{"id":"ef7e9128-ee1d-407d-af1b-7de207cb5eb3","arxiv_id":"2509.06610","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"The paper introduces a Fokker-Planck model whose polynomial drift coefficients are fixed by coupled moment-matching and Fisher-entropy constraints, and shows improved shock-structure agreement with DSMC.","lead":"Rarefied-gas simulations can be sped up by replacing costly binary collisions with a Fokker-Planck drift-diffusion process, but only if the model keeps both the right transport moments and the right entropy behavior. This paper proposes such a model with a Fisher-information entropy constraint and reports shock profiles closer to DSMC than the current cubic-drift Fokker-Planck model.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 4.1's entropy constraint uses a 1D Laplacian factor (2k+2)(2k+1), while in R^3 the radial Laplacian gives (2k+2)(2k+3); the stated cancellation in Eq. (50) therefore does not hold and the Fisher entropy decay (26) does not follow.","rationale":"The reader's weakest_assumption is exactly the load-bearing flaw I identify: Theorem 4.1 is the bridge from the drift ansatz to the H-theorem, and Eq. (25) uses the 1D identity for a 3D Laplacian. I verified the radial Laplacian calculation in R^3 and the consequence for Eq. (50); the cancellation does not go through. I also checked the two other concerns mentioned in the reader's rationale. Theorem 4.3's proof indeed treats \\bar R^{-1} as if it were the corresponding subblock of R^{-1}, which is generally false, and Eq. (47) equates two sums that are equal only near isotropy. These additional flaws strengthen the case, but they are secondary to the entropy-constraint error because the model's central identity is the Fisher decay (26). The numerical shock results are encouraging, but they do not compensate for the unsupported theoretical mechanism, since the improved shock profile is attributed to the entropy structure in Sections 1.2 and 7. Since my reading confirms the reader's rejection rather than introducing a different verdict, I leave the verdict unchanged.","tokens_in":22105,"tokens_out":15264,"duration_ms":125692,"concrete_test":"Independently re-derive the cancellation in Theorem 4.1 using the 3D radial Laplacian: compute Δ_{v,v}||v'||^d_2 for d=2k+2 as (2k+2)(2k+3)||v'||^{2k}_2, substitute into Eq. (50), and check whether the AHO and stabilization terms cancel under the stated Eq. (25). If the residue (2k+2)(2k+3)-(2k+2)(2k+1)=4(k+1) times c_d⟨||v'||^{2k}_2, f⟩ remains, the Fisher entropy decay (26) fails as stated. Then rebuild the linear system in Sec. 4.4 with the corrected factor and verify whether the resulting coefficient matrix L remains invertible and whether h[f] in Eq. (41) must be rescaled.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the FE-FP drift can simultaneously satisfy weak moment consistency and the H-theorem with the Fisher entropy decay rate ∂H/∂t = -D[f]I(f|f0). The mechanism is Theorem 4.1: Eq. (25) imposes sum_α c_α ⟨Δ_{v,v}H_α, f⟩ = (2k+2)(2k+1)c_d ⟨||v'||^{2k}_2, f⟩, which is used in the proof of Sec. 5.1 to cancel the AHO divergence -⟨∇·A_HO, f⟩ against the stabilization term c_d⟨Δ_{v,v}||v'||^d_2, f⟩ in Eq. (50). But the velocity space is R^3 (Section 2.1), where Δ r^d = d(d+1)r^{d-2}. Since d = 2k+2, the stabilization term is (2k+2)(2k+3)c_d⟨||v'||^{2k}_2, f⟩, not (2k+2)(2k+1)c_d⟨...⟩. The factor (2k+1) is the one-dimensional radial value d(d-1). With the constraint as stated, Eq. (50) leaves an uncancelled positive term 2(2k+2)c_d⟨||v'||^{2k}_2, f⟩, so ∂H/∂t = -D[f]I(f|f0) does not follow. The linear-system construction in Sec. 4.4 is built on this constraint, so the inconsistency propagates to the coefficient solve. I am not claiming the idea is irreparable; replacing the factor with (2k+2)(2k+3) and re-deriving the coefficient system might restore the cancellation. But as written, the H-theorem/fisher-decay part of the central claim is unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces the Fisher Entropic Fokker-Planck (FE-FP) model for monatomic rarefied gases. The model uses a drift of the form A = -v'/tau + grad Phi - c_d grad ||v'||_2^d with a polynomial potential Phi, together with a velocity-independent diffusion D = theta/tau. The central claims are: (i) a Fisher-information-based entropy constraint (Theorem 4.1) yields the entropy decay rate partial H / partial t = -D I(f|f0); (ii) the associated linear systems for the drift coefficients admit unique solutions (Theorem 4.3); (iii) the corresponding velocity SDE is well-posed (Theorem 4.4); and (iv) the model reproduces DSMC shock profiles more accurately than the cubic-drift FP model. The paper also derives a time scale tau = 2 mu / p by matching Boltzmann entropy decay for anisotropic Gaussians. Numerical experiments on supersonic flow over a flat plate show close agreement with DSMC, and the authors provide a reproducibility link.","tokens_in":1427,"tokens_out":3411,"duration_ms":193026,"significance":"If the theoretical claims were correct, the FE-FP model would be a significant contribution: it would combine weak moment consistency with a controlled entropy-dissipation structure, two properties that are usually difficult to satisfy simultaneously in Fokker-Planck approximations of the Boltzmann equation. The numerical results are encouraging and the paper is clearly written, with a useful comparative study against DSMC and the cubic-drift FP model. The availability of code and data is a strength. However, the central theoretical results are not supported as written: the entropy constraint contains an incorrect Laplacian factor, the linear-system right-hand side is inconsistent with the stated constraint, and the invertibility proof of the coefficient matrix contains algebraic errors. These issues affect the foundation of the model, not merely its presentation.","major_comments":[{"comment":"The entropy constraint uses the factor (2k+2)(2k+1) for the Laplacian of ||v'||_2^d, but the velocity space is R^3 throughout (Sec. 2.1). In R^3, the radial Laplacian of r^d satisfies Delta r^d = d(d+1) r^{d-2} = (2k+2)(2k+3) r^{2k} for d = 2k+2. With the stated factor, the cancellation in Eq. (50) is incomplete: an uncancelled positive term 2(2k+2) c_d <||v'||_2^{2k}, f> remains. Consequently the Fisher decay (26) does not follow, and the H-theorem in the claimed form is not established.","section":"§4.2, Eq. (25)"},{"comment":"The right-hand side h[f] of the entropy row is set to -c_d <||v'||_2^{2k}, f>. However, to cancel the AHO divergence against the stabilization term in Eq. (50), the entropy constraint must enforce <sum_alpha c_alpha Delta H_alpha, f> = c_d <Delta_{v,v} ||v'||_2^d, f> = (2k+2)(2k+3) c_d <||v'||_2^{2k}, f>. Even ignoring the factor error in Eq. (25), the sign in (41) is opposite to what the proof requires. The assembled linear system therefore does not impose the Fisher entropy decay.","section":"§4.4, Eq. (41)"},{"comment":"The proof of Theorem 4.3 contains invalid algebraic steps. The displayed identity R^{-1}_{phi phi} = <grad ||v'||_2^{2k} dot grad ||v'||_2^{2k}, f> confuses a diagonal entry of the inverse matrix with the corresponding entry of R. Moreover, the reduction of the Schur complement to -R^{-1}_{phi phi} Q_phi^2 is not valid for a general symmetric positive definite R; it holds only for special block structures, for example diagonal R. A simple 2x2 Schur-complement calculation shows that the determinant of L can vanish for admissible R and Q, so the unconditional invertibility claim is not justified by the given proof.","section":"§5.2, Eqs. (52)-(55)"}],"minor_comments":[{"comment":"The expression for S contains index errors: the sum over alpha,beta in the last term of Eq. (39) should involve only the appropriate index, and the subsequent expansion in Eq. (53) mixes indices inconsistently. Please correct the notation.","section":"§4.4, Eq. (39)"},{"comment":"The moment-consistency equation is written with the sum over beta in I_r^3, although the potential Phi in Eq. (20) is expanded over the augmented set I_{2k}^3. State explicitly how the phi component is subsequently incorporated via the relation c = R^{-1}(Q+G) to avoid confusion.","section":"§4.3, Eq. (29)"},{"comment":"The word 'Lyupanov' should be 'Lyapunov'.","section":"§5.3"},{"comment":"Several references have corrupted author names and formatting artifacts (e.g., [12] 'T. Ew art', [27] 'Helf and', [31] 'R. F. Pa wula'). These should be cleaned in revision.","section":"References"}],"recommendation":"reject","confidential_remarks":"The numerical experiments are promising and the reproducibility statement is a positive feature, but the theoretical core of the paper - Theorem 4.1 and Theorem 4.3 - is not reliable as written. The errors are not merely typographical: the entropy constraint is dimensionally inconsistent with the 3D velocity space, the linear-system right-hand side does not match the claimed constraint, and the invertibility proof fails on its own algebra. A substantially revised manuscript that corrects these issues could be reconsidered."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The Fisher Entropic Fokker–Planck model is a genuinely interesting construction: it augments a moment-matched polynomial drift with a stabilizing term and forces an entropy constraint through a linear system on the drift coefficients. The numerical shock profiles agree with DSMC and improve on cubic-drift FP, and the reproducibility package is a plus. If the theory were solid, this would be a useful step toward cheaper rarefied-gas simulation.\n\nThe problem is that the central theorem, as written, does not hold. Theorem 4.1 relies on the cancellation in Eq. (50), which requires Eq. (25) to be the correct Laplacian identity for ||v'||^d in the 3D velocity space. The paper uses (2k+2)(2k+1), the 1D radial value. In R^3 the Laplacian of r^d is d(d+1)r^{d-2}, i.e. (2k+2)(2k+3)r^{2k}. With the stated factor, the AHO divergence and the stabilization term do not cancel; an uncancelled positive term remains, so the Fisher decay rate (26) does not follow. This is load-bearing, not a typo, because the coefficient solve in Sec. 4.4 is built on that constraint.\n\nThe other soft spots are less severe. The time-scale derivation in Sec. 4.5 equates two sums that are not generally equal; τ = 2μ/p only holds near isotropy, so the equality is overclaimed. The invertibility proof in Sec. 5.2 has ugly index typos in the Schur-complement expansion, but the algebra appears salvageable once the notation is cleaned up; I would not call that a fatal flaw. And the H-theorem is imposed by construction rather than derived from first principles—that is a modeling choice, not an error, but the paper should avoid implying it is an independent check.\n\nWho should read this? Kinetic-theory and rarefied-gas researchers who want an FP alternative to DSMC with controlled entropy behavior. The numerical validation is real, but the theoretical foundation needs repair before the method can be trusted in the way the paper claims.\n\nVerdict: engage with it, but do not accept as is. A serious referee should be sent in; the Laplacian factor is fixable (replace (2k+1) with (2k+3) and re-derive the coefficient system), and the numerics and code make the paper worth the effort. My own citation of this work would wait until the corrected version appears.","headline":"Promising FP scheme with a load-bearing Laplacian factor error that undermines the main entropy-decay theorem; worth refereeing but not as is.","tokens_in":23079,"tokens_out":4399,"would_cite":false,"duration_ms":35244,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["76P05","82C40","65C30","35Q84"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper constructs a Fokker-Planck model of rarefied gases that simultaneously enforces Boltzmann moment matching and a Fisher-information entropy decay rate, and validates it against DSMC shock profiles.","keywords":["Fokker-Planck kinetic model","Boltzmann equation","H-theorem","Fisher information","rarefied gas dynamics","moment matching","shock profiles","particle stochastic methods"],"falsifier":"Compute $\\Delta_{v,v}\\|v'\\|_2^d$ in $\\mathbb{R}^3$ for $d=2k+2$ and compare it with $(2k+2)(2k+1)\\|v'\\|_2^{2k}$; equivalently, simulate homogeneous relaxation with FE-FP coefficients and compare the measured $dH/dt$ against $-D I(f|f_0)$. A mismatch in either test would falsify Theorem 4.1's entropy decay claim.","tokens_in":21888,"feed_emoji":"⚛️","tokens_out":10479,"duration_ms":82451,"temperature":0.7,"pith_summary":"Simulating rarefied gas flows near the continuum limit is expensive because direct particle methods must resolve enormous collision counts; replacing collisions by a cheap Fokker-Planck drift-diffusion process works only if the model reproduces Boltzmann moment relaxation and dissipates entropy correctly. This paper claims a new Fisher Entropic Fokker-Planck (FE-FP) model that satisfies both constraints at once: weak moment consistency up to the heat-fluxes, and entropy decay $\\partial H/\\partial t = -D\\, I(f|f_0)$ with the Fisher information as the rate. The drift coefficients come from two linear systems that the paper proves uniquely solvable whenever temperature is positive. Shock simulations at Mach 4 through 7 match DSMC benchmark profiles and improve visibly on the cubic-drift FP model, whose over-diffusive shock structures worsen with Mach number. The reason to care is that this removes the main obstacle to using Fokker-Planck models as efficient, entropy-consistent surrogates for the Boltzmann equation.","feed_headline":"Fokker-Planck model matches Boltzmann entropy decay","feed_subtitle":"Fisher-information entropy rate fixes over-diffusive shock profiles, matching DSMC from Mach 4 to 7.","key_machinery":"The load-bearing object is the Fisher entropic constraint, a linear equality on the drift coefficients that makes the higher-order drift divergence cancel against the stabilization term in the entropy balance, leaving only $-\\frac{\\theta}{\\tau}I(f|f_0)$. It is imposed on the drift ansatz $A = -\\frac{1}{\\tau}v' + \\nabla_v\\Phi^{(r)} - c_d\\nabla_v\\|v'\\|_2^d$, where $\\Phi^{(r)}$ is a polynomial potential built from the augmented moment set. The coefficient construction splits into two linear systems: $\\hat{c} = R^{-1}(Q+G)$ for the normalized high-order coefficients and $c' = L^{-1}b$ for the moment- and entropy-constrained part; the paper proves $R$ is symmetric positive definite and uses a Schur complement argument to show $L$ is invertible for positive temperature. This split is what lets moment matching and entropy dissipation coexist in a single tractable system.","core_discovery":"The central claim is the construction of a Fokker-Planck collision operator whose drift is the gradient of a polynomial potential plus the linear relaxation term $-v'/\\tau$ and a stabilizing term $-c_d\\nabla_v\\|v'\\|_2^d$, with diffusion fixed at $\\theta/\\tau$. Choosing the coefficients so that $\\langle\\sum_{\\alpha} c_\\alpha \\Delta_{v,v}H_\\alpha, f\\rangle = (2k+2)(2k+1)c_d\\langle\\|v'\\|_2^{2k}, f\\rangle$ forces the entropy balance to collapse to $\\partial H/\\partial t = -D\\, I(f|f_0)$, the Fisher-information decay rate. The same coefficients are required to match the Boltzmann production terms for moments up to the heat-fluxes, and the paper proves the resulting linear systems (36) and (37) have a unique solution whenever $T>0$. The time scale is fixed by matching this decay against the Boltzmann entropy decay for anisotropic Gaussian distributions, recovering $\\tau = 2\\mu/p$. Numerical experiments on argon flow over a plate at Mach 4, 5, 6, and 7 show FE-FP shock and temperature profiles following the DSMC reference much more closely than the cubic-drift model, especially in the downstream region.","pith_inferences":["The constraint construction is really a recipe: pick any stabilization polynomial whose Laplacian lies in the span of the augmented test functions, and the Fisher decay rate is enforced by algebra; this suggests a general design principle for other drift ansätze.","A testable extension is to monitor the residual between the measured entropy decay and $-D I(f|f_0)$ during shock simulations; a persistent gap would reveal where moment truncation, rather than the entropy constraint, limits accuracy.","The same coefficient-splitting and Schur-complement technique should extend to anisotropic diffusion tensors, potentially producing entropy-stable ellipsoidal Fokker-Planck models with correct Prandtl number."],"forward_implications":["FP particle methods using the FE-FP operator retain DSMC-level shock accuracy while avoiding collision-frequency scaling, making near-continuum rarefied flows computationally feasible.","Because the entropy decay rate is a well-defined functional of $f$, the model inherits quantitative trend-to-equilibrium bounds rather than only a sign condition on the entropy.","The framework supports arbitrary moment-matching degrees of freedom, so it can be extended to higher-order closures, cascades of FP models, and a posteriori error estimation.","Correct heat-flux relaxation restores the Prandtl number in the hydrodynamic limit, fixing the main defect of linear-drift FP models.","The same moment-plus-entropy constrained SDE design transfers to diffusion-based generative modeling and Schrödinger bridge problems, as the paper notes."],"supporting_citations":[{"why":"Establishes the entropic Fokker-Planck framework and the empirical finding that Fisher-information entropy decay rate improves shock profiles; the direct predecessor this paper builds on.","marker":"[17]"},{"why":"Introduces the cubic-drift FP model and the split-step time integration scheme used here; serves as the main accuracy baseline in the shock tests.","marker":"[18]"},{"why":"Provides the linear-drift FP model with correct viscosity but wrong heat-flux relaxation, setting the moment-consistency agenda and the time scale $\\tau=2\\mu/p$.","marker":"[22]"},{"why":"Supplies an alternative entropy-consistent FP model with anisotropic diffusion, whose diffusion can become negative and motivates the need for a globally consistent construction.","marker":"[28]"},{"why":"The trend-to-equilibrium theorem that justifies closure of the moment system at arbitrary order, on which the weak-consistency equations rest.","marker":"[41]"},{"why":"Provides the convex Sobolev inequality link between Fisher-information entropy production and convergence to equilibrium for Fokker-Planck equations.","marker":"[4]"},{"why":"Defines the DSMC benchmark and the open-boundary treatment used in the validation; the reference solution the FE-FP profiles are compared with.","marker":"[7]"},{"why":"Supplies the automated hard-sphere Boltzmann production terms used as right-hand sides in the moment-matching system.","marker":"[20]"}],"fun_headline_variants":["Fisher entropy makes Fokker-Planck match Boltzmann's H-theorem","Entropic Fokker-Planck with Fisher info honors H-theorem","New FP model uses Fisher info to fix shock over-diffusion","Fokker-Planck with Fisher entropy matches DSMC shocks up to Mach 7","Fisher-information entropy rate fixes Fokker-Planck shock profiles"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entropy-decay theorem rests on a specific calculus identity for the stabilizing term; if that identity has the wrong constant in three-dimensional velocity space, the cancellation that produces the Fisher decay rate no longer goes through.","fun_headline_variants_meta":{"raw":{"variants":["Fisher entropy makes Fokker-Planck match Boltzmann's H-theorem","Entropic Fokker-Planck with Fisher info honors H-theorem","New FP model uses Fisher info to fix shock over-diffusion","Fokker-Planck with Fisher entropy matches DSMC shocks up to Mach 7","Fisher-information entropy rate fixes Fokker-Planck shock profiles"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001076,"raw_usage":{"total_tokens":4565,"prompt_tokens":1069,"completion_tokens":3496,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":685,"completion_tokens_details":{"reasoning_tokens":3395}},"tokens_in":685,"tokens_out":3496,"duration_ms":22164,"temperature":1.0,"reasoning_tokens":3395,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:16:50.972890+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $\\Delta_{v,v}\\|v'\\|_2^d$ in $\\mathbb{R}^3$ for $d=2k+2$ and compare it with $(2k+2)(2k+1)\\|v'\\|_2^{2k}$; equivalently, simulate homogeneous relaxation with FE-FP coefficients and compare the measured $dH/dt$ against $-D I(f|f_0)$. A mismatch in either test would falsify Theorem 4.1's entropy decay claim.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the entropic Fokker-Planck framework and the empirical finding that Fisher-information entropy decay rate improves shock profiles; the direct predecessor this paper builds on."},{"cited_title":"Jenny, M","cited_arxiv_id":null,"evidence_quote":"Provides the linear-drift FP model with correct viscosity but wrong heat-flux relaxation, setting the moment-consistency agenda and the time scale $\\tau=2\\mu/p$."},{"cited_title":"Mathiaud and L","cited_arxiv_id":null,"evidence_quote":"Supplies an alternative entropy-consistent FP model with anisotropic diffusion, whose diffusion can become negative and motivates the need for a globally consistent construction."},{"cited_title":"Truesdell,The Trend to Equilibrium According to the Kinetic Theory of Gases, Springer New York, New York, NY, 1984, pp","cited_arxiv_id":null,"evidence_quote":"The trend-to-equilibrium theorem that justifies closure of the moment system at arbitrary order, on which the weak-consistency equations rest."},{"cited_title":"Arnold, P","cited_arxiv_id":null,"evidence_quote":"Provides the convex Sobolev inequality link between Fisher-information entropy production and convergence to equilibrium for Fokker-Planck equations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the automated hard-sphere Boltzmann production terms used as right-hand sides in the moment-matching system."}],"review_version":1}