{"id":"88be523c-2180-4487-8962-c54735aed8f5","arxiv_id":"2411.13063","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For k≤m, the pushforward of Lebesgue measure under the Gram map is (2^{-k} ∏_{j=1}^k Vol(S^{m-j})) |det G_k|^{(m-k-1)/2} dG, giving the Hilbert measure on the orbit space.","lead":"This paper derives an explicit volume measure on the quotient of k vectors in R^m by the orthogonal group, expressed through the determinant of the Gram matrix. The formula provides a canonical way to integrate rotation-invariant functions and matches the classical Wishart Jacobian used in statistics.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's own theorem contradicts the headline observation: for k=m-2 the density has exponent (m-k-1)/2 = 1/2, so it is not smooth at |G_k|=0 even though k<m.","rationale":"The reader's formal weakest assumption points to the generalized Euler-angle parametrization and Jacobian identity (3.4). I rechecked that computation and found no error: the bookkeeping in (3.4) reproduces the two worked examples, and the same constant factor follows from the standard QR/Wishart change of variables, so the formula in Theorem 1.1 is robust. The load-bearing weakness is instead the paper's advertised observation that singularities occur if and only if k=m. The theorem's own exponent shows this is false: for k<m but m-k even, the exponent (m-k-1)/2 is a positive half-integer, so the density is not smooth at |G_k|=0. This is not merely a disagreement with external consensus; it is internally inconsistent with the proved formula. It does not change the numerical content of Theorem 1.1, but it does require correcting the abstract, the introduction, and the interpretive discussion. Since the reader already reached CONDITIONAL on essentially this basis, my stress-test leaves the verdict unchanged.","tokens_in":14260,"tokens_out":12778,"duration_ms":130068,"concrete_test":"Specialize Theorem 1.1 to k=1,m=3, a coregular case with k<m, and evaluate on the boundary of X: λ_{1,3}(u)=2π√u. Compute the second derivative of λ with respect to u at u=0; it is infinite, so λ is not C^1 there. This directly contradicts the paper's assertion that smoothness fails only for k=m and identifies the exact exponent that invalidates the iff claim.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1.1 gives λ_{k,m} ∝ |G_k|^{(m-k-1)/2}. The introduction and abstract claim λ_{k,m} is a smooth function on R^{(k+1)/2} except when k=m, with singularities only in that case. But for k=m-2 the exponent is 1/2; for example k=1,m=3 gives λ_{1,3}(u)du = 2π√u du on [0,∞), and √u is not differentiable at u=0, a non-principal stratum point. Thus the 'if and only if' singularity statement is false as stated, and the paper's highlighted conclusion is contradicted by its own formula. This is a genuine internal inconsistency in the central narrative, even though the constant factor in Theorem 1.1 and the Jacobian computation (3.4) appear to be correct and match the independent QR/Wishart derivation.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs canonical measures, called Hilbert measures, on the orbit spaces V_k/O_m where V_k = (R^m)^k with diagonal O_m action, for k ≤ m. The main result (Theorem 1.1) gives an explicit formula for the pushforward of Lebesgue measure under the Hilbert embedding u: V_k → R^{k(k+1)/2} given by the inner products ui,j = ⟨vi,vj⟩: the density is (1/2^k) ∏_{j=1}^k Vol(S^{m−j}) |G_k|^{(m−k−1)/2}, where G_k is the Gram matrix. The proof uses a generalized Euler angle parametrization of SO_m to first reduce to a fundamental domain and then performs a second change of variables to the Gram invariants, with the Jacobians computed in Lemmas 3.1 and 3.2. The paper also states a corollary for SO_m and discusses the smoothness of the density along non-principal strata, claiming singularities occur if and only if k = m.","tokens_in":14436,"tokens_out":7219,"duration_ms":66207,"significance":"If Theorem 1.1 is correct, it provides a complete and explicit solution to a natural change-of-variables problem for orthogonal-invariant integrals, generalizing the classical spherical-coordinate example and potentially useful in the Faddeev-Popov approach to gauge systems. The computation is concrete and checkable: the telescoping product in Lemma 3.1, the triangular Jacobian in Lemma 3.2, and the final formula are internally consistent, and the examples in Section 4 (k=2,m=3 and k=2,m=2) match the general formula. The explicit constants in terms of the volumes of spheres are derived cleanly, and the formula reduces correctly to the known single-vector case. However, the advertised smoothness characterization is contradicted by the paper's own formula, as detailed in the major comments, and the proof of Corollary 1.3 contains a circular step that should be repaired.","major_comments":[{"comment":"The claim that the Hilbert measure density λ_{k,m} is smooth on R^{(k+1)/2} in every case except k=m, and that singularities occur if and only if k=m, is contradicted by Theorem 1.1 itself. For k=m−2 the exponent (m−k−1)/2 equals 1/2, so the density is |G_k|^{1/2}, which is not differentiable on the zero set of G_k; for example, the paper's own motivating case k=1, m=3 yields λ_{1,3}(u) du = 2π√u du, which is not differentiable at u=0 even though k<m. The correct condition for smoothness of the density at points where |G_k|=0 is that (m−k−1)/2 be a nonnegative integer (or zero). This error affects the abstract and the highlighted observation, and it must be corrected or qualified.","section":"Abstract and Introduction (paragraph after Corollary 1.2)"}],"minor_comments":[{"comment":"The proof uses Theorem 1.1 to conclude that |G_k| ≥ 0 on u(V_k), but Theorem 1.1's formula defines a genuine positive measure only if the density is real and nonnegative, which already presupposes that |G_k| ≥ 0 on X; this is a circular step. Since the corollary is a direct consequence of the definition of the Gram matrix of a set of vectors, the proof should be replaced by a direct argument, such as the Cholesky decomposition already used in the second half of the proof.","section":"Proof of Corollary 1.3"},{"comment":"In the paragraph after the change of variables, the expression 'w_{2,3} = ρ_2 sin μ' should read 'w_{2,2} = ρ_2 sin μ'; there is no third coordinate of w_2 in the case k=m=2.","section":"Example 4.2"},{"comment":"In the final displayed equation of the proof, the expression '2^k √{G_1 G_2 \\cdots G_k}' is missing absolute values around the determinants; on the principal stratum the determinants are positive so this is harmless, but including absolute values would make the formula valid on the closure.","section":"Lemma 3.2"},{"comment":"The bookkeeping identity (3.4) for the Jacobian of the first change of variables is compressed; a short explanation of how the pattern from Equations (3.1)–(3.3) leads to the general formula, or an indication of the recursive rotation steps, would improve readability.","section":"Section 3.1, Eq. (3.4)"}],"recommendation":"major_revision","confidential_remarks":"The main formula (Theorem 1.1) appears correct and is a solid contribution, but the abstract and introduction overclaim the smoothness results. The error is not merely cosmetic: it concerns the central advertised observation, so it should be fixed before publication. The circularity in Corollary 1.3 is minor because the statement is elementary, but it should still be addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the formula in Theorem 1.1 is correct as far as I can tell, and the derivation is a real proof, not a plausibility sketch. But the paper's headline observation—singularities occur only when k=m—is false, and the paper's own formula disproves it. For k=1, m=3, the exponent (m-k-1)/2 equals 1/2, so λ_{1,3}(u)=2π√u, which is not differentiable at u=0. The same happens for k=m-2 whenever m-k-1 is odd. This is not a wording issue: the abstract, the introduction, and the highlighted conclusion all assert the false claim.\n\nWhat the paper does well: it gives a self-contained computation of the Hilbert measure for all coregular O_m modules, using the generalized Euler angle parametrization. The bookkeeping in Section 3.1 is complicated, and identity (3.4) is the load-bearing step; it appears correct, and the telescoping in Lemmas 3.1 and 3.2 is clean. The examples in Section 4 are consistent with the formula. The constant factor 2^{-k}∏Vol(S^{m-j}) matches an independent QR/Wishart computation, which is another good sign.\n\nSoft spots: the singularity claim is the big one and must be fixed. Second, the formula is the standard Jacobian of the Gram map in multivariate statistics, known through the Wishart distribution, and the paper does not cite that literature. The authors should acknowledge this and explain what is genuinely new—likely the orbit-space framing and the Euler-angle proof. Third, the proof of Corollary 1.3 contains a minor circular step: it derives |G_k|≥0 from Theorem 1.1, but the theorem only refers to the absolute value, so the positivity of the Gram determinant on the principal stratum needs a separate argument.\n\nWho this is for: invariant theorists and anyone integrating invariant functions over orbit spaces. The main formula is useful and citable. It deserves a serious referee; the errors are localized and fixable. I would send it to review with a request that the authors correct the singularities claim and contextualize the Wishart prior.","headline":"A correct and competently derived explicit formula for Hilbert measures on O_m orbit spaces, undermined by a false singularity claim in the abstract that its own theorem contradicts.","tokens_in":14950,"tokens_out":3533,"would_cite":true,"duration_ms":33331,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["57S15","14L30","28A99"],"pacs":[],"model":"deepseek-v4-flash","headline":"For k ≤ m, the Hilbert measure on the O_m orbit space of k vectors is a constant times a power of the Gram determinant.","keywords":["Hilbert measure","orbit space","orthogonal group","coregular representation","Gram matrix","change of variables","invariant theory","Euler angles"],"falsifier":"For k=3, m=4 the formula predicts a constant density $\\lambda = 2^{-3}\\operatorname{Vol}(S^3)\\operatorname{Vol}(S^2)\\operatorname{Vol}(S^1) = 2\\pi^4$. Sample triples of vectors in $\\mathbb{R}^{12}$ from a Gaussian distribution, form their Gram matrices, and compare the empirical density in a small region of the six invariant coordinates with $2\\pi^4$; a mismatch in the normalization or any dependence on the Gram entries would falsify Theorem 1.1.","tokens_in":14061,"feed_emoji":"📐","tokens_out":7010,"duration_ms":66096,"temperature":0.7,"pith_summary":"This paper proves an explicit formula for the Hilbert measure on the orbit space of the orthogonal group O_m acting diagonally on k copies of its defining representation, in the coregular range k ≤ m. The orbit space is coordinatized by the mutual inner products of the k vectors, i.e., by the Gram matrix G_k, and the Hilbert measure is the pushforward of Lebesgue measure on $R^{{mk}}$ to that coordinate space. The formula is $\\lambda_{k,m}(u)\\,du = 2^{-k}\\left(\\prod_{j=1}^{k}\\operatorname{Vol}(S^{m-j})\\right)|G_k|^{(m-k-1)/2}\\,du$. This matters because it turns integration of O_m-invariant functions into an ordinary integral over a semialgebraic domain, the set of positive-semidefinite Gram matrices, and it isolates the only singular case: for k=m the density diverges along the non-principal strata, while for k<m it is smooth. The argument also yields, for k ≤ m−1, the same formula for the special orthogonal group, since the invariants coincide.","feed_headline":"Hilbert measure on O_m quotients is an explicit Gram-determinant power","feed_subtitle":"For k ≤ m the density on invariant coordinates is smooth except at k = m, where it blows up on lower strata.","key_machinery":"The carrying mechanism is a two-step change of variables. First, generalized Euler angles for SO_m—the parametrization in which every unit vector is built by successive rotations—are applied vector by vector to move a generic tuple $(v_1,\\dots,v_k)$ into a fundamental domain where the i-th vector has zero coordinates after position i and positive diagonal coordinate; the Jacobian of this step is the bookkeeping identity (3.4), a product of powers of the diagonal entries and sine factors. Second, the diagonal entries are expressed as ratios of Gram determinants, $w_{i,i} = \\sqrt{|G_i|/|G_{i-1}|}$, and the map from the w-coordinates to the invariants is triangular with determinant $2^{-k}(|G_1|\\cdots|G_k|)^{-1/2}$. The sine factors integrate to products of sphere volumes, and the Gram determinants telescope, leaving the single power $|G_k|^{(m-k-1)/2}$.","core_discovery":"The central discovery is Theorem 1.1: for k ≤ m, the unique measure $\\lambda_{k,m}(u)\\,du$ on the orbit space $V_k/O_m$ that reproduces Lebesgue integration of invariant functions is the Gram-determinant power density above. Equivalently, the measure is $2^{-k}$ times the volume of the Stiefel manifold $O_m/O_{m-k}$ times $|G_k|^{(m-k-1)/2}\\,du$. The proof exhibits an explicit fundamental domain for the O_m action, parametrized by the nonzero entries of a lower-triangular matrix, then changes variables from those entries to the invariant inner products; the Gram determinants telescope to give the stated power. In the boundary case k=m, the exponent is negative, so the density has square-root singularities exactly on the lower-dimensional strata where the vectors become linearly dependent.","pith_inferences":["The same two-step procedure should in principle produce explicit Hilbert measures for every coregular representation of a compact Lie group, since such actions admit generic local cross-sections and the invariants can be used as coordinates; a natural test case is the classification of coregular representations beyond the orthogonal defining modules.","In a zero-dimensional gauge theory, the formula gives the effective integration measure on gauge-invariant degrees of freedom after integrating out the gauge group, making the gauge-fixing determinant explicit; this suggests checking the measure against a direct gauge-fixing computation for a small matrix model.","The observation that k=m is the only singular case suggests a geometric reading: the singularity exponent equals $(m-k-1)/2$, so as the number of vectors approaches the ambient dimension the invariant measure develops a boundary divergence; one could test whether this divergence controls the asymptotic behaviour of invariant integrals near the singular strata."],"forward_implications":["For k ≤ m−1 the SO_m orbit space carries the same Hilbert measure, because the SO_m- and O_m-invariants of these representations coincide and every point is fixed by a reflection.","The density $\\lambda_{k,m}$ is smooth on all of $\\mathbb{R}^{\\binom{k+1}{2}}$ except when k=m; in that case the $|G_m|^{-1/2}$ singularity lies on a measure-zero set, so integrals are still determined by the principal stratum.","The formula, together with the description of the image as the positive-semidefinite cone, makes the integral over $V_k$ of any O_m-invariant function an explicit integral over the positive-semidefinite matrix cone.","Concrete examples: for two vectors in $\\mathbb{R}^3$ the density is the constant $2\\pi^2$, while for two vectors in $\\mathbb{R}^2$ it is $\\pi/\\sqrt{|G_2|}$; both follow from the same general formula."],"supporting_citations":[{"why":"Supplies the generalized Euler-angle parametrization of SO_m whose Jacobian bookkeeping underlies the first change of variables.","marker":"[6]"},{"why":"Provides the differentiable-invariants theorem used to express any smooth invariant function as a function of the Hilbert invariants.","marker":"[13]"},{"why":"Establishes that the k-copy defining representation of O_m is coregular if and only if k ≤ m, fixing the range of the formula.","marker":"[16]"},{"why":"Provides the sphere-volume and Stiefel-manifold volume identities used to package the constant factor.","marker":"[18]"},{"why":"Gives the inequalities defining the image u(V_k), used to identify the integration domain as the positive-semidefinite Gram matrices.","marker":"[11]"}],"fun_headline_variants":["Hilbert measure on O_m quotients: explicit Gram power","Gram-determinant Hilbert measure: singular exactly at k=m","Hilbert measures on O_m orbit spaces smooth unless k=m","Explicit Hilbert measure: Gram power, blow-up on lower strata","O_m orbit measures: smooth for k<m, singular at k=m"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof leans on the generalized Euler-angle parametrization of SO_m and on the unexpanded Jacobian identity (3.4); if that parametrization misses a set of positive measure, or the identity misstates a sine power or constant, the constant factor in the Hilbert measure would be off.","fun_headline_variants_meta":{"raw":{"variants":["Hilbert measure on O_m quotients: explicit Gram power","Gram-determinant Hilbert measure: singular exactly at k=m","Hilbert measures on O_m orbit spaces smooth unless k=m","Explicit Hilbert measure: Gram power, blow-up on lower strata","O_m orbit measures: smooth for k<m, singular at k=m"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000238,"raw_usage":{"total_tokens":1430,"prompt_tokens":782,"completion_tokens":648,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":398,"completion_tokens_details":{"reasoning_tokens":557}},"tokens_in":398,"tokens_out":648,"duration_ms":12083,"temperature":1.0,"reasoning_tokens":557,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T16:53:21.371093+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For k=3, m=4 the formula predicts a constant density $\\lambda = 2^{-3}\\operatorname{Vol}(S^3)\\operatorname{Vol}(S^2)\\operatorname{Vol}(S^1) = 2\\pi^4$. Sample triples of vectors in $\\mathbb{R}^{12}$ from a Gaussian distribution, form their Gram matrices, and compare the empirical density in a small region of the six invariant coordinates with $2\\pi^4$; a mismatch in the normalization or any dependence on the Gram entries would falsify Theorem 1.1.","supporting_citations":[{"cited_title":"Hoffman, Richard C","cited_arxiv_id":null,"evidence_quote":"Supplies the generalized Euler-angle parametrization of SO_m whose Jacobian bookkeeping underlies the first change of variables."},{"cited_title":"Schwarz, Smooth functions invariant under the action of a compact Lie group, Topology 14 (1975), 63–68","cited_arxiv_id":null,"evidence_quote":"Provides the differentiable-invariants theorem used to express any smooth invariant function as a function of the Hilbert invariants."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes that the k-copy defining representation of O_m is coregular if and only if k ≤ m, fixing the range of the formula."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the inequalities defining the image u(V_k), used to identify the integration domain as the positive-semidefinite Gram matrices."}],"review_version":1}