{"id":"751bf1b4-43e9-4777-b147-f38c2b97df14","arxiv_id":"2504.15968","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A Palais-Smale sequence for the mixed local-nonlocal critical problem decomposes into a weak limit and classical Sobolev bubbles, yielding a Coron-type existence theorem for large dimensions.","lead":"This mathematics paper proves a global compactness theorem for a PDE combining the Laplacian with a fractional Laplacian: every bounded energy sequence splits into a main solution plus a handful of concentrating bubbles. It then uses this tool to show positive solutions exist in annular-shaped domains, but only when the space dimension is large.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Global compactness theorem is structurally sound; the load-bearing gap is in Theorem 1.3, where the constrained functional S on M is asserted to inherit the Palais-Smale condition from I_{0,s} without proof.","rationale":"The reader's weakest_assumption (Lemma 2.2) is the right structural fact to examine, but it is true: on a bounded smooth domain, boundedness in X_0^{1,2} plus L^2 convergence gives [u_k-u_0]_s → 0 by the interpolation inequality [u]_s ≤ C ||u||_{L^2}^{1-s} ||∇u||_{L^2}^s, including the R^{2N} tail. Thus the reduction of the remainder to the purely local functional I_∞ is legitimate, and Theorem 1.2 is sound. The genuinely load-bearing gap in the paper is the passage from (PS) for I_{0,s} to (PS) for the constrained functional S on M in the proof of Theorem 1.3. This is stated as an immediate 'Consequently' but requires a Lagrange-multiplier reduction: one must prove that a constrained PS sequence for S produces an unconstrained PS sequence for I_{0,s} after rescaling, and conversely for critical points. The paper supplies no computation of the multiplier or of the energy-level correspondence. This matters because the deformation lemma is applied to S on M; without the (PS) condition, the flow Φ and the contradiction argument have no foundation. The gap is standard and fixable, so the reader's CONDITIONAL verdict is appropriate. I also note that the proof says bubble interaction 'follows directly from [5]' (Bahri-Coron) where Struwe's text is likely intended; this is a citation issue, not a mathematical one. The numerical caveat in Remark 1.1 about Lemma 4.3 failing in lower dimensions is consistent with the large-N hypothesis and does not undermine the theorem.","tokens_in":22362,"tokens_out":47853,"duration_ms":422498,"concrete_test":"Carry out the constrained-to-unconstrained reduction: for a PS sequence u_k ∈ M for S with S(u_k) → c, prove existence of μ_k with S'(u_k)-μ_k N'(u_k) → 0 in X', test against u_k to obtain μ_k → c/2^*, and verify that v_k = S(u_k)^{1/(2^*-2)} u_k satisfies ||I'_{0,s}(v_k)||_{X'} → 0 and I_{0,s}(v_k) → N^{-1} c^{N/2}. If this computation goes through, the deformation argument in Theorem 1.3 is justified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The profile-decomposition proof of Theorem 1.2 is sound: Lemma 2.2 is valid, and the interpolation argument makes the full fractional seminorm of the remainder vanish, so the bubbles are local. The weak point is in the Coron application. In the proof of Theorem 1.3 the authors state: 'By Theorem 1.2, the functional I_{0,s} satisfies (PS)_β ... Consequently, S(·;Ω) satisfies (PS)_β on M.' This consequence is not immediate. A Palais-Smale sequence for S on the L^{2^*}-sphere M is constrained: its differential vanishes only on the tangent space. One must introduce Lagrange multipliers μ_k with S'(u_k)-μ_k N'(u_k) → 0, test against u_k to identify μ_k, then set v_k = S(u_k)^{1/(2^*-2)} u_k and prove that v_k is a PS sequence for I_{0,s} at the level N^{-1} S(u_k)^{N/2}. The paper omits this verification entirely. Without it, the deformation lemma and the topological contradiction are unsupported. This is a gap in Theorem 1.3, not in the global compactness theorem.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper establishes a global compactness (profile decomposition) theorem for Palais–Smale sequences of the functional I_{λ,s}(u) = (1/2)ρ(u)^2 − (λ/2)∫|u|^2 − (1/2*)∫|u|^{2*} associated with the mixed local–nonlocal problem −Δu + (−Δ)^s u − λu = |u|^{2*−2}u in a bounded smooth domain Ω, with zero Dirichlet condition on R^N \\ Ω. The main result, Theorem 1.2, asserts that any PS sequence decomposes, up to a subsequence, as u_k = u0 + Σ_{i=1}^l U_i^k + o(1) in D^{1,2}(R^N), where u0 solves the original problem, each U_i^k is a rescaled solution of the purely local critical equation −ΔU = |U|^{2*−2}U in R^N, the scales and centers of distinct bubbles separate, and energies add. The proof relies on a compact embedding X_0^{1,2}(Ω) ↪ H^s(Ω) (Lemma 2.2), which forces the fractional part of the remainder to vanish. As an application, Theorem 1.3 claims a Coron-type existence result: for domains containing a sufficiently thick annular region and for sufficiently large dimension N, a positive solution exists with energy between the first and second critical thresholds.","tokens_in":22580,"tokens_out":22627,"duration_ms":185247,"significance":"The global compactness theorem is a meaningful extension of Struwe's classical decomposition to the mixed local–nonlocal setting. The proof is structurally transparent and the key mechanism—the compact embedding into H^s—is correctly identified; the external input S_{N,s}=S_N from [8] is clearly acknowledged. The decomposition is parameter-free and the energy additivity is cleanly derived. The Coron-type application follows the standard deformation/topological framework and would be a valuable consequence if the proof is completed. The main weaknesses are concentrated in the statement's bubble scaling and in an unproved Palais–Smale transfer in the application; both are fixable and do not affect the validity of the core compactness mechanism.","major_comments":[{"comment":"The assertion 'By Theorem 1.2, the functional I_{0,s} satisfies the Palais–Smale condition (PS)_β on X_0^{1,2}(Ω) for 1/N S_N^{N/2} < β < 2/N S_N^{N/2}' is not a direct consequence of Theorem 1.2 and is false without the contradiction hypothesis. Under the assumption that no positive solution exists, the claim requires a short argument using the decomposition, Lemma 5.1, and the energy quantization of bubbles: any PS sequence at such a level would have u0 = 0 and exactly one bubble with energy (1/N)S_N^{N/2}, contradicting the strict inequality. The subsequent consequence 'Consequently, S(·;Ω) satisfies (PS)_β on M' also needs the standard link between constrained and unconstrained PS sequences (introducing Lagrange multipliers and rescaling v_k = S(u_k)^{1/(2*−2)} u_k to obtain a PS sequence for I_{0,s}). Both steps are omitted, and the deformation lemma is applied to S on M, so this gap is load-bearing for Theorem 1.3.","section":"Section 4, proof of Theorem 1.3"},{"comment":"The definition U_i^k(x) := (R_i^k)^{-(N−2)/2} U_i((x−x_i^k)/R_i^k) does not solve the critical equation −ΔU = |U|^{2*−2}U in R^N; the correct concentrating bubble is R^{(N−2)/2} U(R(x−x0)) as used in Step 3 of the proof. As written, the statement attributes false scaling to the profiles. The statement should be corrected, for example by writing U_i^k(x) = (R_i^k)^{(N−2)/2} U_i(R_i^k(x−x_i^k)) or by introducing ε_i^k = 1/R_i^k and writing U_i^k(x) = (ε_i^k)^{-(N−2)/2} U_i((x−x_i^k)/ε_i^k). Harmonizing the statement with the proof is necessary for the theorem to be correct as stated.","section":"Theorem 1.2, statement of U_i^k"}],"minor_comments":[{"comment":"The choice of the intermediate exponent s' is not stated explicitly; the proof should specify 0 < s < s' < 1 before applying Lemma 2.1.","section":"Section 2.2, proof of Lemma 2.2"},{"comment":"The inequality |y| ≥ (2R−t)/(1−t) improves to |y| ≥ 2R−1 for t close to 1, not to |y| ≥ 2R; the subsequent integral over R^N \\ B_{2R} is harmless but the estimate should be adjusted or clarified.","section":"Lemma 4.1, estimate for I1"},{"comment":"The proof of Step 3 uses the scaling R_k^{(N−2)/2} v_0(R_k(·−x_k)) without relating it to the U_i^k notation in the theorem statement; after correcting the statement this link should be made explicit for readability.","section":"Theorem 1.2 and Step 3 of its proof"},{"comment":"Reference [8] is cited only as an arXiv preprint; the authors should update it to the published version if one exists.","section":"References"},{"comment":"There are several typographical and formatting issues, including 'in volving' in the abstract, inconsistent use of 'P.S.' and 'PS', and the reference to Coron's paper is missing the period after 'Sér I Math'; a careful proofreading pass is recommended.","section":"Various"}],"recommendation":"major_revision","confidential_remarks":"The global compactness theorem (Theorem 1.2) is essentially correct and valuable. The two main issues—the incorrect scaling in the statement of the bubbles and the missing Palais–Smale transfer in the Coron application—are local and fixable by rewriting; they do not undermine the core mathematical mechanism. I recommend requesting a revision rather than rejecting. I would also encourage the authors to add the short argument that under the contradiction assumption no PS sequence exists at the intermediate level, since this is the subtle step that makes the deformation argument valid."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe thing to know: the global compactness theorem (Theorem 1.2) is a genuine result for the mixed local-nonlocal operator and the proof is structurally sound. The key observation is that the compact embedding X_0^{1,2}(Ω) ↪ H^s(Ω) forces the fractional seminorm of the remainder v_k to vanish, so the bubbles are solutions of the purely local limit equation. That reduces the decomposition to Struwe's classical argument, and the reduction checks out.\n\nWhat is new: this is the first Struwe-type decomposition for −Δ+(−Δ)^s with critical exponent. The boundedness argument, the Brezis–Lieb bookkeeping, and the iterative bubble extraction are all in order. The dependence on [8] for S_{N,s}=S_N is external and legitimate; no circularity.\n\nThe soft spot is Theorem 1.3. The step \"By Theorem 1.2, I_{0,s} satisfies (PS)_β ... Consequently, S(·;Ω) satisfies (PS)_β on M\" is asserted without proof. That consequence is not immediate: a Palais–Smale sequence on the sphere gives a Lagrange multiplier sequence, and the standard rescaling v_k = S(u_k)^{1/(2^*-2)}u_k is needed to convert it into a PS sequence for I_{0,s}. The authors do this at the special level S_N later in the proof, but not for the whole interval on which the deformation lemma relies. As written, the deformation step is unsupported. I think this is a fixable gap, but it must be supplied.\n\nMinor issues: the final \"interaction of the bubbles follows directly from [5]\" should cite Struwe's argument, not Bahri–Coron. The boundedness estimate in Step 1 has a sloppy line where a sublinear term is absorbed as o(1)ρ(uk), but the conclusion is fine. Lemma 4.3's asymptotics look plausible; I did not redo the Stirling bookkeeping.\n\nBottom line: if I were the editor, I would send it to a referee. The central compactness theorem is worth publishing, and the authors need to fill the missing PS inheritance for S on M and fix the small presentation issues. This is a serious paper.","headline":"The global compactness theorem is real and mostly sound; the Coron application has a fillable gap in transferring the Palais–Smale condition to the sphere.","tokens_in":23141,"tokens_out":6368,"would_cite":true,"duration_ms":54039,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35B33","35B38","35J20","35J61"],"pacs":[],"model":"deepseek-v4-flash","headline":"In a mixed local-nonlocal problem, bubbles stay purely local","keywords":["global compactness","profile decomposition","PS sequence","mixed local nonlocal operator","critical Sobolev exponent","fractional Laplacian","annular domain","high-energy solutions"],"falsifier":"A concrete check is to evaluate the inequality in Lemma 4.3, namely $[u_0]_s^2 < (2^{2/N}-1)\\int_{\\mathbb{R}^N}|\\nabla u_0|^2\\,dx$ for the standard bubble $u_0(x)=(1+|x|^2)^{-(N-2)/2}$, numerically for $N=3,4,5,\\dots$ and a fixed $s\\in(0,1)$. If it already holds in low dimensions, the restriction to large $N$ in the annular-domain theorem is not optimal; if it fails for some $N$, the claimed threshold $N_0(s)$ is real and can be located.","tokens_in":1992,"feed_emoji":"🫧","tokens_out":4069,"duration_ms":97531,"temperature":0.7,"pith_summary":"This paper proves a global compactness (profile decomposition) theorem for PS sequences of a critical exponent problem driven by the mixed local nonlocal operator $-\\Delta+(-\\Delta)^s$ with a linear spectral term on a smooth bounded domain. It shows that every bounded PS sequence decomposes, up to a subsequence, into a weak solution of the original problem plus finitely many rescaled bubbles, where each bubble solves the purely local equation $-\\Delta U=|U|^{2^*-2}U$ in $\\mathbb{R}^N$, and the energy of the sequence is the sum of the energy of the weak limit and the bubble energies. The proof rests on a compact embedding of the mixed space into the fractional Sobolev space $H^s$, which kills the fractional part of the remainder. As an application, the authors solve an annular-domain existence problem: when the domain has a hole, the problem with $\\lambda=0$ admits a positive high-energy solution in sufficiently large dimension.","feed_headline":"Bubbles in a mixed local-nonlocal problem are purely local","feed_subtitle":"Profile decomposition pins down exactly when compactness fails and yields a positive solution on annular domains.","key_machinery":"The load-bearing mechanism is the compact embedding of Lemma 2.2: the space $X_0^{1,2}(\\Omega)$, the completion of $C_c^\\infty(\\Omega)$ in the norm built from the gradient $L^2$ norm plus the fractional seminorm, embeds compactly into the fractional Sobolev space $H^s(\\Omega)$. This forces $[u_k-u_0]_s\\to0$ for the weakly convergent remainder, so the nonlocal part of the energy and of the equation disappears on the bubbling scales. That reduction converts the mixed problem into the classical local global compactness theorem, whose bubbles are the extremal functions for the sharp constant in the standard Sobolev inequality, i.e., solutions of the limiting equation (1.4). The same embedding is used in the application to show that the fractional seminorm of approximate bubbles vanishes as the cut-off radius tends to infinity.","core_discovery":"The central discovery is Theorem 1.2: for every Palais-Smale (PS) sequence $\\{u_k\\}$ for the functional $I_{\\lambda,s}$ at level $\\beta$, there exist a weak solution $u_0$ of the original problem, a finite index $l$, sequences of scales and centers, and solutions $U_i$ of $-\\Delta U_i=|U_i|^{2^*-2}U_i$ in $\\mathbb{R}^N$ such that, up to a subsequence, $u_k = u_0 + \\sum_{i=1}^l U_i^k + o(1)$ in $D^{1,2}(\\mathbb{R}^N)$, with the scales and centers of distinct bubbles separating, and the energies adding: $I_{\\lambda,s}(u_k) \\to I_{\\lambda,s}(u_0)+\\sum_{i=1}^l I_\\infty(U_i)$. The proof shows that the remainder $v_k=u_k-u_0$ has vanishing fractional seminorm $[v_k]_s\\to0$, so $v_k$ becomes a PS sequence for the pure Laplacian functional at infinity, and the classical local global compactness theorem then applies to extract the bubbles. The fractional Laplacian therefore plays no role in the concentration profiles.","pith_inferences":["The decomposition suggests a general principle: lower-order nonlocal terms are subcritical in the concentration regime, so any mixed operator whose nonlocal part scales like the fractional Laplacian will have purely local bubbles; this could extend to more general mixed operators with different fractional orders.","The large-dimension assumption in the annular-domain theorem is tied to the numerical inequality in Lemma 4.3; checking that inequality for low dimensions would pinpoint the exact threshold N0(s) and possibly allow a perturbation argument with a small coefficient on the fractional term to lower the required dimension.","Because the compact embedding into H^s is the structural fact, one could test whether the same decomposition holds on unbounded domains or with rough boundaries, where the embedding may fail and mixed bubbles could appear.","The energy additivity could be exploited for multiplicity results: if one can construct PS sequences with several well-separated bubbles, their limits should yield multiple positive solutions in suitable topological settings."],"forward_implications":["Every PS sequence converges up to finitely many purely local bubbles, so compactness is lost only at integer multiples of the classical critical level.","The functional satisfies the Palais-Smale condition at all energies strictly below the first critical level, extending the Brezis-Nirenberg threshold to the mixed operator.","The decomposition yields a positive high-energy solution on an annular domain with a hole, provided the dimension is large and the radius ratio is large, with energy between the first and second critical levels.","Any sign-changing solution with lambda=0 has energy at least twice the first critical level, so the solution obtained is necessarily positive."],"supporting_citations":[{"why":"Supplies the classical global compactness theorem for the pure Laplacian, to which the mixed problem is reduced after the fractional part vanishes.","marker":"[47]"},{"why":"Establishes that the sharp constant for the mixed operator equals the classical Sobolev constant and is not attained, identifying the limit profiles as solutions of the local equation.","marker":"[8]"},{"why":"Provides the interpolation and fractional embedding results (Lemma 2.1, Theorem 6.21) used to prove the compact embedding Lemma 2.2 and the estimates in Lemma 4.3.","marker":"[33]"},{"why":"Gives a global compactness type result for PS sequences in fractional Sobolev spaces, the nonlocal analogue that motivates the profile analysis.","marker":"[39]"},{"why":"Supplies the local profile decomposition step (Lemma 3.3) and the deformation lemma used in the proof of the annular-domain existence theorem.","marker":"[48]"},{"why":"Founded the critical exponent problem with linear perturbation and the compactness-threshold strategy that frames the analysis.","marker":"[10]"},{"why":"Introduced the topological method for annular domains with a hole, which the application adapts to the mixed operator.","marker":"[15]"},{"why":"Provides the bubble interaction result used in Step 4 of the proof to separate the scales and centers of distinct bubbles.","marker":"[5]"}],"fun_headline_variants":["Bubbles stay local in mixed local-nonlocal Brezis-Nirenberg problem","Nonlocal operator doesn't shape bubbles: global compactness proof","Mixed operator problem: bubbles are purely local, compactness recovered","Fractional Laplacian varnish: bubbles are local in mixed problem","Global compactness: fractional part irrelevant for bubble profiles"],"cache_read_input_tokens":25344,"weakest_assumption_plain":"The argument collapses if the mixed space $X_0^{1,2}(\\Omega)$ did not embed compactly into the fractional Sobolev space $H^s(\\Omega)$, because then the fractional part of the remainder would not vanish and the bubbles could be solutions of the mixed equation in $\\mathbb{R}^N$ instead of purely local ones.","fun_headline_variants_meta":{"raw":{"variants":["Bubbles stay local in mixed local-nonlocal Brezis-Nirenberg problem","Nonlocal operator doesn't shape bubbles: global compactness proof","Mixed operator problem: bubbles are purely local, compactness recovered","Fractional Laplacian varnish: bubbles are local in mixed problem","Global compactness: fractional part irrelevant for bubble profiles"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000424,"raw_usage":{"total_tokens":2219,"prompt_tokens":1033,"completion_tokens":1186,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":649,"completion_tokens_details":{"reasoning_tokens":1094}},"tokens_in":649,"tokens_out":1186,"duration_ms":8307,"temperature":1.0,"reasoning_tokens":1094,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T11:15:19.392052+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete check is to evaluate the inequality in Lemma 4.3, namely $[u_0]_s^2 < (2^{2/N}-1)\\int_{\\mathbb{R}^N}|\\nabla u_0|^2\\,dx$ for the standard bubble $u_0(x)=(1+|x|^2)^{-(N-2)/2}$, numerically for $N=3,4,5,\\dots$ and a fixed $s\\in(0,1)$. If it already holds in low dimensions, the restriction to large $N$ in the annular-domain theorem is not optimal; if it fails for some $N$, the claimed threshold $N_0(s)$ is real and can be located.","supporting_citations":[{"cited_title":"A global compactness result for ellipt ic boundary value problems involving limiting nonlinearities","cited_arxiv_id":null,"evidence_quote":"Supplies the classical global compactness theorem for the pure Laplacian, to which the mixed problem is reduced after the fractional part vanishes."},{"cited_title":"A first course in fractional Sobolev spaces , volume 229 of Graduate Studies in Mathematics","cited_arxiv_id":null,"evidence_quote":"Provides the interpolation and fractional embedding results (Lemma 2.1, Theorem 6.21) used to prove the compact embedding Lemma 2.2 and the estimates in Lemma 4.3."},{"cited_title":"A global comp actness type result for Palais-Smale sequences in fractional Sobolev spaces","cited_arxiv_id":null,"evidence_quote":"Gives a global compactness type result for PS sequences in fractional Sobolev spaces, the nonlocal analogue that motivates the profile analysis."},{"cited_title":"Variational methods, volume 34 of Ergebnisse der Mathematik und ihrer Gren- zgebiete","cited_arxiv_id":null,"evidence_quote":"Supplies the local profile decomposition step (Lemma 3.3) and the deformation lemma used in the proof of the annular-domain existence theorem."},{"cited_title":"Topologie et cas limite des injecti ons de Sobolev","cited_arxiv_id":null,"evidence_quote":"Introduced the topological method for annular domains with a hole, which the application adapts to the mixed operator."},{"cited_title":"On a nonlinear ellipt ic equation involving the criti- cal Sobolev exponent: the effect of the topology of the domain","cited_arxiv_id":null,"evidence_quote":"Provides the bubble interaction result used in Step 4 of the proof to separate the scales and centers of distinct bubbles."}],"review_version":1}