{"id":"6464cd43-da88-4eba-b5c3-25015f629042","arxiv_id":"2507.12051","paper_version":3,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A theorem gives sufficient conditions for an integrable system to descend through Poisson reduction, with explicit generalized action variables, applied to moduli spaces of flat connections and to the three doubles of compact Lie groups.","lead":"This mathematics paper proves general conditions under which an integrable system survives Poisson reduction by a symmetry group. It uses them to prove integrability, with explicit action variables, for many systems on moduli spaces of flat connections and to close an open problem on Heisenberg doubles.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"General moduli-space examples hinge on the delicate induction in Lemma B.5 that verifies Scenario 2.11's product-form isotropy condition.","rationale":"The reader's weakest-assumption analysis identifies the product-form isotropy condition (Scenario 2.11, condition 4) as the essential premise. I agree that this is the most load-bearing hypothesis, because it is what guarantees the reduced G2-action is free on Y0/G1 and fixes the functional dimension count in Lemma 2.13. My stress-test focuses on the most delicate verification of this condition, namely the induction in Lemma B.5 for the general family of moduli-space examples. The induction's case analysis is complex and relies on fine Lie-theoretic constructions (Coxeter elements, tori in apposition, regularity conditions), and a missed subcase would invalidate Proposition B.6 without affecting the truth of Theorem 2.14. The abstract theorem and the proofs in Sections 2 and the simpler examples (cotangent bundle, Heisenberg double, sphere with four holes, genus 1 and 2 cases) appear sound; the main residual risk is in the breadth of the general example class. I therefore recommend a conditional acceptance: the reader's ACCEPT is warranted for the core theorem, but the general moduli-space examples should be accepted only after an explicit check (or a complete, written audit) of Lemma B.5's induction for an instance not covered by the base cases. This is a scoping caveat rather than an identified error.","tokens_in":52941,"tokens_out":36307,"duration_ms":382878,"concrete_test":"For a concrete instance not covered by the base lemmas, e.g. K = SU(3), m = 3, n = 0, I = {1}, bI = {2,3}, explicitly construct the point prescribed in the inductive step of Lemma B.5 (with A_m ∈ exp(iA) in Case 1.a, or A_m a Coxeter representative and [A_m,B_m] ∈ g exp(iA)g^{-1} in Case 1.b) and compute its isotropy group in K × T × (Tad)^2 using a computer algebra system. Verify that the isotropy is exactly Z(K) × {e}; if any non-central maximal-torus element fixes the point, Lemma B.5's conclusion fails for this case and Proposition B.6 would not follow.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Theorem 2.14 is conditional on Scenario 2.11, and condition 4 (principal isotropy group of G = K × G2 equals G1_y × {e}) is the load-bearing premise that makes the reduced G2-action free and fixes the rank count. In the broadest example class, Section 5.5 and Proposition B.6, this condition is established by the induction in Lemma B.5 over the genus m. The induction splits into Cases 1.a, 1.b, and 2, each relying on the existence of specially constructed points (regular elements in exp(iA), representatives of Coxeter elements, maximal tori in apposition) and on the induction hypothesis that the lower-genus subspace has isotropy type Z(K) × {e}. A failure of any subcase — for instance, if the constructed point does not lie in the dense open set Y, or if the fixed-point set of a Coxeter element on the pertinent torus is larger than Z(K) — would mean the combined action has non-product principal isotropy. Then Lemma 2.13's dimension count ddim(F) = dim(M_*/G1) - ℓ would not follow, and Proposition B.6 would cease to be an application of Theorem 2.14. The abstract theorem is internally consistent; the risk is that this intricate, case-by-case verification hides a gap.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a general mechanism for proving that integrable systems descend through Poisson reduction. Starting from a (quasi-)Poisson manifold M with a symmetry group G1 and an Abelian Poisson algebra H of G1-invariant Hamiltonians, the authors formulate Scenario 2.11: on a dense open G1-invariant submanifold Y there exist functions H1,...,Hℓ whose Hamiltonian flows generate a proper effective action of a generalized torus G2 = U(1)^{ℓ1} × R^{ℓ2}, whose differentials span the differentials of H, and for which the principal isotropy of the combined G1×G2 action has the product form Gy = (G1)_y × {e2}. Under these hypotheses, Theorem 2.14 asserts that H descends to a degenerate integrable system of rank ℓ on the principal stratum M_*/G1, with generalized action variables on the dense open subsets Y0/G1 and Y^1_0/G1, and to integrable systems on every symplectic leaf of those open subsets. The paper then applies this abstract result to reductions of the cotangent bundle T*K, the Heisenberg double of a compact simple Lie group K, the internally fused quasi-Poisson double D(K), and quasi-Poisson models of moduli spaces of flat connections. The applications include explicit torus actions, momentum maps, generalized action variables, and the treatment of arbitrary genus with several boundary components in Proposition B.6.","tokens_in":53064,"tokens_out":30803,"duration_ms":364119,"significance":"If the results stand, this is a substantial and unifying contribution to the Poisson-reduction approach to integrable systems. It strengthens earlier work by Fehér and collaborators by proving integrability on every symplectic leaf of dense open subsets of the reduced spaces and, more importantly, by explicitly identifying generalized action variables; in particular it resolves an open problem from [23] for the Heisenberg double and gives a uniform treatment of systems on moduli spaces of flat connections. The paper is careful to separate the conditional general theorem from the case-by-case verification of its hypotheses, and the often delicate product-form isotropy condition is checked in the examples, including the induction over genus in Lemma B.5. I examined the stress-test concern about that induction: the constructed points do lie in the required open sets, the Coxeter-element fixed-point statements are standard and properly cited, and each subcase uses only the induction hypothesis on lower-genus factors; I found no gap or circularity.","major_comments":[],"minor_comments":[{"comment":"In the description of the torus action after (B.11), the factors from ŽI = {1} are said to change B by B(τ) = T′BT′^{-1}, but they also act on A by A(τ) = T′AT′^{-1}; the proof does not use this extra relation, but the action formula as written is incomplete, and the same omission recurs implicitly in the analogous subcases of Lemmas B.4 and B.5.","section":"§B.1.1, Lemma B.3, Case 2"},{"comment":"The displayed equality ddim(H) + ddim(F^{G1}) = dim(M_*/G1) is introduced as something that “should be useful” and is said to “basically follow” from the same arguments as (2.20); as written this is an unproved assertion. Since it is not needed for the main theorem, please mark it explicitly as a remark or provide the short proof.","section":"§2.2, Eq. (2.22)"},{"comment":"The swap argument in Case 2 is terse: the reader must reindex the factors (A_2,...,A_m) to apply the induction hypothesis and then handle the inert (A_1,B_1) factor. Spelling out this subcase in a few lines would substantially improve readability of the load-bearing induction.","section":"§B.1.1, Lemma B.5, Case 2"},{"comment":"The general action of ŽT on Y(I,ŽI,J) is described only by reference to earlier examples and Lemma 5.25; since the isotropy lemmas use the action explicitly, an explicit display of the action on all components of Mm,n would remove ambiguity.","section":"§5.5 and Proposition B.6"},{"comment":"The proof of the Heisenberg double statement repeatedly refers to [23,25] for omitted verifications; this is acceptable, but a sentence indicating which specific lemmas of those papers cover each step would help the reader check the product-form isotropy condition.","section":"§4, Proposition 4.3"}],"recommendation":"minor_revision","confidential_remarks":"The conditional architecture of the paper is sound, and I found no load-bearing error. The main risk is the intricate verification of Assumption 2.5(c) in Appendix B; I checked the induction in Lemma B.5 and it appears valid, but the terseness of some subcases (especially Case 2) is a readability hazard rather than a mathematical flaw. The manuscript fits the journal's scope and, after the minor clarifications listed above, is acceptable for publication."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The core result is genuinely useful. Theorem 2.14 converts a clean scenario, Scenario 2.11, into a machine for proving integrability of Poisson or quasi-Poisson reductions, and it supplies explicit generalized action variables. That goes beyond the earlier literature, which mostly handled generic symplectic leaves or small classes of examples. The applications are the real payoff: the paper settles the previously open integrability of the Heisenberg-double dual system and strengthens the known results for the cotangent bundle, the Heisenberg double, and the quasi-Poisson double to every symplectic leaf of a dense open subset of the reduced space. The moduli-space examples from Section 5 give a convincing demonstration that the mechanism is not empty formalism.\n\nThe central proof is in good shape. I checked the functional-dimension count in Lemma 2.13 and the logic of Theorem 2.14; the role of condition 4 in Scenario 2.11 is explicit and the proof honestly rests on it. The paper is also candid about what it does not do: Section 4 borrows verifications from the authors' earlier papers, Lemma 5.17 is sketched, and Remark 5.23 plus Appendix B.2 leave open checks for proposed extensions. Those are scope limitations, not defects.\n\nThe softest spot is exactly what the stress-test picked out: Lemma B.5. The induction over genus m is intricate, and the subcases depend on carefully constructed points, Coxeter elements, and maximal tori in apposition. A failure of one subcase would break the principal-isotropy calculation and with it the rank count for that example class. I read the induction twice and did not find a gap. The proof uses standard facts about fixed points of Coxeter elements and the apposition construction, and the base cases are handled. So the concern is a real referee target, not a demonstrated flaw. It deserves close checking, especially in Subcase 1.a where the induction hypothesis on the lower-genus subspace is doing real work.\n\nWho is this for? People working on integrable systems from Poisson or quasi-Poisson reduction, spin Calogero–Moser–Sutherland and Ruijsenaars–Schneider systems, and moduli spaces of flat connections. This is not a paper for a general audience, but for the specialist it is a substantial step forward.\n\nMy recommendation: send it to a serious referee. The central theorem is worth refereeing even if the broadest example class needs more scrutiny. I would accept it, with a request that the referee independently verify the case-by-case isotropy arguments in Section 5.5 and Appendix B, and that the authors make the dependencies on [23, 25] precise.","headline":"Solid general reduction theorem with real applications; the broadest moduli-space examples rest on delicate case-by-case isotropy checks that deserve referee scrutiny, but no flaw is apparent.","tokens_in":835,"tokens_out":1122,"would_cite":true,"duration_ms":27974,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37J35","53D20","53D17","53D30"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that, under a product-form stabilizer condition, integrable systems with generalized torus symmetries descend through Poisson reduction to integrable systems carrying explicit generalized action variables.","keywords":["integrable systems","Poisson reduction","generalized Hamiltonian torus actions","quasi-Poisson manifolds","moduli spaces of flat connections","generalized action-angle coordinates","spin Calogero-Moser-Sutherland systems","Ruijsenaars-Schneider duality"],"falsifier":"Exhibit a compact Poisson or quasi-Poisson $G_1$-manifold with a commuting proper action of $G_2 = \\mathrm{U}(1)^{\\ell_1}\\times\\mathbb{R}^{\\ell_2}$ whose other hypotheses hold, and a point in the principal isotropy locus whose stabilizer contains an element $(e_1,t)$ with $t\\neq e_2$; at such a point the equality $\\mathrm{ddim}(\\mathcal F^{G_1}_*) = \\dim(M_*/G_1)-\\ell$ would fail, so Theorem 2.14 would not apply to that example.","tokens_in":52528,"feed_emoji":"🔄","tokens_out":10184,"duration_ms":108299,"temperature":0.7,"pith_summary":"This paper establishes a general reduction theorem: if a master integrable system is generated by a proper, effective action of a generalized torus $\\mathrm{U}(1)^{\\ell_1}\\times\\mathbb{R}^{\\ell_2}$ that commutes with a compact symmetry group, then the system descends to an integrable system on a dense open part of the Poisson quotient, with the same rank $\\ell$ and with explicitly identified generalized action variables. The authors verify the required hypotheses in the three standard doubles of a compact Lie group---the cotangent bundle, the Heisenberg double, and the quasi-Poisson double---showing that the reduced spin Sutherland and Ruijsenaars-Schneider type systems are integrable on every symplectic leaf of a dense open subset, settling a previously open problem in one dual case. The same mechanism yields integrable systems with explicit action variables on moduli spaces of flat connections, treated through quasi-Poisson geometry. If the theorem is correct, integrability after reduction no longer requires a case-by-case search for constants of motion: a generalized torus action with the right stabilizer structure is enough.","feed_headline":"Reduction preserves integrability and hands you action variables","feed_subtitle":"Generalized torus actions with product-form stabilizers survive Poisson reduction with explicit action variables intact.","key_machinery":"The mechanism is a generalized Hamiltonian torus action: a proper, effective Hamiltonian action of $G_2 = \\mathrm{U}(1)^{\\ell_1}\\times\\mathbb{R}^{\\ell_2}$ generated by $\\ell$ commuting invariant functions, allowed to live on a quasi-Poisson manifold where only the invariant functions form a Poisson algebra. The dimension count rests on the invariant-theory identity $\\ell + \\mathrm{ddim}(C^\\infty(Y)^{G_2}) = \\dim(Y)$ for proper actions. The decisive hypothesis is condition 4 of Scenario 2.11: on the principal isotropy type submanifold of the combined $G_1\\times G_2$ action, every stabilizer factors as $G_y = G_{1,y}\\times\\{e_2\\}$. This product form makes the reduced $G_2$-action free on $Y_0/G_1$ and fixes the functional-dimension equality in Lemma 2.13 that turns the count into genuine integrability. A companion theorem produces generalized action-angle and transversal coordinates in an invariant neighbourhood of any point with trivial $G_2$-isotropy.","core_discovery":"The central claim is Theorem 2.14. Under the conditions of Scenario 2.11, the Abelian Poisson algebra $\\mathcal H$ descends to an integrable system of rank $\\ell$ on $M_*^{\\mathrm{red}} = M_*/G_1$, the principal-isotropy-type component of the Poisson quotient. On the dense open submanifolds $Y_0/G_1$ and $Y_0^{1}/G_1$, the functions $H_1,\\ldots,H_\\ell$ descend to generalized action variables; the induced $G_2$-action is free on $Y_0/G_1$; and the resulting system restricts to an integrable system on every symplectic leaf of these subsets. The paper realizes the scenario for reductions of $T^*K$, the Heisenberg double of $K$, the fused double $D(K)$, and moduli spaces of flat connections, in each case producing generalized action variables where earlier results only established integrability on generic symplectic leaves.","pith_inferences":["A boundary test the authors leave implicit is whether the product-form stabilizer condition can fail for the extended Hamiltonian families mentioned in Remark 5.23 and Appendix B.2; checking one such family by direct stabilizer computation would map out the method's limits.","Because compactness of $G_1$ is not essential and only properness of the combined action is used, a concrete extension would be to run Scenario 2.11 with a non-compact symmetry group in Toda-type reductions.","Since the action variables are smooth only on a dense open set, the reduced systems likely admit singular extensions to non-principal strata, and describing the behaviour there is a natural continuation.","The permutation construction in Appendix B.3 produces systems not covered by the main family, and testing whether those systems are genuinely inequivalent would be a concrete next step."],"forward_implications":["Every realization of Scenario 2.11 gives an integrable system of rank $\\ell$ on $M_*/G_1$, with $H_1,\\ldots,H_\\ell$ serving as generalized action variables on $Y_0/G_1$ and $Y_0^{1}/G_1$, and with integrability on every symplectic leaf of those subsets.","The reductions of the master systems on $T^*K$ and on the Heisenberg double are integrable in this stronger leaf-by-leaf sense; in particular, the dual Ruijsenaars-type system whose integrability was previously incomplete is now covered.","The quasi-Poisson reductions of $D(K)$ yield integrable systems with explicit action variables on moduli spaces of flat connections for the sphere with four holes, the torus with one hole, and the genus-two surface with one hole.","The general family over $M_{m,n}$ produces integrable systems of rank $(p+b_p+q)\\ell$ on moduli spaces for arbitrary genus and number of holes whenever the interval and commutator data satisfy Assumption B.1.","In these systems the constants of motion are explicit invariants of the reduced free $G_2$-action, and generalized action-angle coordinates exist around points with trivial $G_2$-isotropy."],"supporting_citations":[{"why":"defines quasi-Poisson manifolds, the setting in which only invariant functions form the Poisson algebra used by the scenario.","marker":"[1]"},{"why":"supplies the group-valued momentum-map framework and the connectedness of fibers used in the double and moduli-space reductions.","marker":"[2]"},{"why":"gives the moduli-space integrable systems that the paper's quasi-Poisson examples complement and compare against.","marker":"[3]"},{"why":"provides the foundational facts on proper actions and principal isotropy types used throughout Section 2.","marker":"[15]"},{"why":"is the previous treatment of reductions of master systems on doubles whose open problems the paper solves.","marker":"[23]"},{"why":"supplies the action-angle theorem on Poisson manifolds that Theorem 2.15 generalizes to generalized tori.","marker":"[45]"},{"why":"established degenerate integrability of spin Calogero-Moser systems and Ruijsenaars duals on generic leaves, the statement the paper sharpens.","marker":"[60]"},{"why":"links torus actions to integrable systems, the principle behind generating integrable systems from generalized torus actions.","marker":"[75]"}],"fun_headline_variants":["Poisson reduction preserves integrability and action variables","Generalized torus reductions yield integrable systems","Action variables survive Poisson reduction intact","Reduction builds integrable systems with explicit actions","Superintegrable systems from Poisson quotients"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that on the principal-isotropy part of the combined action, every stabilizer is exactly the product of its $G_1$-part with the identity of the generalized torus; if that product form fails, the dimension count that produces the correct number of integrals breaks down.","fun_headline_variants_meta":{"raw":{"variants":["Poisson reduction preserves integrability and action variables","Generalized torus reductions yield integrable systems","Action variables survive Poisson reduction intact","Reduction builds integrable systems with explicit actions","Superintegrable systems from Poisson quotients"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000602,"raw_usage":{"total_tokens":2850,"prompt_tokens":1027,"completion_tokens":1823,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":643,"completion_tokens_details":{"reasoning_tokens":1755}},"tokens_in":643,"tokens_out":1823,"duration_ms":15842,"temperature":1.0,"reasoning_tokens":1755,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T16:57:26.562264+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a compact Poisson or quasi-Poisson $G_1$-manifold with a commuting proper action of $G_2 = \\mathrm{U}(1)^{\\ell_1}\\times\\mathbb{R}^{\\ell_2}$ whose other hypotheses hold, and a point in the principal isotropy locus whose stabilizer contains an element $(e_1,t)$ with $t\\neq e_2$; at such a point the equality $\\mathrm{ddim}(\\mathcal F^{G_1}_*) = \\dim(M_*/G_1)-\\ell$ would fail, so Theorem 2.14 would not apply to that example.","supporting_citations":[{"cited_title":"Quasi-Poisson Manifolds","cited_arxiv_id":"math/0006168","evidence_quote":"defines quasi-Poisson manifolds, the setting in which only invariant functions form the Poisson algebra used by the scenario."},{"cited_title":"Lie Group Valued Moment Maps","cited_arxiv_id":"dg-ga/9707021","evidence_quote":"supplies the group-valued momentum-map framework and the connectedness of fibers used in the double and moduli-space reductions."},{"cited_title":"Dwyer and C.W","cited_arxiv_id":null,"evidence_quote":"provides the foundational facts on proper actions and principal isotropy types used throughout Section 2."},{"cited_title":"Feh´ er, Notes on the degenerate integrability of reduced systems obtained from the master systems of free motion on cotangent bundles of compact Lie groups","cited_arxiv_id":null,"evidence_quote":"is the previous treatment of reductions of master systems on doubles whose open problems the paper solves."},{"cited_title":"Liashyk, G","cited_arxiv_id":null,"evidence_quote":"supplies the action-angle theorem on Poisson manifolds that Theorem 2.15 generalizes to generalized tori."}],"review_version":3}