{"id":"98cb9a19-aac1-4e5a-9aa4-60fb27f333a8","arxiv_id":"2506.11656","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Existence, uniqueness, and summability gains are established for singular local-nonlocal problems with absorption via a Talenti-type comparison.","lead":"This paper proves that singular elliptic equations combining local and nonlocal diffusion, with an absorption term, have unique finite-energy solutions. It also obtains sharp integrability bounds through a comparison with a symmetrized Laplacian problem.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"For 1/2<s<1, the proof of Theorem 1.2 depends essentially on (1.5): without f>0 a.e., the vanishing of the nonlocal term in (3.32) does not follow, and Appendix A only removes (1.5) under extra hypotheses.","rationale":"I read the paper in good faith and checked the main lines of the proofs of Theorems 1.2, 1.3, and 4.1. The existence argument is coherent under the standing assumptions: the approximating scheme, the a priori estimates in Lemma 3.2, and the passage to the limit in the local and absorption terms are standard and appear correct. The uniqueness proof via the T_k test function is also sound under the monotonicity assumptions. The Talenti comparison in Theorem 4.1 is plausible and internally consistent: the nonlocal term and the absorption term both drop out with the correct sign, and the symmetrized Laplacian comparison yields the stated L^p, L∞ and Orlicz bounds. The single weakest point is the use of assumption (1.5) in the proof of Theorem 1.2 for s>1/2. In Case II, the proof needs |{u=0}|=0 to make the limiting nonlocal integral vanish; without (1.5), that set can be a positive-measure subset of {f=0}, and the sign of the limiting nonlocal term is not controlled. The authors are transparent about this, but calling the assumption 'purely technical' overstates the situation: Appendix A removes it only for the model problem and only under additional regularity and growth hypotheses. The concrete test with an explicit piecewise-affine u and linear φ in one dimension shows that the required sign inequality is false in general when |{u=0}|>0, confirming that the proof as written depends essentially on (1.5). This is exactly the concern the reader flagged in their weakest_assumption, so I agree with their conditional verdict and see no reason to move it.","tokens_in":34626,"tokens_out":16502,"duration_ms":160615,"concrete_test":"Evaluate the double limit in Case II on an explicit configuration: take n=1, Ω=(0,1), s=3/4, u(x)=max{x-1/2,0} (so {u=0}=(0,1/2) has positive measure), φ(x)=x, and compute I=∫_0^{1/2}∫_0^1 (u(x)-u(y))(φ(x)-φ(y))/|x-y|^{5/2} dy dx. For x<1/2<y one has u(x)-u(y)=-u(y)<0 and φ(x)-φ(y)=x-y<0, so the integrand is positive on a set of positive measure; hence I>0. Since the proof of (3.32) for s>1/2 reduces to showing this quantity is ≤0 when |{u=0}|>0, a positive I demonstrates that the argument collapses without (1.5), and any repair must use additional structure as in Appendix A.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 1.2 for 1/2<s<1 reaches the crucial claim (3.32) only through the chain (3.29)-(3.41). From (3.28) one obtains {u=0}⊆{f=0} up to null sets, and (1.5) then forces |{u=0}|=0, which makes the integral in (3.41) vanish. If f is allowed to vanish on a set of positive measure, {u=0} may have positive measure. In the estimate (3.33), the nonlocal term is controlled by the limsup as δ→0 of ∫_{u=0}×R^n (u(x)-u(y))(φ(x)-φ(y))|x-y|^{-n-2s}. For x∈{u=0} and y∉{u=0}, this integrand equals -u(y)(φ(x)-φ(y)), which has no sign; hence the conclusion (3.32) is not justified in that setting. The paper itself labels (1.5) as 'purely technical' in Appendix A, but then supplies an alternative proof only for the model nonlinearity and only under the extra hypotheses q≤2^*-1 and f∈C^δ_loc. Thus the general existence theorem carries a real restriction on the data that is load-bearing for the argument as written; it is not a harmless normalization. Uniqueness and the Talenti comparison are not affected by this concern.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the mixed local-nonlocal singular elliptic problem L u + g(u) = h(u) f in Ω, u > 0 in Ω, u = 0 outside Ω, where L = -Δ + (-Δ)^s and s ∈ (0,1). Under structural assumptions on h, g, and f, it proves existence of an energy solution in X^{1,2}(Ω) with g(u)u ∈ L^1(Ω), uniqueness under monotonicity of h and g, and, for the model problem, a Talenti-type pointwise comparison that yields explicit L^p, L∞, and Orlicz-type estimates. The proof passes through a Leray-Lions theorem for the mixed operator, a two-step approximation scheme, and a new interpolation argument that controls the nonlocal term in the limiting procedure.","tokens_in":34948,"tokens_out":11616,"duration_ms":112795,"significance":"If the results hold, they extend the known regularizing effect of absorption terms in singular problems from the purely local setting of Oliva to local-nonlocal operators, and they provide explicit quantitative bounds via symmetrization. The paper's technical core—the limiting argument for the nonlocal term, especially the interpolation inequality used to prove (3.32)—is original and makes essential use of the local part of the operator. The proofs are detailed and, apart from the caveat below, self-contained. The main limitation is that the existence theorem for 1/2 < s < 1 requires the datum f to be positive almost everywhere, a condition the authors label 'purely technical' but remove only partially in an appendix; this restriction is load-bearing for the proof of the central limit.","major_comments":[{"comment":"The crucial limit (3.32) is proved only through the chain (3.29)-(3.41), where (1.5) forces |{u = 0}| = 0 and hence makes the integral in (3.41) vanish. If f vanishes on a set of positive measure, the limsup in (3.40) need not be zero, and the squeeze argument for B_{n,δ} collapses. The text in the Introduction is honest about needing (1.5), but Appendix A's assertion that (1.5) is 'purely technical and plays no essential role' is not supported: the appendix removes it only for the model problem and only under the extra hypotheses (A.2)-(A.3), namely q ≤ 2^* - 1 and f ∈ C^δ_loc. This is a real restriction on the generality of Theorem 1.2, not a harmless normalization; I recommend stating the theorem with (1.5) explicitly marked as essential for s > 1/2 and reformulating the general removal as an open problem.","section":"Section 3, proof of Theorem 1.2, Case II (1/2 < s < 1), Eqs. (3.29)-(3.41)"},{"comment":"The abstract claims existence and uniqueness of energy solutions for 'singular problems with absorption driven by local-nonlocal operators' without mentioning the positivity condition (1.5). Since for 1/2 < s < 1 the general existence theorem is proved only under this additional datum assumption, the summary of results overstates the scope. Please qualify the abstract and the introductory description of Theorem 1.2 to reflect the dependence on (1.5).","section":"Abstract and Introduction"}],"minor_comments":[{"comment":"There is a typo in the abstract: 'solutionns' should be 'solutions'.","section":"Abstract"},{"comment":"The double limit notation 'lim_{\\delta\\to0} lim_{n\\to\\infty}' is used several times; consider clarifying the order of the limits and, in (3.40), specifying that the limiting integral after 'limsup' is taken in the sense that the upper bound is meant for the iterated limit.","section":"Section 3.1, around Eq. (3.38)"},{"comment":"The references to 'Lemma 3.2-(1)' and 'Lemma 3.2-(2)' are ambiguous because Lemma 3.2 is not stated with numbered subclaims; the intended referents are the estimates (3.12) and (3.13), and the text should say so explicitly.","section":"Appendix A, after Eq. (A.8)"},{"comment":"In the statement of estimate (1.8), the factor 'n2ω2/n_n' is hard to read; the proof of (1.8) should be checked against the displayed constant to ensure the notation for ω_n is consistent.","section":"Theorem 1.5 and its proof"}],"recommendation":"major_revision","confidential_remarks":"The mathematical core is sound under the stated assumptions, and the interpolation argument for the nonlocal term is a genuine contribution. The main issue is the mismatch between the 'purely technical' label for (1.5) and the partial nature of Appendix A; this should be resolved either by proving the general removal or by clearly presenting (1.5) as an essential hypothesis. The paper is within the scope of the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Stefano,\n\nHere's my read on arXiv:2506.11656. The paper proves existence, uniqueness, and a Talenti-type comparison for singular problems with absorption for the mixed operator L=-Δ+(-Δ)^s, s∈(0,1). The genuinely new piece is the comparison principle for this operator, which yields explicit L^p, L∞ and Orlicz bounds for the model problem. The existence theory covers a reasonably general class of singularities and absorption terms, and the proofs are detailed and mostly clean: the functional setting for L is set up properly, the Leray-Lions step is adapted, and the nonlocal term in the limit is handled via a fractional interpolation inequality that uses the local diffusion in an essential way. No fitted parameters, and the self-citations are to results that are actually used.\n\nThe main soft spot is assumption (1.5): for 1/2<s<1, the proof of Theorem 1.2 requires f>0 almost everywhere. This enters at (3.32) through (3.29)-(3.41). Without it, {u=0} may have positive measure and the nonlocal term in the limit does not vanish; the stress-test concern is correct. The authors call (1.5) 'purely technical', but Appendix A removes it only for the model problem under extra hypotheses (q≤2^*-1 and f∈C^δ_loc). So the general existence theorem carries a real restriction on the data. It is not a harmless normalization, and the paper's own appendix shows the authors are aware. That said, the restriction is explicit and is natural for many applications; the s≤1/2 case is unaffected, and uniqueness and the comparison are untouched.\n\nMinor: the abstract has a typo ('solutionns'), and the approximation scheme follows Oliva [57] closely, but the nonlocal term is not a trivial perturbation—the treatment of (3.32) is the real technical contribution.\n\nWho is this for: researchers working on mixed local-nonlocal operators, particularly singular problems and symmetrization. They will get a useful tool and a careful baseline. It deserves a serious referee; I would not desk-reject. Recommend to engage, with the (1.5) caveat made prominent.","headline":"Solid extension of Oliva's theory to mixed local-nonlocal operators, with a new Talenti comparison; one real technical restriction (f>0 a.e. for s>1/2) that should be stated more honestly.","tokens_in":35456,"tokens_out":5011,"would_cite":false,"duration_ms":46657,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["35J75","35M12","35J61","35B51"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves existence, uniqueness, and Talenti-type comparison estimates for singular elliptic problems driven by the mixed local-nonlocal operator $L=-\\Delta+(-\\Delta)^s$.","keywords":["local-nonlocal operators","singular elliptic problems","absorption terms","energy solutions","Talenti comparison","rearrangements","fractional Laplacian","regularizing effects"],"falsifier":"Solve the model problem numerically in a non-radial bounded domain with $s=3/4$, $\\gamma=1/2$, $q=1$, and $f\\equiv 1$, compute the decreasing rearrangement $u^*$, and compare it with $((\\gamma+1)v^*)^{1/(\\gamma+1)}$, where $v^*$ is given by the explicit formula (4.17) for the symmetrized Laplacian. If any $\\tau$ satisfies $u^*(\\tau) > ((\\gamma+1)v^*(\\tau))^{1/(\\gamma+1)}$, the comparison principle in Theorem 1.5 is false.","tokens_in":34413,"feed_emoji":"📐","tokens_out":18439,"duration_ms":161858,"temperature":0.7,"pith_summary":"The paper studies the mixed local-nonlocal operator $L=-\\Delta+(-\\Delta)^s$, the sum of the usual Laplacian and a fractional Laplacian, and proves that singular problems with absorption, $Lu+g(u)=h(u)f$ in a bounded domain with $u>0$ and $u=0$ outside it, admit finite-energy solutions under broad structural assumptions on the singular term $h$, the absorption term $g$, and the datum $f$. For the model case $h(s)=s^{-\\gamma}$ and $g(s)=s^q$, it also proves a Talenti-type pointwise comparison between the decreasing rearrangement of the solution and the solution of a symmetrized Laplacian problem, which yields explicit $L^p$, $L^\\infty$, and Orlicz estimates. The reason to care is that absorption and singularity together produce a gain of summability beyond either effect alone, and the paper establishes this combined regularizing effect for local-nonlocal operators.","feed_headline":"Absorption tames singular local-nonlocal PDEs into energy solutions","feed_subtitle":"A Talenti comparison to a symmetrized Laplacian yields explicit Lp, L∞, and Orlicz bounds.","key_machinery":"The central object is the energy space $X^{1,2}(\\Omega)$, the completion of $C_c^\\infty(\\Omega)$ under the global norm that combines the $H^1$ gradient with the fractional seminorm of the nonlocal term; the mixed operator $L$ is monotone, coercive, and pseudomonotone on it, and a monotone-operator existence theorem (Theorem 2.6) supplies solutions of the approximating problems. The existence proof then runs through uniform $L^\\infty$ and energy estimates obtained by standard truncation and by testing with the absorption term, with the delicate nonlocal limit step handled by a fractional interpolation inequality that works precisely because of the presence of the local Laplacian. The comparison argument is carried by rearrangement machinery: coarea formula, isoperimetric inequality, Hardy-Littlewood inequality, and Bliss inequality transfer the pointwise bound to the symmetrized Laplacian problem.","core_discovery":"Under the standing assumptions (H)_h, (H)_f, (H)_g, which allow h to be singular at zero like $s^{-\\gamma}$ with $\\gamma\\le 1$ and to grow at most polynomially at infinity, allow f to be merely $L^1$ when $\\theta\\ge 1$ or $L^m$ when $\\theta<1$, require $f>0$ almost everywhere when $s>1/2$, and require $g(0)=0$ with a power lower bound when $\\theta<1$, Theorem 1.2 constructs a distributional solution u with $u|_\\Omega\\in H_0^1(\\Omega)$ and $g(u)u\\in L^1(\\Omega)$. Theorem 1.3 gives uniqueness whenever h is non-increasing and g is non-decreasing. For the model problem with $h(s)=s^{-\\gamma}$ and $g(s)=s^q$, Theorem 1.5 establishes the rearrangement comparison $u^*(\\tau) \\le ((\\gamma+1)v^*(\\tau))^{1/(\\gamma+1)}$ for almost every $\\tau\\in(0,|\\Omega|)$, where v solves the symmetrized Laplacian problem $-\\Delta v=f^\\sharp$ in the ball $\\Omega^\\sharp$, and from it derives explicit $L^p$ bounds with $p=nm(\\gamma+1)/(n-2m)$ when $1<m<n/2$, an $L^\\infty$ bound when $m>n/2$, and membership in an exponential Orlicz space when $m=n/2$.","pith_inferences":["Because the comparison proof in Section 4 simply discards the nonlocal term, the rearrangement bound is likely to survive for more general local-nonlocal operators, such as quasilinear variants or other fractional orders, provided the nonlocal term stays nonnegative under the chosen test functions; this is a natural but unproved extension.","The positivity assumption on f for $s>1/2$ looks like an artifact of the proof's handling of the nonlocal term, and the appendix's removal under subcritical growth and local smoothness suggests a strong maximum principle for L would remove it entirely, but no such principle is currently available for this operator.","Because the constants in Theorem 1.5 are independent of s, the local-nonlocal singular problem may inherit the Laplacian's regularizing effects exactly, which would give practical a priori bounds that do not require knowing the fractional order."],"forward_implications":["For the model problem there is a unique energy solution $u\\in H_0^1(\\Omega)\\cap L^{q+1}(\\Omega)$ whenever $q\\ge\\max\\{0,(1-m\\gamma)/(m-1)\\}$.","The rearrangement comparison gives the same explicit $L^p$, $L^\\infty$, and Orlicz integrability that the pure Laplacian singular problem would have, so the fractional part does not worsen the final summability bounds.","When m is below the threshold $m_\\gamma=2n/(n(\\gamma+1)-2(1-\\gamma))$, the absorption term produces a stronger regularizing effect than the Talenti estimate alone, as noted in Remark 1.6.","The existence and uniqueness results extend the classical singular-problem and absorption-type theorems for the pure Laplacian to the local-nonlocal operator $L$, with uniqueness under natural monotonicity assumptions."],"supporting_citations":[{"why":"Supplies the approximation scheme, the a priori estimate Lemma 3.2, and the local test-function argument that the paper adapts to the mixed operator.","marker":"[57]"},{"why":"Establishes the singular-only regularizing effect for -Δu=f/u^γ that this paper combines with absorption.","marker":"[20]"},{"why":"First combined absorption and singular lower-order terms in classical Dirichlet problems; the paper extends that combined regularizing effect.","marker":"[36]"},{"why":"Provides the symmetrization approach for singular semilinear Laplacian problems from which the Talenti comparison in Section 4 is adapted.","marker":"[24]"},{"why":"Bliss inequality converts the rearrangement comparison into the explicit L^p estimate in Theorem 1.5(i).","marker":"[17]"},{"why":"Supplies the fractional Sobolev interpolation inequality used to pass to the limit in the nonlocal term.","marker":"[52]"},{"why":"Gives the local-nonlocal regularity estimates used for L∞ bounds of approximating solutions and for the Appendix A classical-regularity argument.","marker":"[9]"},{"why":"Truncation method yields the uniform L∞ bounds in Lemma 3.1 and the a priori estimates.","marker":"[62]"}],"fun_headline_variants":["Absorption tames singular PDEs into explicit Lp and Orlicz bounds","Talenti comparison for singular local-nonlocal problems: summability gains","Existence, uniqueness, and sharper bounds for singular PDEs with absorption","Local-nonlocal singular problems tamed by absorption: new energy solutions"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"When the fractional exponent s is greater than 1/2, the proof relies on f being nonzero almost everywhere in Ω; allowing f to vanish on a set of positive measure breaks the step that forces the solution to be positive everywhere, and removing this restriction is shown only under extra smoothness and subcritical-growth hypotheses.","fun_headline_variants_meta":{"raw":{"variants":["Absorption tames singular PDEs into explicit Lp and Orlicz bounds","Talenti comparison for singular local-nonlocal problems: summability gains","Existence, uniqueness, and sharper bounds for singular PDEs with absorption","Local-nonlocal singular problems tamed by absorption: new energy solutions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000392,"raw_usage":{"total_tokens":2045,"prompt_tokens":912,"completion_tokens":1133,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":528,"completion_tokens_details":{"reasoning_tokens":1052}},"tokens_in":528,"tokens_out":1133,"duration_ms":10464,"temperature":1.0,"reasoning_tokens":1052,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T04:04:14.964459+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Solve the model problem numerically in a non-radial bounded domain with $s=3/4$, $\\gamma=1/2$, $q=1$, and $f\\equiv 1$, compute the decreasing rearrangement $u^*$, and compare it with $((\\gamma+1)v^*)^{1/(\\gamma+1)}$, where $v^*$ is given by the explicit formula (4.17) for the symmetrized Laplacian. If any $\\tau$ satisfies $u^*(\\tau) > ((\\gamma+1)v^*(\\tau))^{1/(\\gamma+1)}$, the comparison principle in Theorem 1.5 is false.","supporting_citations":[{"cited_title":"Oliva,Regularizing effect of absorption terms in singular problems, J","cited_arxiv_id":null,"evidence_quote":"Supplies the approximation scheme, the a priori estimate Lemma 3.2, and the local test-function argument that the paper adapts to the mixed operator."},{"cited_title":"Boccardo, L","cited_arxiv_id":null,"evidence_quote":"Establishes the singular-only regularizing effect for -Δu=f/u^γ that this paper combines with absorption."},{"cited_title":"De Cave, F","cited_arxiv_id":null,"evidence_quote":"First combined absorption and singular lower-order terms in classical Dirichlet problems; the paper extends that combined regularizing effect."},{"cited_title":"Brandolini, F","cited_arxiv_id":null,"evidence_quote":"Provides the symmetrization approach for singular semilinear Laplacian problems from which the Talenti comparison in Section 4 is adapted."},{"cited_title":"Bliss,An Integral Inequality, J","cited_arxiv_id":null,"evidence_quote":"Bliss inequality converts the rearrangement comparison into the explicit L^p estimate in Theorem 1.5(i)."},{"cited_title":"Leoni,A First Course in Fractional Sobolev Spaces, Grad","cited_arxiv_id":null,"evidence_quote":"Supplies the fractional Sobolev interpolation inequality used to pass to the limit in the nonlocal term."},{"cited_title":"Biagi, S","cited_arxiv_id":null,"evidence_quote":"Gives the local-nonlocal regularity estimates used for L∞ bounds of approximating solutions and for the Appendix A classical-regularity argument."},{"cited_title":"Stampacchia,Le probl` eme de Dirichlet pour les ´ equations elliptiques du second ordre ` a coefficients discontinus, Ann","cited_arxiv_id":null,"evidence_quote":"Truncation method yields the uniform L∞ bounds in Lemma 3.1 and the a priori estimates."}],"review_version":1}