{"id":"bb144bfe-ff0a-4a84-aacc-b208227375ce","arxiv_id":"2608.13264","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The Gromov-Hausdorff distance between S^n and S^{n+1} with geodesic metrics is exactly one half of arccos(-1/(n+1)).","lead":"This paper proves the exact Gromov-Hausdorff distance between unit round spheres of consecutive dimensions, settling a conjecture that was open in all but the lowest three cases. The result provides a benchmark for shape comparison algorithms and introduces a new join operation for correspondences with an exact distortion formula.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Proposition 2.10 defines K with floor, but the existence claim requires ceiling; e.g., m=2,n=5 is impossible, so the claim is false as stated. Theorem A is unaffected.","rationale":"The reader's weakest assumption is the algebraic Lemma 5.9. While that lemma is very complex, I could not identify a specific flaw in it or in the surrounding argument. Instead, I found a concrete, verifiable error in the secondary circle-factor construction: the floor/ceiling confusion in Proposition 2.10. This is load-bearing for the paper's advertised bounds on nonconsecutive spheres and creates an internal inconsistency between the definition of K, the proof, and the figures. Because Theorem A does not rely on Section 2.5, the central claim remains supported. However, a manuscript containing a false proposition as stated should not receive an unconditional accept; a correction of the definition (floor to ceiling) is needed. Hence I recommend CONDITIONAL, and I do not agree that Lemma 5.9 is the weakest point of the paper.","tokens_in":70228,"tokens_out":26437,"duration_ms":237591,"concrete_test":"Change the definition of K(m,n) to K=ceil((n−m)/L_m) and recompute the existence condition and Corollary 2.11. For m=2,n=5, the corrected K=2 gives a valid choice d_1=2 with distortion δ(2)=4π/5, matching the staircase bound. Verify that the corrected definition reproduces the claims of Proposition 2.10 and the labels in Figures 4–5 for all small m,n; if it does, the floor definition is confirmed as a typo.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"In Proposition 2.10, K(m,n) is defined as floor((n−m)/L_m) with L_m=2b(m,2). The proof asserts a choice of integers d_i≥0 with sum d_i=n−m and d_i≤2K exists because 2K b(m,2) ≥ n−m. For floor this inequality is reversed: K ≤ (n−m)/(2b), so 2Kb ≤ n−m. Example m=2,n=5: b=1, L=2, K=floor(3/2)=1, so the cap is 2, but n−m=3 requires d_1=3. This also breaks Corollary 2.11's claimed equivalence (k−1)L_m < n−m ≤ kL_m ⇒ K=k; with floor, n−m=3,L=2,k=2 gives K=1, not 2. The proof and Figures 4–5 implicitly use K=ceil((n−m)/L_m), so the manuscript is internally inconsistent. This error is concrete and load-bearing for the statements about nonconsecutive spheres, but it does not affect Theorem A, whose proof is independent.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper determines the Gromov–Hausdorff distance between consecutive unit round spheres with their geodesic metrics, proving that d_GH(S^n, S^{n+1}) = (1/2) arccos(-1/(n+1)) for every n ≥ 1. The upper bound is established by an explicit anchored–chord correspondence R_n whose distortion is exactly the simplex edge length ζ_n, and the matching lower bound is re-proved self-containedly through a quantitative Borsuk–Ulam argument. The paper also introduces synchronized spherical joins and suspensions of correspondences, proves that distortion is preserved under the maximum of factor distortions, and derives bounds for spheres of nonconsecutive dimensions, including a π/4 stabilization for fixed or sublinear codimension.","tokens_in":70429,"tokens_out":8322,"duration_ms":91017,"significance":"If correct, the main theorem resolves a conjecture of Lim, Mémoli, and Smith and provides a uniform construction of an optimal correspondence in every dimension. A noteworthy strength is that the construction is genuinely parameter-free: the gain is forced by sharpness on the swapped-anchor family and by the involution F_n, rather than fitted to the target value. The self-contained lower-bound proof, the exact distortion formula for spherical joins, and the explicit suspension results are independently useful. I found no defect in the proof of Theorem A; however, the nonconsecutive-sphere section contains a concrete false statement that must be corrected before publication.","major_comments":[{"comment":"The definition K(m,n) := floor((n-m)/L_m) is internally inconsistent with the existence claim made in the proof. For the floor definition, the inequality 2K(m,n) b(m,2) ≥ n-m used in the proof is reversed: K ≤ (n-m)/(2b(m,2)) implies 2K b(m,2) ≤ n-m. The counterexample m=2, n=5 has b(m,2)=1, L_m=2, K=1, so the constraint d_i ≤ 2K bounds the sum of the d_i by 2, whereas n-m=3; no admissible choice of d_i exists. The proof and the claimed equivalence in Corollary 2.11 use K = ceil((n-m)/L_m), not floor, and Figures 4–5 are drawn according to the ceiling convention. As written, Proposition 2.10, Eqs. (29)–(30), and Corollary 2.11 are false. The fix is local: replace floor by ceiling in the definition of K(m,n) and adjust the surrounding inequalities accordingly. This issue does not affect Theorem A, whose proof is independent of Section 2.","section":"§2.5, Proposition 2.10"}],"minor_comments":[{"comment":"The sign of the polynomial P_{ρ_n} is the decisive algebraic step in Proposition 5.5. I checked the printed factorization and the endpoint cases for several small values of n and found them consistent, but the verification is lengthy enough that a reader cannot realistically check every coefficient by hand. I recommend that the authors supply a computer-algebra verification or a notebook certifying the factorization in Eq. (58) and the sign analysis in Appendix B.2.","section":"Appendix B.2, Lemma 5.9"},{"comment":"The notation bζ_n for π - ζ_n is easy to misread as a product b ζ_n, especially in plain-text discussion. Please introduce a symbol such as \\widehat\\zeta_n or \\bar\\zeta_n and use it consistently in the final version.","section":"Notation throughout"},{"comment":"In the caption of Figure 2, the sentence 'the highlighted pair has x ∈ M_2 and r_1(α)=2' could be misunderstood because r_1(α) is an extended-real gain; specifying that this is the finite gain value for the drawn α, rather than the bounded chord fraction λ_1(α), would improve clarity.","section":"Figure 2"}],"recommendation":"major_revision","confidential_remarks":"The floor/ceiling slip in Proposition 2.10 is the only substantive defect I found in reading the manuscript. It is localized to the nonconsecutive-sphere applications and does not undermine Theorem A, so this is not a reject decision. I request a major revision because the defective statements are advertised in the abstract and used in comparisons with prior work, and because the manuscript as written is internally inconsistent in Section 2.5. Once the definition is corrected to the ceiling convention, the claims in Propositions 2.10–2.11 and the figures should be rechecked against the corrected formula."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things. The central result is very likely right: the explicit anchored-chord correspondence gives the sharp upper bound matching the quantitative Borsuk-Ulam lower bound, and the proof is self-contained. The secondary material on nonconsecutive spheres contains a concrete error that is real but isolated.\n\nWhat is new and good: the equality d_GH(S^n,S^{n+1}) = \\zeta_n/2 for all n resolves a conjecture open for n\\ge 4. The construction is uniform, the lower bound is re-proved directly, and the synchronized spherical join theorem (distortion of a join = max of distortions) is a genuinely useful tool. The upper-bound proof is long but honest; every estimate is tied to an explicit correspondence, and the auxiliary algebra is relegated to appendices rather than hand-waved. The citation pattern also looks fine: prior work is cited contextually, and the self-citations are legitimate.\n\nSoft spots: Proposition 2.10 defines K(m,n) = floor((n-m)/L_m), but the existence proof needs K = ceil. With floor, the inequality 2K b(m,2) \\ge n-m is false in general: m=2, n=5 gives K=1 and the claimed sum n-m=3 cannot be split into one integer at most 2. Corollary 2.11's equivalence also breaks. The text and Figures 4-5 implicitly use the ceiling version, so the manuscript is internally inconsistent. This does not touch Theorem A, whose proof is independent, but it does mean the advertised all-pairs bounds and the comparison with Harrison-Jeffs need repair. Also, Lemma 5.9 is load-bearing for n\\ge 2 and is verified by a long factorization; I did not find a gap, and your moderate confidence is reasonable, but a referee should check that polynomial numerically before the proof is trusted.\n\nBottom line: the paper deserves a serious referee. Send it to review, but ask the referee to verify Lemma 5.9 and to make Section 2 internally consistent, either by fixing K to be ceil or by adjusting the claims and figures. This is a publishable main theorem with a fixable secondary bug.","headline":"The main theorem, exact GH distance between consecutive spheres, looks solid and is a genuine advance; but the secondary circle-factor constructions in Section 2 have a floor/ceil error that needs fixing before publication.","tokens_in":70937,"tokens_out":2478,"would_cite":true,"duration_ms":25757,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["53C23","51F30","55M20"],"pacs":[],"model":"deepseek-v4-flash","headline":"For every n ≥ 1, the Gromov–Hausdorff distance between consecutive unit spheres equals half of ζ_n = arccos(−1/(n+1)), the edge length of a regular simplex inscribed in S^n; the paper exhibits the correspondence that attains it.","keywords":["Gromov–Hausdorff distance","round spheres","geodesic metric","metric distortion","correspondences","Borsuk–Ulam theorem","spherical joins","regular simplex"],"falsifier":"Check the explicit relation $R_4$ directly: either evaluate the polynomial $P_{\\rho_n}$ of Lemma 5.9 on a dense grid of parameters satisfying condition (52) for $n = 4$ and look for a positive value, or compute the realized defect maxima of $R_4$ on fine finite nets of $S^4$ and $S^5$. Any parameter point that satisfies the condition yet gives a positive polynomial would refute the lower-boundary proof, and any matched pair whose defect exceeds $\\zeta_4$ would refute optimality of the constructed correspondence.","tokens_in":70029,"feed_emoji":"🔵","tokens_out":25238,"duration_ms":203389,"temperature":0.7,"pith_summary":"This paper determines the exact Gromov–Hausdorff distance between consecutive unit round spheres with their geodesic metrics: for every $n \\geq 1$, the distance from $S^n$ to $S^{n+1}$ is $\\zeta_n/2$, where $\\zeta_n = \\arccos(-1/(n+1))$ is the common geodesic distance between distinct vertices of a regular simplex with $n+2$ vertices inscribed in $S^n$. The Gromov–Hausdorff distance measures how far two metric spaces are from being isometric; here it is computed via the worst distortion forced by any pairing (correspondence) of points between the two spheres. The lower bound was already known from a quantitative Borsuk–Ulam principle; the paper's contribution is an explicit correspondence whose distortion hits that bound exactly, proving equality in every dimension at once, including all cases $n \\geq 4$ that were previously open after $n = 1, 2, 3$ had been settled separately. The construction also feeds a second, general result: synchronized spherical joins of correspondences multiply sphere dimensions while their distortion is exactly the maximum distortion of the factors, which yields new upper bounds for spheres of nonconsecutive dimensions and shows that every fixed-gap distance stabilizes to $\\pi/4$ as the dimension grows.","feed_headline":"The gap between adjacent spheres is exactly half a simplex edge","feed_subtitle":"One explicit pairing of points matches the topological lower bound, settling all previously open cases n ≥ 4.","key_machinery":"The load-bearing object is the anchored–chord correspondence $R_n$, built from the vertices $v_1, \\ldots, v_{n+2}$ of a regular simplex inscribed in $S^n$ (the anchors), their closed spherical Voronoi cells $M_i$, and the normalized-chord map $H_{\\alpha,i}(x)$, which radially projects the Euclidean chord point $(v_i + r_n(\\alpha)x)/(1 + r_n(\\alpha))$ onto the sphere. The scalar $r_n(\\alpha)$ is the canonical gain: the unique continuous choice that makes the equal-level swapped-anchor comparisons sharp and satisfies the reciprocal identity $r_n(\\alpha) r_n(F_n(\\alpha)) = 1$, where the critical involution $F_n$ is defined by $\\cos\\alpha \\cos F_n(\\alpha) - \\rho_n \\sin\\alpha \\sin F_n(\\alpha) = \\rho_n$ with $\\rho_n = 1/(n+1)$. The proof that this correspondence attains distortion $\\zeta_n$ runs through the admissible $(Q,R)$-region of source and target inner products: same-anchor pairs are controlled by a vertex-dependent isometric embedding of $S^n$ into $S^{n+1}$ that keeps each embedded target within $(\\pi-\\zeta_n)/2$ of its source; distinct-anchor pairs with $Q \\geq \\rho_n$ are controlled by a lower-boundary estimate whose proof uses Ptolemy's inequality and reduces to the sign of a specific polynomial (Lemma 5.9); and pairs with $Q \\leq -\\rho_n$ are controlled by an exact dimension reduction of an optimization over two spherical caps to a one-variable concave maximization. The matching lower bound is proved self-containedly as a quantitative Borsuk–Ulam theorem for adjacent spheres, combining a helmet trick that converts arbitrary maps into odd maps with a sharp spherical Jung lemma; on top of the consecutive case, the synchronized spherical join operation is the mechanism that converts the sharp relations into bounds for nonconsecutive dimensions.","core_discovery":"The central claim is Theorem A: for every integer $n \\geq 1$, $d_{\\mathrm{GH}}(S^n, S^{n+1}) = \\zeta_n/2 = \\tfrac{1}{2}\\arccos(-1/(n+1))$, where the two spheres carry their geodesic distances. The paper proves this by constructing, for each $n$, one anchored–chord correspondence $R_n \\subseteq S^{n+1} \\times S^n$ whose distortion equals $\\zeta_n$, exactly matching the quantitative Borsuk–Ulam lower bound that any correspondence between adjacent spheres has distortion at least $\\zeta_n$. The matching construction is uniform: a point of the upper hemisphere at colatitude $\\alpha$ from the north pole is paired with the radial projection onto $S^n$ of a point on the Euclidean chord joining a regular-simplex vertex to its target, with a canonical gain function $r_n(\\alpha)$ forced by sharpness on the swapped-anchor family and by a reciprocal identity under a critical radial involution. The distortion analysis splits into same-anchor comparisons, where an isometric embedding of $S^n$ into $S^{n+1}$ keeps each embedded target within $(\\pi-\\zeta_n)/2$ of its source, and distinct-anchor comparisons, controlled by two boundary estimates in the plane $(Q,R)$ of source and target inner products: a lower boundary proved through Ptolemy's inequality and a polynomial sign lemma, and an upper boundary proved by reducing an optimization over two spherical caps to a one-variable concave maximization. The paper also proves that synchronized spherical joins of correspondences have distortion exactly equal to the maximum distortion of their factors, so suspension preserves distortion; joining and suspending the sharp relations yields bounds for arbitrary sphere pairs, including $\\lim_{m\\to\\infty} d_{\\mathrm{GH}}(S^m, S^{m+d(m)}) = \\pi/4$ whenever $d(m) \\geq 1$ and $d(m) = o(m)$.","pith_inferences":["Read as a general principle, the join theorem turns the sphere-distance problem into a finite minimax allocation — distribute the dimension gap as evenly as possible among sharp factors — and the same max-rule is a natural test case for other symmetric families, such as real or complex projective spaces, for which no exact consecutive-pair distances are known.","Because R_n is fully explicit, it supplies a concrete ground-truth object: numerically evaluating defect maxima over fine nets of S^4 and S^5 and comparing with ζ_4 would give a computation-based check of the algebraic bottleneck, and the same nets could calibrate practical shape-matching heuristics that approximate the Gromov–Hausdorff distance.","The fact that sharpness on one extremal family (the swapped-anchor equal-level pairs) determines the gain uniquely suggests that near-optimal correspondences might be characterized by their (Q,R)-boundary curves; a testable consequence is that any correspondence with distortion within ε of ζ_n must have its realized inner-product pairs confined to a thin neighborhood of the region bounded by the t"],"forward_implications":["Every consecutive-sphere distance is now exactly known from a single construction, including the previously open cases n ≥ 4, and the same relation also reproduces the three cases n = 1, 2, 3 that had been treated by different arguments.","For every fixed gap d ≥ 1, the distance m ↦ d_GH(S^m, S^{m+d}) is nonincreasing in m and converges to π/4, and the same limit holds when the gap grows sublinearly with d = o(m).","Suspension of a correspondence preserves its distortion, so any comparison between S^m and S^n transfers to S^{m+1} and S^{n+1} at no extra cost.","Balanced joins of the sharp relations yield the upper bound d_GH(S^m, S^{m+d}) ≤ (1/2) arccos(−1/⌊(m+1)/d⌋) whenever 2d ≤ m+1, while joins of exact circle correspondences give the staircase bound d_GH(S^m, S^n) ≤ πk/(2k+1) whenever the codimension n−m lies in the k-th band of width 2⌊(m+1)/2⌋.","The stabilization to π/4 as m grows contrasts with the infinite-dimensional sphere, from which every finite-dimensional sphere remains at the maximal possible distance π/2."],"supporting_citations":[{"why":"States the consecutive-sphere conjecture, proves the quantitative Borsuk–Ulam lower bound d_GH(S^m, S^n) ≥ ζ_m/2, and supplies the n = 1 and n = 2 cases that the paper re-derives.","marker":"[LMS23]"},{"why":"Establishes the quantitative Borsuk–Ulam theorem (equidiscontinuity of odd maps) that the self-contained lower-bound proof adapts.","marker":"[DS81]"},{"why":"Provides the dimension-independent upper bound π/3 and the circle-correspondence lower bounds δ(⌈d/2⌉) that the all-pairs joins match.","marker":"[ABC+26]"},{"why":"Proves the previously known n = 3 case by continuous geodesic interpolation, which the normalized-chord construction extends uniformly.","marker":"[RM26]"},{"why":"Constructs the exact circle correspondences P_d with distortion δ(⌈d/2⌉), used as factors in the all-pairs staircase bound.","marker":"[HJ23]"},{"why":"Supplies the sharp spherical Jung-type lemma used in the proof of the adjacent-sphere lower bound.","marker":"[Dek95]"}],"fun_headline_variants":["Adjacent sphere gap is exactly half a simplex edge","Distance between neighboring spheres: half a simplex edge","Consecutive spheres: GH distance equals half a simplex edge","Half a simplex edge: exact distance between adjacent spheres","Explicit matching shows sphere gap is half a simplex edge"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is a single algebraic inequality at the heart of the upper-bound proof: an explicitly written polynomial must never be positive when its parameters satisfy a stated condition, a sign claim verified only by a long symbolic factorization — if that sign claim were wrong for some n ≥ 2, the lower-boundary estimate and with it the main theorem would collapse.","fun_headline_variants_meta":{"raw":{"variants":["Adjacent sphere gap is exactly half a simplex edge","Distance between neighboring spheres: half a simplex edge","Consecutive spheres: GH distance equals half a simplex edge","Half a simplex edge: exact distance between adjacent spheres","Explicit matching shows sphere gap is half a simplex edge"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000901,"raw_usage":{"total_tokens":4029,"prompt_tokens":1245,"completion_tokens":2784,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":861,"completion_tokens_details":{"reasoning_tokens":2706}},"tokens_in":861,"tokens_out":2784,"duration_ms":23704,"temperature":1.0,"reasoning_tokens":2706,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:03:32.357113+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Check the explicit relation $R_4$ directly: either evaluate the polynomial $P_{\\rho_n}$ of Lemma 5.9 on a dense grid of parameters satisfying condition (52) for $n = 4$ and look for a positive value, or compute the realized defect maxima of $R_4$ on fine finite nets of $S^4$ and $S^5$. Any parameter point that satisfies the condition yet gives a positive polynomial would refute the lower-boundary proof, and any matched pair whose defect exceeds $\\zeta_4$ would refute optimality of the constructed correspondence.","supporting_citations":[],"review_version":1}