{"id":"205a3c7f-26ff-4cfb-bb2e-9c5fc0fae877","arxiv_id":"2412.12414","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The hydrodynamic limit of a generalized exclusion process with long jumps and a slow barrier is ∂tρ = L^γ_κ F(ρ), with F any convergent power series on [0,Ne].","lead":"This paper rigorously derives nonlinear fractional diffusion equations from a random particle system with long jumps and a slow barrier. The hydrodynamic limit supports a power-series nonlinearity F(ρ), not just polynomials, opening a new class of fractional porous-medium equations to microscopic justification.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Replacement Lemma is not proved for the constant-profile entropy bound actually used in Theorem 2.27; Section 7 explicitly defers Hypothesis 3.12(i), yet Propositions 5.18 and 6.5 rely on it.","rationale":"The paper is a serious and technically detailed contribution, and the reader's CONDITIONAL verdict seems right. My stress-test sharpens the reason: the load-bearing gap is not just that f'_∞ < ∞ is a strong assumption, but that the Replacement Lemma is proved for a non-constant-profile entropy bound while the main theorem uses the constant-profile bound (2.41). This is explicitly acknowledged in Section 7 as left for the reader, and the lemma is genuinely used at two critical points: in Proposition 5.18 to obtain the H^{γ/2}_κ regularity of F(ρ), and in Proposition 6.5 to close the nonlinear terms in the integral equation. Without a completed proof of Lemma 3.15 under Hypothesis 3.12(i), Theorem 2.27 is incomplete for the central nonlinear, non-polynomial case. I also note the statement of Proposition 5.18 omits Hypothesis 2.25(ii), which Lemma 3.15 requires; this is probably an implicit reference to the standing assumptions of Theorem 2.27, but it should be stated. The omission is likely fixable: taking h ≡ θ makes the nonconstant-profile proof a special case when (2.39) holds, and identity (5.26) simplifies the Dirichlet-form estimate. So the concern does not overturn the paper's conclusion; it identifies a precise missing argument. The fast-diffusion cases such as F(ρ)=ρ^m with m∈(0,1) are explicitly outside Theorem 2.27, so they are a limitation, not a flaw, and the paper honestly states the weaker statements available there. No other issue I examined appeared to threaten the central claim: the operator L^γ_κ is dissipative even for κ>1, the lower-bound hypothesis plausibly enforces monotonicity of F, and the non-uniqueness regime γ∈(1,2) for (2.35) is disclosed. Therefore the reader's CONDITIONAL verdict should stand, pending the missing replacement-lemma proof.","tokens_in":89120,"tokens_out":15673,"duration_ms":142846,"concrete_test":"Write out the proof of Lemma 3.15 for Hypothesis 3.12(i): set h ≡ θ in Section 7, replace Proposition 7.5 by the exact identity (5.26), and verify that the one-block and two-block estimates (7.5)-(7.6), together with the global moving particle lemma (Lemma 7.10), go through using only Hypothesis 2.25(ii) and (2.41). If the two-block estimate secretly needs the f'_n bounds supplied by Proposition 7.5, then Theorem 2.27's nonlinear claim requires an extra assumption; if the argument closes without them, the omitted case is resolved and the theorem stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The main theorem assumes (2.41), i.e. relative entropy bounded against the constant product measure νθ. Under the paper's own definitions this is Hypothesis 3.12(i). The Replacement Lemma 3.15, the tool that converts microscopic product terms into powers of the density, is stated under Hypothesis 3.12, but Section 7 proves it only for Hypothesis 3.12(ii), the non-constant-profile condition (3.29) combined with the full Hypothesis 2.25. Immediately after Proposition 7.5 the authors write: 'We leave the case where Hypothesis 3.12 (i) holds for the reader, but we observe that the arguments are totally analogous and, in fact, simpler.' This is an explicit omitted case, and it is not cosmetic: Lemma 3.15 is applied in Proposition 6.5 to characterize limit points and in Proposition 5.18, Step III, to prove the energy estimate (5.3). Until the constant-profile case is written out, Theorem 2.27 is not fully proved for nonlinear F. There is a second, smaller wrinkle: Proposition 5.18 states only (2.39) and (2.41), but invokes Lemma 3.15, whose Hypothesis 3.12(i) additionally requires Hypothesis 2.25(ii); as written that assumption is missing. The gap is probably fillable, since h ≡ θ is a valid element of Ref and (5.26) gives an exact Dirichlet-form identity, so the missing case is a special case of the nonconstant argument when (2.39) holds; nonetheless, the proof as written is incomplete.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper derives, by means of the entropy method, the hydrodynamic limit of a generalized exclusion process with long jumps and a slow barrier. The limiting density is shown to solve a fractional nonlinear equation ∂tρ = L^γ_κ F(ρ), where F is a power series (2.8), L^γ_κ = κΔ^{γ/2} + (1−κ)Δ^{γ/2}_⋆, and the boundary conditions depend on α = lim α_n through Definitions 2.16–2.19. The main result, Theorem 2.27, covers polynomial and non-polynomial F under Hypotheses 2.25 and the entropy bound (2.41), with a partial statement (tightness only rather than full convergence) in the regime α = 0, γ ∈ (1,2), lim α_n r_n^γ ∈ (0,∞] because uniqueness of weak solutions for (2.35) is left open. The proof is organized as tightness (Section 4), energy estimates (Section 5), characterization of limit points via Dynkin's martingale (Section 6), and a replacement lemma (Section 7), with substantial appendices on fractional operators, discrete convergences, and measure-theoretic tools.","tokens_in":89360,"tokens_out":6373,"duration_ms":58452,"significance":"If completed, the paper would be a significant contribution: it gives rigorous microscopic derivations of non-polynomial fractional porous-medium-type equations, extends the slow-barrier analysis to a general scaling sequence α_n, and treats arbitrary dimension d ≥ 1 with occupation number Ne ≥ 1. The paper is carefully structured, with the solution notions, test-function spaces, and hypotheses set out precisely, and with detailed appendix material on the infinite-volume technicalities. A particular strength is the explicit separation of the hypotheses that are needed for tightness, energy estimates, uniqueness, and the replacement lemma, rather than hiding them in vague regularity conditions. The identified gap concerns the proof of the replacement lemma under one of its stated hypotheses; it is localized and likely repairable, but it currently affects the proof of the main theorem for nonlinear F.","major_comments":[{"comment":"Theorem 2.27 assumes the relative entropy bound (2.41), which together with Hypothesis 2.25(ii) is exactly Hypothesis 3.12(i). The Replacement Lemma 3.15 is stated under Hypothesis 3.12, but Section 7 proves it only for Hypothesis 3.12(ii): after Proposition 7.5 the text says “We leave the case where Hypothesis 3.12 (i) holds for the reader”. This is not a cosmetic omission, because Lemma 3.15 is used in the nonlinear case of Proposition 5.18 (Step III, for the energy estimate (5.3)) and in Proposition 6.5 (characterization of limit points). Consequently Theorem 2.27 is not fully proved for nonlinear F under its own hypotheses. Since h ≡ θ is an admissible element of Ref and (5.26) gives an exact Dirichlet-form identity for constant profiles, I expect the missing case to be repairable, but the proof as written should not rely on an unproved case of a central lemma.","section":"Section 7 / Lemma 3.15"},{"comment":"Proposition 5.18 is stated under (2.39) and (2.41) (plus the relevant choice of O), but its proof for nonlinear F applies Lemma 3.15. Under Hypothesis 3.12(i), Lemma 3.15 additionally requires Hypothesis 2.25(ii); this hypothesis is not listed among the assumptions of Proposition 5.18. The gap disappears inside Theorem 2.27 because Hypothesis 2.25(ii) is assumed there, but as a standalone proposition the statement is missing an assumption. Please add Hypothesis 2.25(ii) to Proposition 5.18, or replace the reference to Lemma 3.15 by a version whose hypotheses are met.","section":"Proposition 5.18"}],"minor_comments":[{"comment":"The phrase “by means of a of a partial differential equation” contains a duplicated article and should be corrected.","section":"Section 1.1"},{"comment":"The statement writes “Y^{n,G}_F ≲ …” while the definitions in (3.26) and (3.27) use the notation Y^{n,γ}_F(G) and Y^{n,γ}_S(G); please unify the notation.","section":"Proposition 3.8"},{"comment":"In the definition of Ω_w, the second case appears to repeat “if k⋆ = k+” where it should read “if k⋆ = k−”.","section":"Equation (7.37)"},{"comment":"The notation Λ^γ_κ is introduced in (2.5) but the text sometimes writes Λ^γ_α, Λ^γ_{n,α}, and Λ^γ_κ in adjacent formulas; a notational index or table would improve readability.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The missing replacement-lemma case is a genuine gap in the current version, but it is localized and likely fillable by specializing the nonconstant-profile argument to h ≡ θ or by using the exact Dirichlet-form identity (5.26). I would not recommend rejection on this basis; the paper's main contribution is substantive and the remaining architecture is coherent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real advance, but the main theorem is incomplete as written. The Replacement Lemma 3.15, which is the tool that converts local products into functions of density, is proved only under Hypothesis 3.12(ii), the non-constant profile entropy bound. Theorem 2.27 assumes (2.41), which is Hypothesis 3.12(i). Section 7 explicitly leaves that case 'for the reader', with the claim that the arguments are analogous and simpler. That is not a cosmetic omission: Lemma 3.15 is applied in Proposition 5.18 and Proposition 6.5, both feeding directly into the proof of Theorem 2.27. Until the constant-profile case is written out, the theorem is unproved for nonlinear F. The missing case probably goes through—h ≡ θ is a perfectly good element of Ref, and the non-constant argument should specialize—but as it stands, the proof has a hole.\n\nThere is a second, smaller wrinkle: Proposition 5.18 is stated with only (2.39) and (2.41), but it invokes Lemma 3.15, whose Hypothesis 3.12(i) also requires Hypothesis 2.25(ii). The main theorem includes (ii), so the proposition as stated is under-specified.\n\nNow the positives. The setting is genuinely new: a generalized exclusion process with Ne > 1, long jumps, and a slow barrier, producing hydrodynamic limits of the form ∂tρ = L^γ_κ F(ρ) where F is a convergent power series, not just a polynomial. The multidimensional case is handled, and the interpolating model of [17] is extended to non-polynomial F. The proof strategy follows the entropy method, and the paper is commendably detailed: tightness, energy estimates, characterization of limit points, and five appendices covering the analysis. The authors also state the open uniqueness case for γ ∈ (1,2) with the regional fractional Laplacian, which is honest.\n\nThe citation pattern is fine: prior works [7, 8, 10, 17] are used as tools, not as imported target results. Self-citations are to the authors' own earlier tools, used appropriately.\n\nWho should read this: anyone working on hydrodynamic limits of long-range exclusion processes, and PDE people interested in fractional diffusion equations as rigorous scaling limits. It deserves a serious referee, but the referee should ask the authors to supply the missing replacement-lemma case before acceptance.\n\nMy recommendation: send to peer review. If the constant-profile replacement lemma can be written out, this is a strong paper; if not, the main theorem stays conditional.","headline":"Genuinely new hydrodynamic-limit result, but the proof as written has a real gap: the Replacement Lemma is not proved for the constant-profile entropy bound that Theorem 2.27 actually uses.","tokens_in":90001,"tokens_out":3052,"would_cite":true,"duration_ms":27603,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["60K35","35R11","35S15"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that a generalized exclusion process with long jumps and a slow barrier converges, in the hydrodynamic limit, to the unique weak solution of ∂_tρ = L^γ_κ F(ρ), where F is an absolutely convergent power series and the…","keywords":["hydrodynamic limit","generalized exclusion process","fractional Laplacian","regional fractional Laplacian","slow barrier","fractional porous medium equation","power series nonlinearity","replacement lemma"],"falsifier":"Take a case covered by the theorem, e.g. F(ρ)=e^ρ (b_k^+ = 1/k!, b_k^- = 0), d=1, γ=3/2, N_e=1, α_n → 1/2, and initial density a step function. Simulate the accelerated process at two large system sizes and compare the empirical density with a high-resolution numerical solution of (2.32); agreement at the barrier would confirm the boundary-condition classification, while disagreement in the bulk or across the barrier beyond discretization error would falsify the hydrodynamic-limit claim.","tokens_in":88827,"feed_emoji":"🧮","tokens_out":11724,"duration_ms":99514,"temperature":0.7,"pith_summary":"Hydrodynamic limits of interacting particle systems usually produce linear or polynomial equations. This paper constructs a particle system whose limiting density obeys a fractional diffusion equation with a genuinely non-polynomial nonlinearity: ∂_tρ = L^γ_κ F(ρ), where L^γ_κ is the distorted fractional Laplacian $κΔ^{{γ/2}}$ + (1-κ)$Δ^{{γ/2}}$_⋆ and F is an absolutely convergent power series of the form (2.8), so F can be an exponential rather than just x or $x^{2}$. The microscopic model is a symmetric generalized exclusion process with long jumps, up to N_e particles per site, and a slow barrier between two half-spaces; the barrier strength α_n becomes the parameter κ in the limiting operator, and the boundary condition at the barrier is fixed by the limit α = lim α_n and by whether α_n r_n^γ vanishes or stays positive. The result matters because it provides a constructive dictionary: choose the coefficients of F, build the jump rates from products of occupation variables, and the hydrodynamic limit returns the chosen PDE. The proof is an adaptation of the entropy method to infinite volume, with a replacement lemma that holds under entropy bounds relative to non-constant profiles.","feed_headline":"Fractional PDEs with power-series nonlinearity from particle systems","feed_subtitle":"The limiting density obeys a distorted fractional Laplacian applied to any power-series function, such as an exponential.","key_machinery":"The argument is carried by a long-range gradient property of the rates. Identity (3.5) writes the current $c^{{(k),j}}$_{x,y}(η)[η(y)-η(x)] as $P^{{(k),j}}$(τ_y η) - $P^{{(k),j}}$(τ_x η) - ∇_j $A^{{(k),j}}$_{x,y}(η), with A antisymmetric; after summation by parts this turns the generator's integral term into a discrete version of the distorted fractional Laplacian acting on the local function F_n(η). The other essential mechanism is the Replacement Lemma (Lemma 3.15), proved by one-block and two-block estimates, which allows products of occupation variables to be replaced by products of empirical box averages under entropy bounds relative to non-constant profiles; this is what extracts the nonlinear function F from the microscopic products.","core_discovery":"Under Hypotheses 2.25, the n^γ-time-accelerated process converges, in probability and with respect to the empirical measure, to the unique weak solution of ∂_tρ = L^γ_κ F(ρ), where L^γ_κ = $κΔ^{{γ/2}}$ + (1-κ)$Δ^{{γ/2}}$_⋆ and F is the power series (2.8). The limiting equation is (2.32) when α = lim α_n lies in R_+\\setminus{1}, (2.33) when α=1, and (2.34) or (2.35) when α=0, depending on whether α_n r_n^γ tends to 0 or has a positive limit, with r_n^γ defined in (2.42). The nonlinearity is realized microscopically by jump constraints that are linear combinations of products of occupation variables; the novel step is a replacement lemma that, under an entropy bound with non-constant profiles, replaces products of occupancies by products of local density averages, so the martingale's integral term converges to ∫ L^γ_κ F(ρ) G. Uniqueness is proved in the regimes covered by the theorem; in the α=0, γ∈(1,2) case with lim α_n r_n^γ ∈ (0,∞], the paper proves tightness and that all limit points solve (2.35) but leaves uniqueness open.","pith_inferences":["The dictionary F ↔ jump rates is likely reversible: any sufficiently regular increasing diffusion coefficient with an absolutely convergent expansion of the type (2.8) should be realizable by some choice of b_k^±, which would extend the construction to nonlinearities such as D(ρ)=1/(1+ρ) or D(ρ)=ρ^m with m>2.","The threshold r_n^γ = 1_{γ<1} + log n 1_{γ=1} + n^{γ-1} 1_{γ>1} probably governs slow bonds in other long-range systems: the barrier is effectively insulating when α_n r_n^γ → 0 and transparent when the product has a positive limit, a classification one could test in zero-range or inclusion processes.","A practical check of the replacement mechanism: choose F with moderate coefficients so that (2.39) is close to failing, for example b_k^+ ≈ 1/k^{1+δ} with δ small, and look for finite-n corrections near the barrier; the proof predicts these corrections vanish only after the two limits n→∞ then ǫ→0, so simulations should show slow convergence in ǫ."],"forward_implications":["Any absolutely convergent F of the form (2.8), including F(ρ)=e^ρ, can be realized as the hydrodynamic nonlinearity of a symmetric long-range exclusion model, so rigorously derivable fractional PDEs are no longer limited to polynomials.","The slow barrier's limiting strength α selects the operator: α=1 gives the standard fractional Laplacian, α=0 gives the regional fractional Laplacian, and other values give a distorted mixture of the two.","The speed at which α_n approaches its limit selects the boundary condition: α_n r_n^γ → 0 yields vanishing fractional flux through the barrier, while a positive limit yields equality of the two one-sided fractional derivatives.","For fast diffusion F(ρ)=ρ^m with m∈(0,1), the weaker Hypotheses 2.28–2.29 still give tightness and the integral equations, even though full uniqueness uses the stronger slow-growth condition (2.39).","The replacement lemma works under the weaker entropy bound (3.29) with non-constant profiles, so the hydrodynamic limit does not require initial measures close to a constant-density equilibrium."],"supporting_citations":[{"why":"Supplies the long-range jump kernel p_γ(z)=c_γ|z|^{-d-γ} and the fractional Laplacian hydrodynamic limit for linear F.","marker":"[21]"},{"why":"Introduces the nonlinear constraints c^{(k)} from products of occupancies and the replacement-lemma route to ∂_tρ=Δ^{γ/2}ρ^2.","marker":"[7]"},{"why":"Extends the method to F(ρ)=ρ^m and provides the polynomial-rate framework the present paper generalizes to power series.","marker":"[8]"},{"why":"Introduces the slow bonds and the distorted fractional Laplacian L^γ_κ with the boundary-condition hierarchy.","marker":"[10]"},{"why":"Gives the generalized binomial expansion of ρ^m whose coefficients motivate the power-series F and the fast-diffusion coefficients.","marker":"[17]"},{"why":"Supplies the kinetic constraint rates r^{(k)} for the nearest-neighbor porous medium model underlying the multidimensional definitions.","marker":"[16]"},{"why":"Provides the product-form generalized exclusion setup with N_e > 1 and entropy estimates relative to non-constant profiles.","marker":"[13]"},{"why":"Provides the entropy method, the overall convergence strategy adapted here.","marker":"[20]"},{"why":"Supplies the tightness and martingale characterization framework for hydrodynamic limits, adapted by the authors to infinite volume.","marker":"[23]"}],"fun_headline_variants":["Fractional diffusions with power-series nonlinearity from exclusion with slow barrier","Non-polynomial fractional diffusion from particle exclusion with barrier","Power-series nonlinearity in fractional porous media from slow-barrier exclusion","Generalized fractional porous medium from exclusion with a slow barrier","Hydrodynamic limit yields fractional PDE with arbitrary power-series nonlinearity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the coefficient-weighted derivative sum f'_∞ = Σ_{k≥1} k(|b_k^+|+|b_k^-|) $N_e^{{k-1}}$ is finite, which keeps the microscopic rates bounded and the macroscopic F regular enough for energy estimates and uniqueness; this fails for fast-diffusion nonlinearities such as F(ρ)=ρ^m with m∈(0,1).","fun_headline_variants_meta":{"raw":{"variants":["Fractional diffusions with power-series nonlinearity from exclusion with slow barrier","Non-polynomial fractional diffusion from particle exclusion with barrier","Power-series nonlinearity in fractional porous media from slow-barrier exclusion","Generalized fractional porous medium from exclusion with a slow barrier","Hydrodynamic limit yields fractional PDE with arbitrary power-series nonlinearity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000575,"raw_usage":{"total_tokens":2707,"prompt_tokens":929,"completion_tokens":1778,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":545,"completion_tokens_details":{"reasoning_tokens":1692}},"tokens_in":545,"tokens_out":1778,"duration_ms":10628,"temperature":1.0,"reasoning_tokens":1692,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T14:07:31.623097+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a case covered by the theorem, e.g. F(ρ)=e^ρ (b_k^+ = 1/k!, b_k^- = 0), d=1, γ=3/2, N_e=1, α_n → 1/2, and initial density a step function. Simulate the accelerated process at two large system sizes and compare the empirical density with a high-resolution numerical solution of (2.32); agreement at the barrier would confirm the boundary-condition classification, while disagreement in the bulk or across the barrier beyond discretization error would falsify the hydrodynamic-limit claim.","supporting_citations":[{"cited_title":"Hydrodynamic limit of particle systems with long jumps","cited_arxiv_id":"0805.1326","evidence_quote":"Supplies the long-range jump kernel p_γ(z)=c_γ|z|^{-d-γ} and the fractional Laplacian hydrodynamic limit for linear F."},{"cited_title":"Cardoso, R","cited_arxiv_id":null,"evidence_quote":"Introduces the nonlinear constraints c^{(k)} from products of occupancies and the replacement-lemma route to ∂_tρ=Δ^{γ/2}ρ^2."},{"cited_title":"Cardoso and P","cited_arxiv_id":null,"evidence_quote":"Extends the method to F(ρ)=ρ^m and provides the polynomial-rate framework the present paper generalizes to power series."},{"cited_title":"Cardoso, P","cited_arxiv_id":null,"evidence_quote":"Introduces the slow bonds and the distorted fractional Laplacian L^γ_κ with the boundary-condition hierarchy."},{"cited_title":"From Exclusion to Slow and Fast Diffusion","cited_arxiv_id":"2301.06585","evidence_quote":"Gives the generalized binomial expansion of ρ^m whose coefficients motivate the power-series F and the fast-diffusion coefficients."},{"cited_title":"Gonçalves, C","cited_arxiv_id":null,"evidence_quote":"Supplies the kinetic constraint rates r^{(k)} for the nearest-neighbor porous medium model underlying the multidimensional definitions."},{"cited_title":"Hydrodynamical behavior for the generalized symmetric exclusion with open boundary","cited_arxiv_id":"2201.10241","evidence_quote":"Provides the product-form generalized exclusion setup with N_e > 1 and entropy estimates relative to non-constant profiles."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the entropy method, the overall convergence strategy adapted here."},{"cited_title":"Kipnis and C","cited_arxiv_id":null,"evidence_quote":"Supplies the tightness and martingale characterization framework for hydrodynamic limits, adapted by the authors to infinite volume."}],"review_version":1}