{"id":"4db084c8-5d20-45cb-9608-ab24eba4f467","arxiv_id":"2411.13896","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Forced Navier-Stokes solutions in the energy space can blow up in finite time under critical-order forces, including an explicit small force in L∞_t L^{3/2}_x.","lead":"A construction produces a smooth, finite-energy fluid flow that becomes unbounded in finite time under a force whose scaling matches point-source forces. A small explicit version of the force, living in the standard critical space, is also included.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof that the perturbation u has square-integrable gradient is asserted after (2.48) but not demonstrated; since v=Z+u belongs to the energy space only if ∇u∈L^2_{t,x}, this is the load-bearing step that needs an explicit check.","rationale":"The reader's weakest assumption correctly identifies the unproved passage from (2.48) to ∇u∈L^2_t L^2_x, and I agree this is the central load-bearing step for the energy-space conclusion. The fixed point in X gives only a weighted L∞ bound; the gradient regularity is a separate claim that must be verified before v=Z+u can be declared an energy-class solution. My concrete test isolates exactly that claim: check that the four nonlinear products lie in L^2_{t,x} and that the kernel multiplier argument applies. I do not see evidence that the claim is false—the decay in (2.27) and the η-suppression in u suggest all four products should be L^2—but the paper does not supply the necessary estimates. The rest of the argument, including the contraction estimates and the blow-up mechanism, appears internally consistent. The exact identities and the explicit linear blow-up solution Z are strong supporting evidence. The concern is therefore a substantial but fixable gap, not a demonstrated contradiction, so the appropriate verdict remains the reader's CONDITIONAL.","tokens_in":13574,"tokens_out":38790,"duration_ms":361998,"concrete_test":"Starting from the fixed-point equation (2.48), write f=u⊗u+Z⊗u+u⊗Z+Z⊗Z and apply the L^2_{t,x} Fourier-multiplier bound for the Stokes kernel, whose first-derivative symbol is ξ_jξ_i/(iτ+|ξ|^2). Compute or bound the four space-time L^2 norms of the products using (2.27) and |u|≤η/(1+|x|^2), explicitly handling the log terms near |x|=0 and near |x|=√t where |ln(|x|^2+1-t)| is small. If all four norms are finite, the assertion ∇u∈L^2_{t,x} is justified and v is in the energy space; if any product fails, the theorem's energy-space claim is unsupported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1.2 claims v=Z+u is a smooth solution in the energy space that blows up at T. The linear part Z is controlled by the pointwise estimates (2.27)-(2.28), and ∇Z is square-integrable in space-time. The remaining requirement is ∇u∈L^2_t L^2_x. The fixed point is obtained only in the weighted sup-norm space X defined in (2.32), so it gives no gradient information. The passage from (2.48) to the conclusion '|∇u|∈L^2_t L^2_x' is a single unproved assertion: it relies on the claim that the second derivatives of the Stokes kernel are bounded from space-time L^2 to itself, applied to the four product terms u⊗u, Z⊗u, u⊗Z, and Z⊗Z. For this to work, each product must lie in L^2_{t,x}; in particular Z⊗Z must satisfy ∫_0^1∫ |Z|^4 dx dt < ∞, and the mixed products need ∫ |Z|^2|u|^2 dx dt < ∞. These are plausible from (2.27) and the bound |u|≤η/(1+|x|^2), but no computation is given. Moreover, the later integration by parts yielding (2.49) already presupposes ∇u∈L^2, so if the singular-integral step were invalid the regularity bootstrap would be circular. Because membership of v in the energy space is a stated conclusion of the theorem, this omitted verification is the most load-bearing gap.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs finite-time blow-up solutions of the forced, incompressible Navier-Stokes equations in three dimensions. Theorem 1.1(a) gives, on any domain containing the origin, a smooth compactly supported solution in the energy space whose force has principal part -e^{-|x|^2}/(|x|^2+T-t)(1,0,0). Theorem 1.1(b) gives a smooth solution on R^3 with force in L^\\infty_t L^{3/2}_x whose principal part contains an additional logarithmic factor. Theorem 1.2 gives, for all sufficiently small \\delta>0, a smooth energy-space solution with zero initial data and the fully explicit force F=-\\delta e^{-|x|^2}/((|x|^2+T-t)(1+|\\ln(|x|^2+T-t)|))(1,0,0), which blows up at time T. The method solves a scalar forced heat equation, applies the transformation v=curl(curl((-\\Delta)^{-1}h)), absorbs gradient errors into the pressure, and for Theorem 1.2 writes v=Z+u where Z is the linear Stokes solution and u is obtained by a contraction mapping in a weighted sup-norm space.","tokens_in":13877,"tokens_out":9357,"duration_ms":83581,"significance":"If the gaps identified below are repaired, this would be a significant advance: the forcing in Theorems 1.1(b) and 1.2 is at or below the critical scaling order -2, unlike earlier supercritical constructions, and Theorem 1.2 provides an explicit force with zero initial data and finite-time blow-up in the energy space. The paper is also careful to avoid the trivial 'define F after v' objection by prescribing F before solving for the perturbation in Theorem 1.2. The proof contains several clean and useful estimates, especially the Riesz-transform argument showing blow-up of v at the origin and the weighted contraction argument. However, the manuscript as it stands has a load-bearing gap in the justification of the energy-space membership of the perturbation u in Theorem 1.2, and part (b) of Theorem 1.1 is not proved with sufficient detail.","major_comments":[{"comment":"The statement 'Due to the boundedness of u and ln type singularity of Z, the standard parabolic singular integral theory tells us |\\nabla u| \\in L^2_t L^2_x' is asserted without proof. The fixed point is obtained in the sup-norm space X of (2.32), which gives no gradient information. To apply the L^2-boundedness of second derivatives of the Stokes kernel, one must first show that each product u_i u, Z_i u, u_i Z, Z_i Z lies in L^2_t L^2_x. For Z_i Z this requires checking \\int_0^1\\int |Z|^4 dxdt<\\infty; this is plausible from (2.27) and the fact that (1+|x|^2)^{-4} is integrable in R^3, but no computation is given. For the mixed products one needs \\int_0^1\\int |Z|^2|u|^2 dxdt<\\infty, which likewise is not written out. Moreover, the integration by parts that yields (2.49) already presupposes \\nabla u \\in L^2, so if the singular-integral step is not justified first, the regularity bootstrap is circular. Since membership of v in the energy space is an explicit conclusion of Theorem 1.2, this omitted verification is a load-bearing gap and must be supplied with explicit estimates.","section":"Section 2, proof of Theorem 1.2, after Eq. (2.48)"},{"comment":"The proof of part (b) is summarized as 'The rest of the proof follows that of part (a) step by step with only small changes.' This is insufficient for the theorem's central claim that F=f-v\\nabla v belongs to L^\\infty_t L^{3/2}_x. In particular, one needs explicit bounds showing that both the heat forcing f in (2.20) and the nonlinear term v\\nabla v, estimated through the log-log analogues of (2.18) and (2.19), have finite L^{3/2}_x norm uniformly in t. The sentence 'It is now easy to see the forcing term F=f-v\\nabla v is in the space L^\\infty_t L^{3/2}_x' is too terse for a theorem-level assertion; a short verification should be included.","section":"Section 2, proof of Theorem 1.1(b), around Eq. (2.23)"},{"comment":"The contraction argument itself is mostly sound, but the notation for the Stokes kernel convolution is imprecise: in (2.33) the expression '\\partial_i K u_i u(y,s)' should be written with parentheses, e.g. '\\partial_{y_i}K(x,t;y,s)\\, u_i(y,s)u(y,s)', and similarly in (2.45). More substantively, the proof of the containment M(B(0,\\eta))\\subset B(0,\\eta) uses the estimate (2.42) for T_4 that contains a factor C\\delta^2; the subsequent choice of \\eta and \\delta is legitimate, but the dependence of the constants on the logarithmic factors should be tracked to ensure that the smallness condition is uniform in \\delta and \\eta. This is a presentation issue rather than a logical flaw, but it would help the reader to see the explicit constants.","section":"Section 2, proof of Theorem 1.2, Eqs. (2.33) and (2.45)"}],"minor_comments":[{"comment":"The sentence 'Just like Theorem 1.2, Z is in the energy space and it blows up at t=1' should refer to Theorem 1.1, not Theorem 1.2.","section":"Section 2, proof of Theorem 1.2, first paragraph"},{"comment":"The denominator in the last inequality is written as '|1+|x|^2'; it should be '1+|x|^2'.","section":"Equation (2.47)"},{"comment":"Minor language issues: 'scales logarithmic worse than -1' should be 'scales logarithmically worse than -1'; '0 initial value' should be 'zero initial value'; 'through out' should be 'throughout'.","section":"Throughout"},{"comment":"The phrase 'smooth, compactly supported solution v' is potentially confusing because v is constructed with a spatial cutoff \\varphi and is compactly supported for each time, but it is not compactly supported in time on [0,1). The statement could be clarified.","section":"Theorem 1.1(a) statement"}],"recommendation":"major_revision","confidential_remarks":"The manuscript contains a genuine and interesting construction, and the central gap in Theorem 1.2 appears to be repairable with explicit estimates for the L^2 integrability of the gradient of u. Part (b) of Theorem 1.1 also needs a short but explicit verification of the critical-space membership of the force. I recommend major revision with a request to include these details; I do not see a fundamental obstruction to the claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real step past the author's own earlier supercritical example, and Theorem 1.2 is the part that matters—an explicit force of critical order −2, fixed before the solution is constructed, giving a smooth blow-up in the energy space with zero initial data. I think the paper deserves refereeing. It also has one genuinely under-verified step that a referee should push on.\n\nWhat's new: the force scaling is now critical rather than supercritical, and the log-subcritical variant in Theorem 1.1(b) plus the explicit L∞_t L^{3/2}_x force in Theorem 1.2 are genuine improvements over [9] and Scheffer's \"bizarre function\". The construction itself is straightforward—solve the heat equation with the forcing, turn the solution into a divergence-free field via curl curl, absorb the error into pressure—but the details matter and most of them are checked. The blow-up proof via radial symmetry is clean, and the energy bound for the linear solution through Riesz transforms is solid. The fixed point for u in Theorem 1.2 is done with care; the bounds on T4, T3, T2, T1 are explicit and the estimates (2.39)–(2.44) check out.\n\nWhere I'd be cautious: the step after (2.48) is the load-bearing one for the energy-space claim. The contraction gives u only in the weighted sup-norm space X. To conclude v = Z + u is in the energy space you need ∇u ∈ L^2_t L^2_x, and the paper asserts this from \"standard parabolic singular integral theory\" without showing that the four products u⊗u, Z⊗u, u⊗Z, Z⊗Z are in L^2_{t,x}. The decay bounds (2.27) and |u| ≤ η/(1+|x|^2) make that plausible, and I don't think it's wrong, but it's exactly the kind of estimate that should be written down, especially since the later integration by parts in (2.49) already assumes ∇u is integrable. A referee should ask for a lemma here.\n\nPart (b) of Theorem 1.1 is also summarized rather than proved; the phrase \"step by step with only small changes\" is doing a lot of work. I believe it follows, but it should be laid out.\n\nThe circularity concern is mostly handled: Theorem 1.1 is reverse-engineered by design, and the paper says so. Theorem 1.2 is the honest version, with the force fixed first. The remaining question is only the technical check above.\n\nWho is this for? People working on forced Navier-Stokes blow-up, Scheffer's program, or critical regularity thresholds. It doesn't speak directly to the unforced regularity problem, and the authors are appropriately careful about that.\n\nRecommendation: send it out. It's a constructive example with an explicit force, and the central idea is sound. The referee should require a complete proof of the gradient estimate for u and a fuller treatment of Theorem 1.1(b).","headline":"A genuine improvement over earlier forced blow-up constructions: an explicit critical-order force with zero initial data, but the proof skips the one estimate that puts the solution in the energy space.","tokens_in":14468,"tokens_out":2813,"would_cite":true,"duration_ms":54367,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35Q30","76N10"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper constructs a forced Navier-Stokes solution that stays in the energy space but blows up at a finite time, with a forcing term at the critical scaling order -2.","keywords":["Navier-Stokes equations","finite-time blow-up","critical forcing","energy space","Stokes kernel","fixed point","log-subcritical force"],"falsifier":"Take a fixed small $\\delta>0$, evaluate the heat-kernel integral defining $h_1$ at $x=0$ as $t\\to1^-$, and check that $|h_1(0,t)|$ diverges at least like $\\ln\\ln(1/(1-t))$; if it stays bounded, the linear solution $Z$ does not blow up. Then compute the fixed point $u$ of the integral equation and test whether $\\int_0^1\\int_{\\mathbb{R}^3}|\\nabla u|^2\\,dx\\,dt$ is finite; if that integral diverges, the claimed energy-space blow-up solution collapses.","tokens_in":13318,"feed_emoji":"🌊","tokens_out":16006,"duration_ms":133013,"temperature":0.7,"pith_summary":"This paper aims to show that a forcing term at the critical scaling order $-2$ can drive a smooth solution of the Navier-Stokes equations to blow up in finite time, even when the initial velocity is zero and the solution remains in the energy space (finite kinetic energy, finite accumulated dissipation) throughout. The construction starts from a blow-up solution of a linear heat equation with the critical force, turns it into a divergence-free velocity via a curl-curl transform, and then adds a small correction obtained by a fixed-point argument. If the proof is right, the explicit force $-\\delta \\frac{e^{-|x|^2}}{(|x|^2+T-t)[1+|\\ln(|x|^2+T-t)|]}(1,0,0)$ is enough to produce a singularity, suggesting that natural point-source forces with the same scaling could also cause singularity formation.","feed_headline":"Critical force drives Navier-Stokes to finite-time blow-up","feed_subtitle":"A critical-scale force, with zero initial data, still yields a smooth energy-class flow that blows up at time T.","key_machinery":"The carrying object is the linear Stokes solution $Z=\\operatorname{curl}\\operatorname{curl}((-\\Delta)^{-1}h)$, where $h$ solves the vector heat equation $\\Delta h-\\partial_t h=F$ with the critical force; the curl-curl operator makes $Z$ divergence-free and leaves a gradient error that is absorbed into the pressure. The correction $u$ is produced as the fixed point of an integral equation built from the Stokes kernel $K$, the fundamental solution of the linearized Stokes system, whose decay $|\\nabla K|\\le C/(|x-y|+\\sqrt{t-s})^4$ makes the map a contraction in the weighted space with norm $\\sup(1+|x|)^2|u|$. The pointwise estimates $|Z(x,t)|\\le C\\delta(1+|x|^2)^{-1}(1+|\\ln(|x|^2+1-t)|)$ and a similar bound for $\\nabla Z$ are what make the nonlinear terms in the equation for $u$ subcritical, so a small-data argument applies.","core_discovery":"The central claim is Theorem 1.2: there exists $\\delta_0>0$ such that for every $\\delta\\in(0,\\delta_0]$, the forced Navier-Stokes equations with zero initial data and force $F=-\\delta \\frac{e^{-|x|^2}}{(|x|^2+T-t)[1+|\\ln(|x|^2+T-t)|]}(1,0,0)$ have a smooth solution $v$ in the energy space on $\\mathbb{R}^3\\times[0,T)$ that blows up at time $T$. The solution is written $v=Z+u$, where $Z$ is the explicitly defined linear Stokes solution produced by the heat-kernel formula and $u$ is the fixed point of a contraction in a weighted space. A stronger but less explicit version works in any open domain containing the origin with no-slip boundary conditions and also produces a force in the standard critical space $L^\\infty_t L^{3/2}_x$. The proof uses the pressure to absorb a gradient error and checks that the nonlinear term $v\\nabla v$ stays below the critical scaling, so the blow-up is driven by the linear, force-induced term.","pith_inferences":["If the contraction proof is stable, the same $v=Z+u$ splitting should yield blow-up for a family of forces obtained by adding small, smooth, divergence-free perturbations to the explicit $F$, as long as they stay at or below the critical scale.","The paper leaves the sign of $F\\cdot v$ uncontrolled; a natural next step would be to construct a version with $F\\cdot v\\ge0$, which would make the force plausibly 'slowing' rather than 'pushing' the flow and strengthen the physical interpretation.","The result is a step toward singular solutions of the unforced equations only if the pressure and force can eventually be removed; nothing in this paper shows that, so the step remains incomplete.","A numerical check of the predicted $\\ln\\ln(1/(1-t))$ blow-up rate at $x=0$ could test whether the linear mechanism is the one actually operating for finite $\\delta$."],"forward_implications":["For the explicit force with any sufficiently small amplitude $\\delta$, the forced Navier-Stokes equations with zero initial data admit a smooth solution in the energy space that blows up at $T$.","The blow-up is driven by the linear Stokes solution $Z$, which diverges like $\\ln\\ln(1/(1-t))$ for the explicit force (or faster in the non-explicit variant), while the correction $u$ stays bounded.","The same construction works in any open domain containing the origin with no-slip boundary conditions when the force's principal part is the explicit one, and it also yields forces in the critical space $L^\\infty_t L^{3/2}_x$.","Because the force has scaling order $-2$, matching point-source forces, the result shows that a critical (not supercritical) driving term can produce a finite-time singularity."],"supporting_citations":[{"why":"Supplies the Stokes kernel decay estimates used to bound the integral terms in the contraction argument.","marker":"[8]"},{"why":"Provides the function-space and fixed-point framework that justifies solving the perturbed equation for u in a broader class.","marker":"[10]"},{"why":"The earlier construction with a supercritical force whose heat-equation-to-Stokes transform the present proof takes over and modifies.","marker":"[9]"}],"fun_headline_variants":["Critical force triggers Navier-Stokes blow-up","Forced Navier-Stokes blows up at critical scale","Zero data still yields blow-up under critical force","Explicit critical force causes finite-time singularity","Navier-Stokes singularity from critical scaling force"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof stands on two pointwise decay estimates for the linear solution $Z$ and on the statement, not fully proven in the text, that the fixed-point correction $u$ has a square-integrable gradient on $\\mathbb{R}^3\\times[0,1)$; if either fails, $v=Z+u$ is not guaranteed to lie in the energy space and the blow-up solution is not established.","fun_headline_variants_meta":{"raw":{"variants":["Critical force triggers Navier-Stokes blow-up","Forced Navier-Stokes blows up at critical scale","Zero data still yields blow-up under critical force","Explicit critical force causes finite-time singularity","Navier-Stokes singularity from critical scaling force"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000539,"raw_usage":{"total_tokens":2606,"prompt_tokens":987,"completion_tokens":1619,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":603,"completion_tokens_details":{"reasoning_tokens":1547}},"tokens_in":603,"tokens_out":1619,"duration_ms":10322,"temperature":1.0,"reasoning_tokens":1547,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:47:47.961572+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a fixed small $\\delta>0$, evaluate the heat-kernel integral defining $h_1$ at $x=0$ as $t\\to1^-$, and check that $|h_1(0,t)|$ diverges at least like $\\ln\\ln(1/(1-t))$; if it stays bounded, the linear solution $Z$ does not blow up. Then compute the fixed point $u$ of the integral equation and test whether $\\int_0^1\\int_{\\mathbb{R}^3}|\\nabla u|^2\\,dx\\,dt$ is finite; if that integral diverges, the claimed energy-space blow-up solution collapses.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Stokes kernel decay estimates used to bound the integral terms in the contraction argument."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the function-space and fixed-point framework that justifies solving the perturbed equation for u in a broader class."},{"cited_title":"A blow up solution of the Navier-Stokes equations with a super critical forcing term","cited_arxiv_id":"2311.12306","evidence_quote":"The earlier construction with a supercritical force whose heat-equation-to-Stokes transform the present proof takes over and modifies."}],"review_version":1}