{"id":"50d03d5e-38a3-40ba-8e13-3b0dfb372398","arxiv_id":"2607.29373","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Integrability of a gl-regular hydrodynamic-type system forces the Haantjes tensor of its operator and of every symmetry to vanish locally.","lead":"The paper proves the Bolsinov–Konyaev–Matveev conjecture: integrable hydrodynamic-type systems whose defining operator is gl-regular must have vanishing Haantjes torsion. This gives a local canonical description of such integrable systems as regular F-systems.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Load-bearing reliance on Theorem 3.1 from unpublished preprint [3]; if its proof uses F-manifold structure beyond cyclicity, Lemma 4.1 and the main theorem fail.","rationale":"The reader's weakest_assumption correctly identifies the dependency on Theorems 3.1 and 3.3 from [3]. My stress-test narrows this: the truly load-bearing external step is Theorem 3.1, since Theorem 3.3's characterization of commuting flows is only used for interpretation, while the polynomial expansion part is proved in the paper and the criterion (3.2) comes from the published reference [19]. The sketch of Theorem 3.1 is plausible and the dimension count/injectivity argument appears correct, so I do not see an internal inconsistency. However, the proof is not self-contained and rests on an unpublished preprint that may contain hidden F-manifold assumptions. This is a legitimate concern but not, on the evidence, a fatal flaw: the authors have provided enough of the argument to suggest the generalization is valid, and the reader's ACCEPT with moderate confidence is reasonable. Hence I recommend no change to the verdict, while suggesting a concrete check to close the gap.","tokens_in":16645,"tokens_out":25441,"duration_ms":243278,"concrete_test":"Verify Theorem 3.1 directly in a non-F-manifold setting: take n=3, let A be a constant nilpotent Jordan block (e.g., A^3=0, A^2≠0) and v a constant cyclic vector, on a coordinate chart with the flat connection. Explicitly construct the natural connection by solving the linear equations ∇v=0 and d∇A=0, and check existence and uniqueness. If the construction works here, or if the proof in [3] is checked to use only cyclicity, the generalization is sound. Otherwise, Lemma 4.1 fails and the conjecture proof collapses.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central proof of Theorem 4.3 depends crucially on Theorem 3.1, which asserts the existence and uniqueness of a torsionless connection ∇(A,v) with ∇v=0 and d∇A=0 for an arbitrary cyclic pair (A,v). This is the step that yields d∇A=0, which is then used to derive d∇K_j=0 in Lemma 4.1 and to evaluate d∇K̃ in Theorem 4.3. The proof of Theorem 3.1 is only sketched in this paper: it reduces the problem to a fibrewise isomorphism M_A:E→Λ²T*⊗TM and gives an injectivity argument using the cyclic vector v. The full justification is relegated to the authors' unpublished preprint [3], where the theorem was originally proved for cyclic F-manifolds (with A=X∘ and v=e, the unit). If the proof in [3] inadvertently uses the product structure, the associativity, the Hertling-Manin condition, or the special role of the unit e, then the generalization to an arbitrary cyclic pair (A,v) may fail, breaking Lemma 4.1 and the entire proof of the conjecture. By contrast, the other imported ingredient—Theorem 3.3's equivalence (i)⇔(ii)—is less load-bearing: the 'Moreover' part (polynomial expansion of B) is proved in this paper using Lemma 2.4, and the commutativity criterion (3.2) is cited to the published [19], not to [3]. Thus the single most critical external dependency is Theorem 3.1.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves the Bolsinov-Konyaev-Matveev conjecture: if K_1,...,K_n are mutual symmetries of hydrodynamic-type systems, linearly independent at every point, and A = Σ c_i K_i is gl-regular at a point p, then the Haantjes tensor of A vanishes on a neighbourhood of p, and consequently the Haantjes tensor of every K_i vanishes there. The proof is built around a natural torsionless connection ∇ for a cyclic pair (A,v) (Theorem 3.1), a flow-commutation criterion (Theorem 3.3), and a sequence of algebraic lemmas: Lemma 4.1 upgrades commutativity of the K_i to d∇K_j=0, Lemma 4.10 selects at each point a constant-coefficient combination K̃ with K̃_p=A_p^2, and Lemma 4.9 shows that the resulting expression for (N_A)_p has vanishing Haantjes-ization. The n=2 case is handled separately. With the companion paper [8], this yields the structural conclusion that integrable gl-regular systems are regular F-systems near algebraically generic points.","tokens_in":16975,"tokens_out":17213,"duration_ms":157153,"significance":"If correct, the paper settles a conjecture that had previously been verified only in semisimple and special Jordan-block cases. The proof is clear and largely self-contained: the cyclic-vector mechanism (Lemma 2.5) and the algebraic cancellation lemma (Lemma 4.9) are elegant and presented in full detail. I specifically examined the stress-test concern about Theorem 3.1, which is imported from the authors' unpublished preprint [3]. In the present paper, the proof of Theorem 3.1 is summarized but is actually complete: it reduces the problem to the linear map M_A, proves injectivity by the same argument as Lemma 2.5, and uses the dimension count to conclude surjectivity. Thus the dependence on [3] is not a load-bearing gap in the argument. The flow-commutation criterion of Theorem 3.3 is likewise supported by the published reference [19] and by a proof sketch. The result is significant for the theory of integrable hydrodynamic-type systems.","major_comments":[],"minor_comments":[{"comment":"The proof is introduced as 'the proof given in [3]' and 'summarized'. Although the summary is mathematically complete, the journal version should either present it as a full proof in this paper or give a published/DOI reference for [3]; otherwise a reader without access to the preprint may not recognize that the 'algebraic fact' is fully established here.","section":"Section 3, Theorem 3.1"},{"comment":"There are several typos: 'existance' and 'existnce' in the concluding comments; 'complemnent' in the paragraph after Theorem 3.3 (should be 'complement'); 'complement' is used elsewhere where 'complement' is meant. These should be corrected.","section":"Concluding comments and Section 2"},{"comment":"The proof is dense; a short sentence explaining that N_A, as a vector-valued 2-form on a 2-manifold, is necessarily of the form ω⊗N, would improve readability. The pointwise identities for ω(AY,AZ) and ω(AY,Z)+ω(Y,AZ) are correct but could be stated as such.","section":"Section 4, Observation 4.4"},{"comment":"References [3] and [7] appear as arXiv preprints / journal articles dated 2026. Please update their publication status (accepted, DOI, or arXiv version) at the proof stage, since the text relies on [3] for two structural theorems.","section":"References"}],"recommendation":"minor_revision","confidential_remarks":"I found the central proof sound and the summaries of the imported theorems sufficient. The only editorial concern is the reliance on the authors' own unpublished preprint [3]; although the present paper contains enough of the argument, the editor may wish to confirm the status of [3] before publication. The paper fits the scope of the journal and, subject to the minor revisions above, is suitable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper genuinely proves the Bolsinov–Konyaev–Matveev conjecture, and the main argument is coherent. The new technique—using a cyclic pair and its natural connection to reduce the Haantjes tensor to a purely algebraic cancellation—is a real step beyond the partial cases in [7].\n\nWhat it does well: the structure is clear. Lemma 2.5 upgrades a partial vanishing condition to full vanishing using cyclicity; Lemma 4.9 is a purely algebraic cancellation that holds for any operator; Lemma 4.10 selects a constant-coefficient combination of the symmetries that equals A^2 at a single point. The proof of Theorem 4.3 then runs smoothly. I don't see a hidden fitting step or circularity. The dependence on the authors' own [3] for the natural connection is genuine, but it is not circular—it is an independent prior result about cyclic F-manifolds, and the sketch in Section 3 makes a plausible case that only cyclicity is used. The stress-test worry about F-manifold structure is worth taking seriously, but it is not a demonstrated flaw.\n\nSoft spots: Theorem 3.1 is only sketched, and it is load-bearing. A referee should read [3] and confirm that the proof does not secretly use the product, the unit, or the Hertling–Manin condition. The polynomial identity from [5] used in Proposition 4.2 is quoted without proof; presumably correct but worth checking. There are small typos (\"existnce\", \"complemnt\"), and the exposition is dense in places, but those don't affect the math.\n\nWho this is for: people working on integrable systems of hydrodynamic type, Haantjes and Nijenhuis geometry, or F-manifolds. The consequence that gl-regular integrable systems coincide with regular F-systems satisfying (1.7) is significant if the cited [8] holds.\n\nRecommendation: send to a serious referee. The referee should verify Theorem 3.1 in [3] and the cited identity, but the paper is honest, clear, and likely correct. I would not desk-reject it.","headline":"A genuine proof of the BKM conjecture, with the main argument sound; the only real risk is the borrowed natural-connection theorem from the authors' unpublished preprint.","tokens_in":17464,"tokens_out":1852,"would_cite":true,"duration_ms":18363,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37K10","35L60","53B05"],"pacs":[],"model":"deepseek-v4-flash","headline":"Integrability of a gl-regular hydrodynamic system forces the Haantjes tensor of the operator and all its symmetries to vanish locally.","keywords":["Haantjes tensor","Haantjes torsion","integrable systems of hydrodynamic type","gl-regular operator","cyclic pair","natural connection","Nijenhuis torsion","F-manifolds"],"falsifier":"Generate, with a computer algebra system, a 3×3 operator field A that is gl-regular at a point p and a commuting operator field B with B_p = A^2_p such that the flows of A and B satisfy the commutativity criterion of Theorem 3.3 (e.g., by imposing [A,B]=0 and d_∇B=0 for the natural connection). Compute the Haantjes tensor H_A at p. The paper's Corollary 4.11 predicts H_A|_p = 0; a single nonzero component would refute the proof.","tokens_in":16504,"feed_emoji":"🌊","tokens_out":9021,"duration_ms":76120,"temperature":0.7,"pith_summary":"The paper proves a conjecture about quasilinear first-order systems of PDEs, u_t = A(u) u_x, where A is a (1,1)-tensor field. The conjecture says that if such a system has n mutually commuting symmetries K_1, ..., K_n that are linearly independent at every point, and if some constant-coefficient linear combination A = Σ c_i K_i is gl-regular at a point p (meaning the pair (A,v) admits a cyclic vector near p), then the Haantjes tensor of A — and of every K_i — vanishes in a neighbourhood of p. The proof builds a canonical torsionless connection associated with the cyclic pair (A,v), converts flow commutativity into the covariant-linear condition d∇K_j = 0, and then shows that the remaining algebraic part of the Nijenhuis tensor at any point is of a form that is annihilated by the operation producing the Haantjes tensor. As a corollary, at an algebraically generic point every integrable gl-regular system is locally a regular F-system, i.e., of the form u_t = X(u) ◦ u_x with a commutative associative product satisfying the Hertling-Manin conditions.","feed_headline":"Integrable hydrodynamic systems must have vanishing Haantjes torsion","feed_subtitle":"New proof shows cyclic integrability kills the Haantjes tensor of every symmetry, unifying local geometry of such systems.","key_machinery":"The natural torsionless connection ∇^{(A,v)} of a cyclic pair (A,v), characterized by ∇v = 0 and d_∇A = 0, together with the algebraic cancellation lemma stating that any tensor N(Y,Z) = Σ_k (α_k(Y) A^k Z − α_k(Z) A^k Y) has zero Haantjes-ization. The connection converts the flow-commutativity condition into the pointwise equation d_∇B = 0, and the cancellation lemma removes the coefficient part of d_∇K̃, leaving the Nijenhuis tensor at a point to be of a form that cannot contribute to the Haantjes tensor.","core_discovery":"The central discovery is Theorem 4.3: let K_1,...,K_n be mutual symmetries on an n-manifold, linearly independent at each point, and let A = Σ c_i K_i be a constant-coefficient linear combination that is gl-regular at p. Then H_A, the Haantjes tensor of A, vanishes on an open neighbourhood of p. Since each K_j is shown to be a polynomial in A with functional coefficients, the vanishing of H_A implies the vanishing of H_{K_j} (Proposition 4.2), proving the conjecture. The proof relies on the existence, for any cyclic pair (A,v), of a unique torsionless connection ∇ with ∇v = 0 and d_∇ A = 0. For such a connection, commutativity of the flows of A and B is equivalent to [A,B]=0 and d_∇ B=0. The","pith_inferences":["The natural connection of a cyclic pair may be useful beyond this proof: it gives a canonical way to test integrability of a single operator field by checking d_∇B = 0 for candidate symmetries.","The algebraic cancellation Lemma 4.9 suggests a broader principle: any operator field whose Nijenhuis tensor at a point is spanned by polynomial-commutator terms automatically has zero Haantjes tensor there; this might provide a route to classify non-gl-regular cases by approximation.","Corollary 4.11 isolates a surprisingly minimal test: if a symmetry B of A satisfies B_p = A^2_p at a gl-regular point p, then H_A vanishes at p; running this check on random pairs might reveal whether the phenomenon is specific to mutual-symmetry systems or holds more generally."],"forward_implications":["The gl-regular integrability conjecture is proven: for integrable gl-regular hydrodynamic systems, H_A = 0 and H_{K_i} = 0 on the region of gl-regularity.","In a neighbourhood of an algebraically generic point, integrable gl-regular systems of hydrodynamic type coincide with regular F-systems satisfying integrability condition (1.7).","The local statement is strong: gl-regularity of A at a single point p implies vanishing of H_A on a whole neighbourhood of p; if the gl-regular locus is dense, vanishing holds on all of M.","In dimension 2 the conjecture is vacuous because every operator field is Haantjes; the proof handles n ≥ 3 and does not require algebraic genericity or diagonalizability."],"fun_headline_variants":["Proof: integrable hydrodynamic systems have zero Haantjes torsion","Conjecture proven: Haantjes tensor vanishes for integrable systems","New proof: cyclic integrability kills Haantjes torsion","Haantjes torsion vanishes for integrable hydrodynamic systems"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof leans on Theorem 3.1 and Theorem 3.3 — the existence and uniqueness of the natural torsionless connection for a cyclic pair and the equivalence between commuting flows and [A,B]=0 with d_∇B=0 — which are taken from the authors' earlier paper and only summarized; if those theorems fail outside the F-manifold setting, the argument collapses.","fun_headline_variants_meta":{"raw":{"variants":["Proof: integrable hydrodynamic systems have zero Haantjes torsion","Conjecture proven: Haantjes tensor vanishes for integrable systems","New proof: cyclic integrability kills Haantjes torsion","Haantjes torsion vanishes for integrable hydrodynamic systems"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00015,"raw_usage":{"total_tokens":1029,"prompt_tokens":735,"completion_tokens":294,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":479,"completion_tokens_details":{"reasoning_tokens":224}},"tokens_in":479,"tokens_out":294,"duration_ms":3661,"temperature":1.0,"reasoning_tokens":224,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T08:17:06.087449+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Generate, with a computer algebra system, a 3×3 operator field A that is gl-regular at a point p and a commuting operator field B with B_p = A^2_p such that the flows of A and B satisfy the commutativity criterion of Theorem 3.3 (e.g., by imposing [A,B]=0 and d_∇B=0 for the natural connection). Compute the Haantjes tensor H_A at p. The paper's Corollary 4.11 predicts H_A|_p = 0; a single nonzero component would refute the proof.","supporting_citations":[],"review_version":1}