{"id":"e9473a60-e918-421b-9d4c-b4f3245ab650","arxiv_id":"2607.16817","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Forward ideal sub-Finslerian manifolds satisfy interpolation, Brunn-Minkowski and measure-contraction inequalities with distortion coefficients replacing the classical curvature terms.","lead":"This mathematics paper extends sub-Riemannian optimal-transport inequalities to sub-Finslerian manifolds, where distances can be asymmetric because the metric is defined only on a non-integrable distribution. It defines distortion coefficients that replace the classical curvature terms and proves Brunn-Minkowski-type inequalities, with an explicit computation on a Randers sub-Finslerian Heisenberg group.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The interpolation inequality hinges on the unproved positivity lemma (Prop. B.1); the sketch delegates the core matrix inequality to sub-Riemannian references, so Theorem 3.7 and hence Theorem 1.1 are conditional.","rationale":"The paper's main theorem is a chain: optimal transport theory on forward ideal sub-Finslerian manifolds (Theorem 4.5), the Jacobian estimate (Theorem 3.7), and explicit distortion coefficients (Lemma 5.3). The Jacobian estimate is the only step that converts geometric Jacobi information into the determinant inequality needed for the interpolation inequality. Its proof relies on Proposition B.1, whose matrix inequalities are exactly what allow Minkowski's determinant theorem. The appendix provides only a sketch and transfers results from sub-Riemannian references, asserting without proof that the arguments adapt to irreversible sub-Finslerian structures. This is a genuine load-bearing gap: the sub-Finslerian Hamiltonian is not quadratic and not reversible, so the positivity and symmetry properties of the relevant matrices are not immediate. The secondary s↑1 limit in Theorem 3.7 is not fatal if continuity of the Jacobi matrices is supplied, but it is another place where the proof is compressed. No fatal contradiction or circularity appears elsewhere; the explicit Randers Heisenberg computation provides independent support for the framework. The reader's CONDITIONAL verdict is therefore appropriate: the claims are plausible and likely correct, but the proof should be completed before full acceptance.","tokens_in":40570,"tokens_out":33966,"duration_ms":276934,"concrete_test":"Independently re-derive Proposition B.1(b) from the sub-Finslerian Jacobi equation (3.1), the nonnegativity of B(t), and the symplectic Wronskian identities; in particular, show that S(t)=N_0^V(t)^{-1}N_0^H(t) satisfies \\dot S=-N_0^V(t)^{-1}B(t)N_0^V(t)^{-T}≤0 and that this yields the required positive-semidefinite matrices. If the derivation requires the symmetry H(-λ)=H(λ), the irreversible case is not established; if it goes through without reversibility, the concern is reduced to a missing-detail issue rather than a mathematical failure. A numerical cross-check on the Randers Heisenberg example with a≠0 would further confirm the matrix inequalities along explicit geodesics.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central Jacobian estimate (Theorem 3.7) applies Minkowski's determinant theorem to the identity N(t)=N_s^V(t)N_s^V(0)^{-1}+N_0^V(t)N_0^V(s)^{-1}N(s). This requires Proposition B.1: det N_0^V(t)>0, N_0^V(t)^{-1}N_s^V(t)N_s^V(0)^{-1}≥0, and N_0^V(s)^{-1}N(s)≥0 (with symmetry). Appendix B labels this 'crucial' but gives only a sketch: (a) is delegated to a Riccati comparison argument, (b) to [9, Lemma 74] with the assertion that the argument survives irreversibility, and (c) to [17, Claim 2.4]. No full proof of these matrix inequalities in the sub-Finslerian, irreversible, non-quadratic Hamiltonian setting is supplied. If any one fails, the Minkowski step and the lower bound on det(d_xT_t) collapse; since Theorem 1.1 is derived by combining exactly that estimate with the Monge-Ampère equation (Theorem 4.7), the main interpolation inequality stands or falls with B.1. The secondary limit s↑1 in Theorem 3.7 is less serious: once B.1 holds on (0,1), continuity of N_s^V(t) in s and invertibility of N_1^V(t) for t<1 justify the divergence, but the paper should state this explicitly.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper extends Barilari–Rizzi sub-Riemannian interpolation inequalities to forward ideal sub-Finslerian manifolds, i.e., forward-complete sub-Finslerian structures with no nontrivial abnormal minimizers and possibly irreversible norms. The main result, Theorem 1.1, asserts a pointwise interpolation inequality for densities along Wasserstein geodesics, with generalized forward/backward distortion coefficients defined by volume-ratio limits. The proof develops sub-Finslerian Jacobi fields, a Jacobian estimate (Theorem 3.7), and an optimal transport theory culminating in the Brenier–McCann theorem and the Monge–Ampère equation (Theorem 4.7). Applications include Borell–Brascamp–Lieb, p-mean, Brunn–Minkowski, and MCP statements. In Section 6 the paper gives an explicit computation for Randers sub-Finslerian Heisenberg groups and proves an explicit MCP(0,N) exponent N(a)>5 for a≠0.","tokens_in":40899,"tokens_out":7638,"duration_ms":73700,"significance":"If the central Jacobian estimate is fully proved, the paper would be a substantial extension of the sub-Riemannian interpolation framework to a genuinely irreversible, non-quadratic setting, with a nontrivial Randers Heisenberg example and explicit MCP exponents. The paper's structure is coherent and the Randers Heisenberg computation in Appendix C is explicit and checkable; Proposition 6.3 provides a concrete, parameter-free bound on the curvature exponent. The main obstruction is that the crucial positivity lemma (Appendix B) is only sketched and delegates the core matrix inequalities to sub-Riemannian references, so the main theorem is currently conditional.","major_comments":[{"comment":"The proof of Theorem 3.7 reduces the Jacobian estimate to Proposition B.1: det N_0^V(t)>0, N_0^V(t)^{-1}N_s^V(t)N_s^V(0)^{-1}≥0, and N_0^V(t)^{-1}N(t)≥0, which are then used to apply Minkowski's determinant theorem to identity (3.3). Appendix B explicitly labels this lemma crucial, yet it supplies only a sketch: part (b) is transferred from [9, Lemma 74] and part (c) from [17, Claim 2.4], with assertions that the arguments survive in the sub-Finslerian setting. Because the Hamiltonian is non-quadratic and the metric is irreversible, these transfers are not automatic, and the nonnegativity/symmetry of the relevant matrix products is not verified. If any of these inequalities fails, the Minkowski step and the lower bound on det(d_xT_t) collapse, and Theorem 1.1 — which is derived from this estimate in §5.2 — is left unproved. A complete proof of Proposition B.1 must be supplied.","section":"Appendix B; Theorem 3.7"},{"comment":"In the contradiction argument showing γ(1) is not conjugate to γ(0), the paper states that for fixed t<1 the numerator det N_s^V(t) 'remains nonzero' as s↑1 while the denominator det N_s^V(0) tends to 0, concluding that the right-hand side diverges. Nonzero is not sufficient: unless det N_s^V(t) is bounded away from zero, or is shown to converge to a nonzero limit via continuity of N_s^V(t) in s and invertibility of N_1^V(t) for t<1, the claimed divergence is unjustified. This step is needed to extend (3.4) to s=1 and should be proved explicitly.","section":"Theorem 3.7, limit s↑1"}],"minor_comments":[{"comment":"The notation in the sketch of part (c) is ambiguous: 'e^{t\\vec H}_*(-c_s)(X(0))' should presumably read 'e^{t\\vec H}_* d^2_x(-c_s)(X(0))', and 'M^v_s', 'N^v_s' should be 'M_s^V', 'N_s^V'. The current notation makes the claimed computation hard to verify.","section":"Appendix B, near (B.1)"},{"comment":"The proof asserts that shrinking neighborhoods A_r→{x} gives m(Z_t(A_r,B))→m(Z_t(x,B)). This requires a continuity or measure-theoretic justification, especially because B is an arbitrary Borel set and the points in B chosen by geodesics may not have compact closure. Without this, the claimed equivalence between the Brunn–Minkowski inequality and MCP is not fully established.","section":"Theorem 5.8, (ii)⇒(iii)"},{"comment":"The title contains an erroneous spacing: 'INTERPOLA TION' should be 'INTERPOLATION'. Similar spacing issues occur in the running header.","section":"Title and abstract"},{"comment":"When Theorem 3.7 refers to [9, Lemma 29] for the positivity step, it should also state explicitly which hypotheses of Proposition B.1 are being invoked, since the proof currently points to Appendix B only after the positivity assertion is already used in the main text.","section":"Section 3.2"}],"recommendation":"major_revision","confidential_remarks":"The paper deserves a chance after a full proof of Proposition B.1 is supplied; the Randers Heisenberg calculation and the explicit MCP exponent are valuable and appear sound. I would not accept the current version because the main interpolation inequality rests on an unproved matrix-positivity lemma. The limit s↑1 step should also be clarified, though it is secondary."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper does real work. It transplants the Barilari–Rizzi interpolation machinery into irreversible sub-Finslerian geometry, where the distance is asymmetric and CD conditions fail. The new pieces are genuinely new: sub-Finslerian Jacobi fields, distortion coefficients defined as volume-ratio limits, a Brenier–McCann theorem on forward ideal sub-Finslerian manifolds, the interpolation inequality itself, and an explicit computation showing that the Randers sub-Finslerian Heisenberg group satisfies MCP(0, N) with an explicit N(a) > 5. The Randers computation in Appendix C is detailed and reproducible, not a black box. The paper is also honest about what it borrows and what remains open.\n\nThe soft spot is real and load-bearing. Proposition B.1 supplies the matrix positivity needed for Minkowski's determinant theorem in the Jacobian estimate (Theorem 3.7), and Theorem 1.1 stands on that estimate. But Appendix B is a sketch: (a) is asserted from a Riccati comparison, (b) is delegated to [9, Lemma 74] with the claim that the argument survives non-reversibility, and (c) is delegated to [17, Claim 2.4]. No full proof is given in the irreversible, non-quadratic Hamiltonian setting. That is not a fatal flaw on its face — the architecture is coherent and the transfer might well work — but it is exactly the kind of step that needs checking before the main theorem can be accepted as written. The limit s↑1 in Theorem 3.7 is a secondary issue; it is probably fixable by continuity and invertibility of N_1^V(t), but the paper should state the argument.\n\nThe rest of the paper is careful and the citation pattern is appropriate. The reliance on [9], [19], and [17] is explicit and not hidden; the genuinely new content is the sub-Finslerian distortion coefficients and the Randers Heisenberg example. I do not see circularity or fitted parameters.\n\nWho should read this? Anyone working on synthetic curvature in sub-Finslerian or sub-Riemannian geometry, and anyone who wants a concrete non-sub-Riemannian example of interpolation inequalities. The paper deserves a serious referee: the gap is addressable and the framework is valuable enough that referee time is warranted, even if the final version needs a full proof of B.1 and a clean argument for the s=1 limit.","headline":"A serious and valuable sub-Finslerian extension of Barilari–Rizzi, but the main theorem is conditional on a positivity lemma that is only sketched and delegated to sub-Riemannian references.","tokens_in":41399,"tokens_out":1307,"would_cite":true,"duration_ms":15258,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C17","49Q22","53C23","49J15"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that forward ideal sub-Finslerian manifolds support interpolation inequalities for optimal transport, extending the sub-Riemannian result to the sub-Finslerian setting.","keywords":["sub-Finslerian geometry","optimal transport","interpolation inequality","Jacobi fields","distortion coefficients","Brunn-Minkowski inequality","measure contraction property","Randers Heisenberg group"],"falsifier":"For a specific irreversible sub-Finslerian structure, such as the Randers sub-Finslerian Heisenberg group with a≠0, compute the Jacobian matrices N_0^V(t) and N_s^V(t) explicitly and check whether det(N_0^V(t)^{-1})>0 and N_0^V(t)^{-1}N_s^V(t)N_s^V(0)^{-1}≥0 hold for all relevant s,t; a single negative determinant or eigenvalue would invalidate the Minkowski step and violate the claimed inequality.","tokens_in":40436,"feed_emoji":"📐","tokens_out":5005,"duration_ms":42104,"temperature":0.7,"pith_summary":"The paper extends the sub-Riemannian interpolation inequality to sub-Finslerian manifolds, where the distance is allowed to be asymmetric (irreversible). It defines sub-Finslerian Jacobi fields and generalized distortion coefficients, and proves that along the Wasserstein geodesic between two absolutely continuous measures, the density at intermediate times is bounded below by a weighted combination of the endpoint densities. This yields Brunn-Minkowski and Borell-Brascamp-Lieb inequalities in this setting. A worked example, the Randers sub-Finslerian Heisenberg group, provides explicit distortion coefficients and establishes the measure contraction property.","feed_headline":"Sub-Finslerian manifolds support new optimal transport inequalities","feed_subtitle":"New distortion coefficients let Wasserstein densities obey a sharp bound, yielding Brunn-Minkowski and measure-contraction results.","key_machinery":"The argument rests on three connected pieces: (1) a Jacobi-field calculus for sub-Finslerian geodesics via the Hamiltonian flow, leading to a Jacobian estimate (Theorem 3.7) for the derivative of the optimal transport map; (2) the Minkowski determinant theorem applied to matrix identities linking vertical and horizontal Jacobi matrices; and (3) a sub-Finslerian version of the Brenier-McCann theorem that supplies the optimal transport map and the Wasserstein geodesic. The 'forward ideal' condition (no abnormal minimizing geodesics) provides the regularity needed for these tools to work.","core_discovery":"The central claim is Theorem 1.1: on a forward ideal sub-Finslerian manifold (forward complete, with no nontrivial abnormal minimizing geodesics) equipped with a smooth measure, the density along the unique Wasserstein geodesic satisfies 1/ρ_t(T_t(x))^{1/n} ≥ (β^>_t)^{1/n}/ρ_0(x)^{1/n} + (β^<_t)^{1/n}/ρ_1(T(x))^{1/n} for µ0-a.e. x, where β^> and β^< are forward/backward distortion coefficients defined as volume-ratio limits along sub-Finslerian geodesic flows. This places the sub-Finslerian setting in the same framework as metric measure-space interpolation inequalities, with the distortion coefficients encoding the geometry instead of a sectional curvature bound.","pith_inferences":["The hidden load-bearing step is the positivity of certain Jacobian matrix products (Proposition B.1), which the paper only sketches; if these inequalities fail for some irreversible sub-Finslerian structure, the Jacobian estimate and hence the main theorem would need revision.","The explicit Randers Heisenberg computation offers a ready test bed: one can numerically verify the positivity matrices and the distortion bounds for a≠0, providing independent evidence for the Jacobian estimate.","The forward/backward asymmetry in the distortion coefficients suggests that the effective 'geodesic dimension' may differ in the two directions, potentially leading to refined measure-contraction exponents.","A likely extension is to other Carnot-type sub-Finslerian groups, where explicit exponential-map computations could yield distortion coefficients and curvature exponents beyond the Heisenberg case."],"forward_implications":["The interpolation inequality holds for irreversible sub-Finslerian metrics, with asymmetry handled by distinct forward and backward distortion coefficients.","It directly yields Brunn-Minkowski and Borell-Brascamp-Lieb inequalities, and hence the Prékopa-Leindler inequality, for forward ideal sub-Finslerian manifolds.","For the Randers sub-Finslerian Heisenberg group, the paper computes explicit distortion coefficients and proves the measure contraction property MCP(0,N) with an explicit N>5 depending on the drift parameter.","The p-mean inequality (Corollary 5.5) gives a whole family of concentration and volume comparison statements that interpolate between the endpoint densities.","The result provides a concrete step toward synthetic curvature-dimension conditions in nonholonomic geometries, where classical CD conditions are known to fail."],"fun_headline_variants":["Sub-Finslerian Jacobi fields unlock transport inequalities","Measure contraction proven for Randers sub-Finslerian Heisenberg","Optimal transport meets sub-Finslerian geometry","Sharp interpolation on sub-Finslerian manifolds via distortion","New sub-Finslerian tools: Brunn-Minkowski and more"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The Jacobian estimate at the heart of the proof relies on a positivity statement for certain Jacobian matrices (Proposition B.1) that is only sketched and transferred from the sub-Riemannian case; if these matrix inequalities fail for irreversible sub-Finslerian structures, the main interpolation inequality collapses.","fun_headline_variants_meta":{"raw":{"variants":["Sub-Finslerian Jacobi fields unlock transport inequalities","Measure contraction proven for Randers sub-Finslerian Heisenberg","Optimal transport meets sub-Finslerian geometry","Sharp interpolation on sub-Finslerian manifolds via distortion","New sub-Finslerian tools: Brunn-Minkowski and more"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000271,"raw_usage":{"total_tokens":1453,"prompt_tokens":718,"completion_tokens":735,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":462,"completion_tokens_details":{"reasoning_tokens":648}},"tokens_in":462,"tokens_out":735,"duration_ms":6935,"temperature":1.0,"reasoning_tokens":648,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T19:50:31.957391+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For a specific irreversible sub-Finslerian structure, such as the Randers sub-Finslerian Heisenberg group with a≠0, compute the Jacobian matrices N_0^V(t) and N_s^V(t) explicitly and check whether det(N_0^V(t)^{-1})>0 and N_0^V(t)^{-1}N_s^V(t)N_s^V(0)^{-1}≥0 hold for all relevant s,t; a single negative determinant or eigenvalue would invalidate the Minkowski step and violate the claimed inequality.","supporting_citations":[],"review_version":1}