4.8) are asserted as 'verified in a similar way' without a derivation, and the boundedness of ψ=|g|+|B| as well as the exact form of the inequality (4.11) are not checked. For p>2 the exponential factors e−(p−2)s and e−ps make the reduction plausible, but for p=2, which is included in the statement of Theorem 1, the factor e−(p−2)s is identically 1 and the displayed estimates do not lead to (4.11). The manuscript should either give a complete proof of Proposition 4.1 or explicitly restrict the proof to p>2 and quote the classical p=2 uniqueness result from the introduction.

minor comments (4)

[Section 2, proof of Theorem 1] The final compactness step, where a sequence hk in the kernel of L+ is replaced by a limit h∗ with ‖h∗‖L2=1, is not stated explicitly; since the kernel is finite-dimensional (Lemma 5.2), this step is valid but should be spelled out.
[Section 5, Lemma 5.1] The proof of Lemma 5.1 only discusses the case u′(r0)<0 and omits the symmetric case u′(r0)>0; in the case u′(r0)=0 it invokes uniqueness for the Cauchy problem without stating the required regularity or Lipschitz condition on V(|u|)u. This is likely easy to repair, but the lemma is used for the final contradiction in Theorem 1 and should be proved completely.
[Section 3, Theorem 2] The proof of Theorem 2 assumes the existence of a minimizer v for the functional Fσ in (1.13). Since Fσ has infimum 0 and the existence of extremals for the Gagliardo–Nirenberg inequality is nontrivial, the paper should provide a reference or a short argument for this existence.
[Title and abstract] The title and abstract promise orbital stability of solitary waves, but the paper actually proves local uniqueness of minimizers and a characterization of best constants; the connection to orbital stability is only via the classification in [3]. The wording should be adjusted to match the content of the results.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found; the uniqueness argument is self-contained apart from standard asymptotics and an auxiliary integral lemma.

full rationale

The central claim, Theorem 1, is a local uniqueness statement proved by contradiction from the variational problem (1.3). The proof does not assume its conclusion: from a hypothetical second minimizer it derives a nontrivial kernel element h of L_+, then shows that such an h cannot exist using analytic continuation of K(z). The analytic continuation uses the estimate |h|/Q ≤ C from Proposition 4.1, whose proof cites asymptotic expansions from Strauss [11] and Lemma 4.1 from the authors' earlier paper [4]. This self-citation is not load-bearing circularity: Lemma 4.1 is a parameter-free integral inequality whose stated assumptions do not include the target uniqueness result, and the asymptotic expansions are external standard results. No fitted value is renamed as a prediction, and no ansatz is smuggled in via citation. The proof does contain a nontrivial gap in the c1 = 0 case of Proposition 4.1, where the maximum-principle step assumes two zeros of h without proof, but that is a correctness risk rather than a circularity. Accordingly, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

No free parameters are fitted. The argument introduces no new physical entities; the only imported objects are the ground state Q, the Riesz potential, and the sharp constant C*, all standard.

assumptions (5)

domain assumption Existence of radial positive minimizers Q of (1.3) for p in (5/3, 7/3) is imported from [9] and [10].
Theorem 1 concerns minimizers; existence is not reproved.
domain assumption Asymptotic expansions (4.3) and (4.4) for Q and A are imported from Strauss [11].
These decay rates are used in Proposition 4.1 to control h/Q.
domain assumption Lemma 4.1 of [4], an integral inequality, is used without proof in Proposition 4.1.
Self-cited auxiliary lemma; parameter-free and not the target theorem, but unverified here.
standard math Fuchs-Painleve theorem from Hille [5] provides real analytic solutions near the singular point r = 0.
Used in Lemma 5.2 and the appendix to control the dimension of the kernel.
domain assumption The linearized operator L+ has kernel elements satisfying the decay and non-vanishing properties of Proposition 4.1.
This is a proposition of the paper, but its proof rests on imported asymptotics and on [4].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Orbital stability of solitary waves for the generalized Choquard model." pith.science (2026). https://pith.science/paper/YPY7B37D

@misc{pith2026190808106,
  author       = {Pith},
  title        = {Pith review of: Orbital stability of solitary waves for the generalized Choquard model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YPY7B37D}},
  note         = {Machine review of arXiv:1908.08106}
}

abstract

We consider the generalized Choquard equation describing trapped electron gas in 3 dimensional case. The study of orbital stability of the energy minimizers (known as ground states) depends essentially in the local uniqueness of these minimizers. In equivalent way one can optimize the Gagliardo--Nirenberg inequality subject to the constraint fixing the $L^2$ norm. The uniqueness of the minimizers for the case $p=2$, i.e. for the case of Hartree--Choquard is well known. The main difficulty for the case $p > 2$ is connected with possible lack of control on the $L^p$ norm of the minimizers.

Figures

Figures reproduced from arXiv: 1908.08106 by the authors.

**Figure 1.** The domain of analyticity of K(z), here depicted by the shaded region Ωδ. Choosing z(R) = R + iR with R → ∞, we can use the relation 1 + z(R)h(|x|)/Q(|x|) p σ + z(R) 2 → h(|x|) Q(|x|) , combined with Lebesgue dominated convergence theorem to conclude that lim R→∞ K(z(R)) = Ep( √ σ h). The relation Ep( √ σ h) = K(0) = Ep(Q) [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

12 extracted references · 12 canonical work pages

[4]

Georgiev, M

V. Georgiev, M. Tarulli and G. Venkov, Existence and Uniqueness of Ground States for p-Choquard Model, Nonlinear Analysis 179 (2019), 131–145

work page 2019
[1]

Chang, S

S.M. Chang, S. Gustafson, K. Nakanishi and T. P. Tsai, Spectra of linearized operators for NLS solitary waves, SIAM Journal on Mathematical Analysis , 39(4) (2007), 1070–1111

work page 2007
[2]

Genev and G

H. Genev and G. Venkov, Soliton and blow-up solutions to the time- dependent Schr¨ odinger-Hartree equation, Discrete Contin. Dyn. Syst. Ser. S , 5 no. 5 (2012), 903 – 923

work page 2012
[3]

Georgiev and A

V. Georgiev and A. Stefanov, On the classiﬁcation of the spectr ally stable stand- ing waves of the Hartree problem, Physica D: Nonlinear Phenomena, Vol. 370, (2018), 29–39

work page 2018
[5]

Hille, Ordinary diﬀerential equations in the complex domain, repr int of the 1976 original

E. Hille, Ordinary diﬀerential equations in the complex domain, repr int of the 1976 original. Dover Publications, Inc., Mineola, NY, 1997 18 VLADIMIR GEORGIEV, MIRKO TARULLI, AND GEORGE VENKOV

work page 1976
[6]

Kwong, Uniqueness of positive solutions of ∆ u − u + up = 0, Arch

M.K. Kwong, Uniqueness of positive solutions of ∆ u − u + up = 0, Arch. Rational Mech. Anal. Vol. 105, no. 3 (1989), 243 – 266

work page 1989
[7]

Lieb, Existence and uniqueness of the minimizing solution of Cho quard’s nonlinear equation, Stud

E.H. Lieb, Existence and uniqueness of the minimizing solution of Cho quard’s nonlinear equation, Stud. Appl. Math. 57 (2) (1976/1977), 93–105

work page 1976
[8]

McLeod and J

K. McLeod and J. Serrin, Uniqueness of positive radial solutions o f ∆ u + f (u) = 0 in RN , Archive for Rational Mechanics and Analysis , 99(2) (1987), 115–145

work page 1987

Show all 12 references

[9]

Moroz and J

V. Moroz and J. Van Schaftingen, Groundstates of nonlinear Ch oquard equa- tions: existence, qualitative properties and decay asymptotics, J. Funct. Anal. 265, no. 2 (2013), 153–184

2013
[10]

Moroz and J

V. Moroz and J. Van Schaftingen, Existence of ground states for a class of nonlinear Choquard equations. Trans. Amer. Math. Soc. 367, no. 9 (2015), 6557–6579

2015
[11]

W. Strauss. Existence of Solitary Waves in Higher Dimension, Comm. Math. Physics, Vol. 55 (1977), 149 – 162

1977
[12]

Tarulli and G

M. Tarulli and G. Venkov, Decay and Scattering in energy space for the solu- tion of weakly coupled Choquard and Hartree-Fock equations, e-p rint (2019): arXiv:1904.10364. V. Georgiev, Dipartimento di Matematica, Universit `a di Pisa Largo B. Pontecorvo 5, 56127 Pisa, Italy, a...

2019 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [4]

Georgiev, M

V. Georgiev, M. Tarulli and G. Venkov, Existence and Uniqueness of Ground States for p-Choquard Model, Nonlinear Analysis 179 (2019), 131–145

work page 2019

[2] [1]

Chang, S

S.M. Chang, S. Gustafson, K. Nakanishi and T. P. Tsai, Spectra of linearized operators for NLS solitary waves, SIAM Journal on Mathematical Analysis , 39(4) (2007), 1070–1111

work page 2007

[3] [2]

Genev and G

H. Genev and G. Venkov, Soliton and blow-up solutions to the time- dependent Schr¨ odinger-Hartree equation, Discrete Contin. Dyn. Syst. Ser. S , 5 no. 5 (2012), 903 – 923

work page 2012

[4] [3]

Georgiev and A

V. Georgiev and A. Stefanov, On the classiﬁcation of the spectr ally stable stand- ing waves of the Hartree problem, Physica D: Nonlinear Phenomena, Vol. 370, (2018), 29–39

work page 2018

[5] [5]

Hille, Ordinary diﬀerential equations in the complex domain, repr int of the 1976 original

E. Hille, Ordinary diﬀerential equations in the complex domain, repr int of the 1976 original. Dover Publications, Inc., Mineola, NY, 1997 18 VLADIMIR GEORGIEV, MIRKO TARULLI, AND GEORGE VENKOV

work page 1976

[6] [6]

Kwong, Uniqueness of positive solutions of ∆ u − u + up = 0, Arch

M.K. Kwong, Uniqueness of positive solutions of ∆ u − u + up = 0, Arch. Rational Mech. Anal. Vol. 105, no. 3 (1989), 243 – 266

work page 1989

[7] [7]

Lieb, Existence and uniqueness of the minimizing solution of Cho quard’s nonlinear equation, Stud

E.H. Lieb, Existence and uniqueness of the minimizing solution of Cho quard’s nonlinear equation, Stud. Appl. Math. 57 (2) (1976/1977), 93–105

work page 1976

[8] [8]

McLeod and J

K. McLeod and J. Serrin, Uniqueness of positive radial solutions o f ∆ u + f (u) = 0 in RN , Archive for Rational Mechanics and Analysis , 99(2) (1987), 115–145

work page 1987