{"id":"473c688a-0259-414b-b8e2-764db31bfab2","arxiv_id":"1908.01209","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"When the low-frequency Besov norms of the initial data are bounded but not necessarily small, the density and velocity of global strong Navier-Stokes-Poisson solutions decay at the optimal rates predicted by the linearized system.","lead":"This paper proves time-decay rates for the density and velocity of the compressible Navier-Stokes-Poisson equations in critical Besov spaces, using a pure energy argument instead of Fourier spectral analysis. The improvement is that the low-frequency part of the initial data needs to be bounded, not small, for these decay rates to hold.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation (4.5), cited from [31], is a load-bearing product estimate whose endpoint s1=s0 reaches zero-sum regularity; if it fails, Lemma 4.1 and Theorem 1.2 as stated fail.","rationale":"The reader's specific concern about inequality (4.6) appears answerable: for 2≤p≤d, B^{d/p-1}_{p,1} embeds in L^p, the product lies in L^{p/2}, and the low-frequency Sobolev embedding L^{p/2}→ℓ B^{-s0}_{2,∞} gives the second estimate in (4.6) with s0=2d/p-d/2; the first estimate is valid on the ℓ-restricted norm up to a constant depending only on the fixed cutoff j0. Thus (4.6) is not the main hidden assumption. The more delicate point is (4.5), used in Lemma 4.1 to bound g4(a,u^ℓ). At the allowed endpoint s1=s0, the sum of the two regularities on the right side is exactly zero, which is the critical case for the paraproduct remainder; standard product laws require positive sum. Since (4.5) is only cited from the preprint [31] and not proved or checked in this text, the uniform bound (4.12) and hence the Lyapunov closure (5.5) rest on an unverified endpoint estimate. The rest of the paper—energy estimates (3.3) and (3.9), the interpolation (5.9)-(5.10), and the Gronwall argument—is internally coherent, and no contradiction was found in those steps. Therefore the reader's CONDITIONAL verdict is preserved, but the specific load-bearing risk should be relocated from (4.6) to (4.5).","tokens_in":21477,"tokens_out":51743,"duration_ms":462343,"concrete_test":"Independently derive (4.5) from Bony's paraproduct decomposition for the endpoint case s1=s0, with σ=d/p-d/2-s1. Isolate the remainder term R(F,G) and check whether it is bounded in B^{-d/p}_{2,∞} when the sum of the two regularities is zero; a positive result must supply an explicit paraproduct estimate, while a negative result (e.g., a wave-packet example with ||FG||_{B^{-d/p}_{2,∞}} growing while the right side stays bounded) would force the range in Theorem 1.2 to be narrowed to s1<s0.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Lemma 4.1, which produces the uniform negative-Besov bound (4.12) needed for the Lyapunov closure (5.5), depends on the non-classical product estimate (4.5) to control the low-frequency part of g4(a,u^ℓ) in the case p≥2. For the endpoint s1=s0 (which the statement of Theorem 1.2 explicitly allows), (4.5) reads ||FG||_{B^{-d/p}_{2,∞}} ≲ ||F||_{B^{d/p-1}_{p,1}} ||G||_{B^{1-d/p}_{2,∞}}. The two regularities on the right sum to exactly 0, the borderline case for Bony's paraproduct: the remainder R(F,G) is usually only controlled when the sum is positive. The text does not prove (4.5); it cites [31], and (4.5) is not a consequence of the classical estimates in Proposition 2.4 or Proposition 2.5. Since s1=s0 is included in the theorem's range and is used in the interpolation step (5.9)-(5.10), a failure of (4.5) at the endpoint would invalidate Lemma 4.1 and hence break the proof of Theorem 1.2. The issue may be repairable by restricting to s1<s0 or by proving (4.5) when G is low-frequency, but as written the endpoint is asserted without the needed argument.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies the large-time decay of global strong solutions to the compressible Navier-Stokes-Poisson system in critical Besov spaces. The main result, Theorem 1.2, asserts that if the low-frequency parts of the initial density and velocity have bounded norms in B^{-s1-1}_{2,\\infty} and B^{-s1}_{2,\\infty} for 1-d/2 < s1 \\leq s0, then the B^s_{p,1} norms of the density and velocity decay respectively at the rates (1+t)^{-d/2(1/2-1/p)-(s1+s+1)/2} and (1+t)^{-d/2(1/2-1/p)-(s1+s)/2}. The proof avoids spectral analysis and is built instead on a Lyapunov-type inequality obtained by pure energy methods. The key novelty is a Gronwall argument, Lemma 4.1, which gives a uniform bound on the negative low-frequency Besov norm of the solution; this bound replaces the usual smallness assumption on the low frequencies of the initial data.","tokens_in":1682,"tokens_out":1681,"duration_ms":114194,"significance":"If the main theorem is correct, the paper is a meaningful advance: it removes the low-frequency smallness assumption in the Lp critical framework, covers the oscillatory case p > d in dimensions two and three, and gives density decay one half-power faster than velocity, reflecting the effect of the Poisson potential. The Lyapunov closure in Section 5 is elegant, and the interpolation steps leading to (5.11) and (5.12) are algebraically coherent. The main weakness is that the decisive uniform bound (4.12) rests on non-classical product estimates, in particular (4.5) and (4.6), which are quoted from other papers rather than proved here; the endpoint s1 = s0, which is explicitly allowed in Theorem 1.2, is the delicate case for these estimates.","major_comments":[{"comment":"The estimate ||FG||_{B^{d/p-d/2-s1}_{2,\\infty}} \\lesssim ||F||_{B^{d/p-1}_{p,1}} ||G||_{B^{d/p-d/2-s1+1}_{2,\\infty}} is cited from [31] and is used to control the low-frequency term g4(a,u^\\ell), an unavoidable contribution in Lemma 4.1. The theorem's range 1-d/2 < s1 \\leq s0 includes the endpoint s1 = s0, for which the two regularities on the right are d/p-1 and 1-d/p and therefore sum to exactly zero. This is the borderline case for Bony's paraproduct remainder, and the manuscript supplies no proof that the estimate remains valid there. Since (4.12) and hence the Lyapunov inequality (5.5) depend on this estimate, the proof of Theorem 1.2 is incomplete at the advertised endpoint. The author should either prove (4.5) for the full range, especially at s1 = s0, or restrict the statement of Theorem 1.2 to s1 < s0.","section":"Section 4, Eq. (4.5)"},{"comment":"The inequality ||FG^h||_{B^{-s1}_{2,\\infty}} \\lesssim ||F||_{B^{d/p-1}_{p,1}} ||G^h||_{B^{d/p-1}_{p,1}} for 2 \\leq p \\leq d is quoted from the author's own prior paper [26] and is used to control all the high-frequency nonlinear terms in the case 2 \\leq p \\leq d, including \\Lambda^{-1}\\mathrm{div}(a u^h), k(a)\\nabla a^h, g3(a,u^h) and g4(a,u^h). The present text gives neither a proof nor a precise statement of the hypotheses under which (4.6) is valid. This is load-bearing because it is exactly the closure of Lemma 4.1 that removes the low-frequency smallness assumption. The author should supply a self-contained proof of (4.6), or at least a complete statement with all hypotheses, and should verify in particular that the endpoint s1 = s0 is covered.","section":"Section 4, Eq. (4.6)"},{"comment":"The passage from (4.1) to the uniform bound (4.12) invokes 'nonlinear generalisations of the Gronwall inequality' from page 360 of [18] without stating the version used. This is a minor presentation issue in itself, but it becomes more serious because D2_p in (4.1) is only known to be integrable by using the smallness of Ep,0 from Theorem 1.1; the constants and the precise condition on the data should be made explicit so that the reader can verify that (4.12) indeed follows. This is not a fatal objection, but it should be repaired in the revision.","section":"Section 4, Eq. (4.12)"}],"minor_comments":[{"comment":"The displayed definition of \\|z\\|_{\\ell \\dot{B}^s_{2,1}} uses an L^p norm, while the surrounding argument treats it as an L^2 low-frequency norm; the exponent in the Lebesgue norm should be corrected to L^2 for consistency with the rest of the paper.","section":"Section 5, after Eq. (5.1)"},{"comment":"There are several typographical errors, including 'A NA VIER-STOKES-POISSON' in the title and 'NavierStokesequations' after Eq. (1.1); these should be corrected.","section":"Title and Introduction"},{"comment":"The word 'optimal' is used for the decay rates, but only upper bounds are proved; if no matching lower bounds are established, the wording should be softened to 'rates matching the expected optimal rates' or a precise notion of optimality should be stated.","section":"Remark 1.2"},{"comment":"The estimates (4.4)-(4.5) are quoted from the arXiv preprint [31]; since these estimates are load-bearing, the author should ensure that the current published or accessible version of [31] contains them and should state them explicitly in the present notation.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper has a coherent and attractive main idea, and the interpolation argument in Section 5 is sound. The unresolved point is the validity of the quoted product estimates (4.5) and (4.6) at the endpoint s1 = s0, which is explicitly allowed in the main theorem and is emphasized in Remark 1.2. I recommend requesting a revision in which the author either proves these estimates in the full range or restricts the theorem to s1 < s0; the rest of the analysis appears to be in good shape."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: this paper claims exactly what its title says, and on a first pass the claim is plausible. Shi transfers the pure energy method from Xin–Xu's compressible Navier–Stokes preprint to the Poisson system, and the removal of the low-frequency smallness condition is a real extension of Bie–Wang–Yao, Chikami–Danchin, and Shi–Xu. The Lyapunov structure is coherent: low-frequency ~a,u energy, high-frequency ∇a,u energy, negative-Besov evolution, then interpolation. The decay rates in Theorem 1.2 have the right character, with density decaying one-half power faster than velocity due to the Poisson potential.\n\nWhere it gets shaky: Lemma 4.1, the heart of the paper, depends on product estimates (4.4)–(4.5) cited from [31] and (4.6) cited from the author's own [26]. The text does not prove them. That would be acceptable if they were standard, but they are not: at the endpoint s1=s0, (4.5) reads ||FG||_{B^{-d/p}_{2,∞}} ≲ ||F||_{B^{d/p-1}_{p,1}} ||G||_{B^{1-d/p}_{2,∞}}, with the two right-hand regularities summing to zero. This is the borderline for Bony's paraproduct, and the usual remainder estimate fails. The paper uses (4.5) only with G = ∇u^ℓ, the low-frequency part, so the estimate may be true there, but the text neither says so nor gives the argument. Since Theorem 1.2 explicitly allows s1=s0 and the interpolation step uses it, this is a real gap, not a cosmetic one.\n\nThe rest of the argument checks out in structure: the low- and high-frequency estimates in Section 3 are standard, the Gronwall closure (4.12) is a genuine Gronwall argument, and the interpolation steps (5.9)–(5.10) align. I found no circularity or equation-fitting. The citation pattern is normal; the main self-citation supplies a product estimate, which is not a problem if the estimate is correct.\n\nBottom line: this is a serious paper deserving a serious referee, but the referee should insist on seeing proofs of (4.4)–(4.6), especially the low-frequency G case of (4.5), and on either proving the s1=s0 endpoint or restricting the theorem. Right now, the result is conditional on estimates I cannot verify from the text.\n\nI'd send it to review, with a request that the author supply the missing product-law arguments. For a reading group, it's a good example of the pure-energy technique, but the unresolved endpoint makes me hesitate to cite it in the next year.","headline":"A plausible pure-energy proof that removes low-frequency smallness for Navier–Stokes–Poisson, but the endpoint s1=s0 rests on an unproved borderline product law cited from a preprint.","tokens_in":22280,"tokens_out":11952,"would_cite":false,"duration_ms":99924,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q35","35B40","76N15"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves optimal time-decay rates for the compressible Navier-Stokes-Poisson equations, in critical Besov spaces, requiring only boundedness—not smallness—of the low-frequency initial data.","keywords":["compressible Navier-Stokes-Poisson equations","optimal time-decay estimates","critical Besov spaces","pure energy method","low-frequency smallness","negative Besov norms","Lyapunov inequality","nonlinear Gronwall"],"falsifier":"Compute the low-frequency product norm in (4.6) for a specific pair of functions, for example taking $F=G$ equal to a Schwartz function localized at frequency $2^j$, and check whether $\\|FG^h\\|_{\\dot B^{-s_1}_{2,\\infty}}\\lesssim \\|F\\|_{\\dot B^{d/p-1}_{p,1}}\\|G^h\\|_{\\dot B^{d/p-1}_{p,1}}$ holds for every $s_1\\in(1-d/2,s_0]$ in dimensions $d=2,3$; a single violated exponent would break Lemma 4.1.","tokens_in":21250,"feed_emoji":"📉","tokens_out":7268,"duration_ms":60690,"temperature":0.7,"pith_summary":"The paper establishes sharp large-time decay rates for global strong solutions of the compressible Navier-Stokes-Poisson equations near a constant equilibrium, in the same critical Besov spaces where global existence was already known. Its main theorem states that if the low-frequency parts of the initial density and velocity have bounded norms in suitable negative Besov spaces—no smallness required—then the density decays like $(1+t)^{-\\frac d2(\\frac12-\\frac1p)-\\frac{s_1+s+1}{2}}$ and the velocity like $(1+t)^{-\\frac d2(\\frac12-\\frac1p)-\\frac{s_1+s}{2}}$. The proof uses a pure energy argument in place of the spectral analysis used in earlier work. A sympathetic reader would care because removing the smallness condition makes the decay estimates available for a broader class of physically relevant initial data, including highly oscillating velocity fields.","feed_headline":"Navier-Stokes-Poisson: sharp decay without small low-frequency data","feed_subtitle":"Density and velocity in charged fluids decay at sharp rates even when initial low-frequency modes are only bounded.","key_machinery":"The argument runs through a Lyapunov-type inequality for the energy norm $\\|(\\tilde a,u)^\\ell\\|_{\\dot B^{\\frac d2-1}_{2,1}}+\\|(\\nabla a,u)^h\\|_{\\dot B^{\\frac dp-1}_{p,1}}$, where $\\tilde a=\\Lambda^{-1}a$. The inequality is driven by the negative Besov norm $\\|(\\tilde a,u)^\\ell\\|_{\\dot B^{-s_1}_{2,\\infty}}$, which is shown to stay bounded for all time by a nonlinear Gronwall argument once certain product estimates hold. The load-bearing product bounds are (4.4)-(4.5), taken from the companion paper [31], and (4.6), taken from the author's earlier work [26]; these control the low-frequency nonlinear terms and make the Gronwall closure possible. The final decay rates emerge from real interpolation between the bounded negative norm and the energy norm.","core_discovery":"The central claim is Theorem 1.2: under the global-existence assumptions of Theorem 1.1, if the low-frequency parts of the initial data lie in $\\dot B^{-s_1-1}_{2,\\infty}$ (density) and $\\dot B^{-s_1}_{2,\\infty}$ (velocity) with bounded norms, then for all $t\\ge 0$ and all admissible $s$, $\\|(\\rho-1)(t)\\|_{\\dot B^s_{p,1}}\\lesssim(1+t)^{-\\frac d2(\\frac12-\\frac1p)-\\frac{s_1+s+1}{2}}$ and $\\|u(t)\\|_{\\dot B^s_{p,1}}\\lesssim(1+t)^{-\\frac d2(\\frac12-\\frac1p)-\\frac{s_1+s}{2}}$. The density therefore decays half a power faster than the velocity, an effect attributed to the Poisson potential. The innovation is that the low-frequency norms only need to be bounded, not small, whereas earlier critical-space decay results required smallness of the low frequencies.","pith_inferences":["The same pure-energy strategy may extend to other systems with a nonlocal damping term, such as two-fluid plasma or Navier-Stokes-Maxwell models, where spectral analysis is considerably harder; this is an extension the paper does not pursue.","One could test the sharpness of the threshold $s_1>1-d/2$: the proof needs strict inequality for interpolation, so the endpoint $s_1=1-d/2$ might still decay but with a logarithmic correction, a question left open here.","A direct verification of the product estimate (4.6) across the full stated range would settle whether the method genuinely removes smallness in all cases, since a single failure would break the Gronwall closure."],"forward_implications":["The smallness condition on low-frequency data, imposed in earlier critical-space decay results for this system, is no longer needed; only boundedness is required.","The decay rates are described as optimal, matching the behavior predicted by linearized analysis, with the density half a power faster than the velocity.","The estimates cover dimensions $d\\ge 2$ and the case $p>d$, where the velocity regularity exponent $d/p-1$ may be negative, so highly oscillating initial velocity fields are admitted.","Corollary 1.1 converts the Besov estimates into explicit $L^r$ decay rates for derivatives of density and velocity, for $p\\le r\\le \\infty$."],"supporting_citations":[{"why":"Supplies the global-existence theorem (Theorem 1.1) for strong solutions in critical Besov spaces on which the decay result is built.","marker":"[6]"},{"why":"Provides the regularity assumption on low frequencies and the crucial product estimate (4.6) used to bound high-frequency nonlinear terms.","marker":"[26]"},{"why":"Supplies the product estimates (4.4)-(4.5) and the pure-energy method for compressible Navier-Stokes without additional smallness assumptions.","marker":"[31]"},{"why":"Introduces the low-frequency assumption framework for optimal time-decay estimates that the present work refines.","marker":"[30]"},{"why":"Provides the earlier optimal time-decay framework in critical $L^p$ spaces and product estimates used in the energy argument.","marker":"[12]"},{"why":"Establishes $L^q$-$L^2$ decay for the Navier-Stokes-Poisson system, identifying the density half-power faster decay that the paper reproduces in critical spaces.","marker":"[27]"},{"why":"Gives the spectral analysis of the linearized system and the benchmark optimal decay rates that the nonlinear result is compared with.","marker":"[20]"},{"why":"Provides an earlier critical-$L^p$ decay result requiring smallness of low frequencies, serving as the comparison baseline that the paper improves.","marker":"[2]"}],"fun_headline_variants":["NSP optimal decay without low-frequency smallness","Charged fluid decay rates sharp without smallness","Pure energy proof removes NSP smallness assumption","Density decays faster than velocity in NSP","Bounded low frequencies enough for NSP decay"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument's final closure rests on a product estimate, taken from the author's earlier work, being valid for every regularity exponent in the full stated range; if that estimate fails at any one exponent, the Gronwall step that keeps the negative Besov norm bounded collapses.","fun_headline_variants_meta":{"raw":{"variants":["NSP optimal decay without low-frequency smallness","Charged fluid decay rates sharp without smallness","Pure energy proof removes NSP smallness assumption","Density decays faster than velocity in NSP","Bounded low frequencies enough for NSP decay"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000303,"raw_usage":{"total_tokens":1710,"prompt_tokens":878,"completion_tokens":832,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":494,"completion_tokens_details":{"reasoning_tokens":760}},"tokens_in":494,"tokens_out":832,"duration_ms":7881,"temperature":1.0,"reasoning_tokens":760,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:20:38.762716+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the low-frequency product norm in (4.6) for a specific pair of functions, for example taking $F=G$ equal to a Schwartz function localized at frequency $2^j$, and check whether $\\|FG^h\\|_{\\dot B^{-s_1}_{2,\\infty}}\\lesssim \\|F\\|_{\\dot B^{d/p-1}_{p,1}}\\|G^h\\|_{\\dot B^{d/p-1}_{p,1}}$ holds for every $s_1\\in(1-d/2,s_0]$ in dimensions $d=2,3$; a single violated exponent would break Lemma 4.1.","supporting_citations":[{"cited_title":"Chikami, R","cited_arxiv_id":null,"evidence_quote":"Supplies the global-existence theorem (Theorem 1.1) for strong solutions in critical Besov spaces on which the decay result is built."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the regularity assumption on low frequencies and the crucial product estimate (4.6) used to bound high-frequency nonlinear terms."},{"cited_title":"Danchin, J","cited_arxiv_id":null,"evidence_quote":"Provides the earlier optimal time-decay framework in critical $L^p$ spaces and product estimates used in the energy argument."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes $L^q$-$L^2$ decay for the Navier-Stokes-Poisson system, identifying the density half-power faster decay that the paper reproduces in critical spaces."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the spectral analysis of the linearized system and the benchmark optimal decay rates that the nonlinear result is compared with."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides an earlier critical-$L^p$ decay result requiring smallness of low frequencies, serving as the comparison baseline that the paper improves."}],"review_version":1}