Taking the amplituhedron to the limit

Joris Koefler; Rainer Sinn

REVIEW 4 major objections 4 minor 1 cited by

Taking the amplituhedron to the limit

T0 review · 4 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The infinite-particle amplituhedron is a positive geometry with two boundary components.

desk verdict A genuinely new and attractive construction in positive geometry, but the proof of empty residual arrangement has a concrete dimension error in Lemma 4.5 that needs fixing before the main theorem can be trusted. read the letter →

arxiv 2501.08221 v1 pith:3LNXQ4NL submitted 2025-01-14 math.AG hep-th

Joris Koefler , Rainer Sinn This is my paper · ORCID

classification math.AGhep-th MSC 14M1514N1514P10

keywords amplituhedronpositivegeometrycanonicalformChowhypersurfacerationalnormalcurveGrassmanniannonnegativescatteringamplitudes

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that when the number of particles n is sent to infinity, the m=2 amplituhedron converges, in a well-defined direct-limit sense, to a semialgebraic set A∞_k inside the Grassmannian Gr(k,k+2), and that this limit amplituhedron is a positive geometry. If true, this means the limiting object carries a unique rational canonical form with simple poles exactly along its algebraic boundary, a property that is known for polytopes and for the k=m=2 amplituhedron but was open for larger k in this limit. The boundary is shown to be the union of two Chow hypersurfaces: one attached to the rational normal curve C_{k+1} and one to the secant line S01 spanned by its endpoints. The paper works out the full singularity stratification of this boundary and proves the residual arrangement is empty, so the canonical form is the reciprocal of the product of the two Chow forms.

What carries the argument

The argument is carried by the limit amplituhedron itself, defined as the union, over all partitions of the interval [0,1], of the amplituhedra obtained from totally positive matrices whose rows are sampled from the rational normal curve γ_{k+1}(t) = (1,t,...,$t^{{k+1}}$). The algebraic boundary is computed by taking Zariski closures of Euclidean boundary strata, which are characterized by k-spaces meeting the rational normal curve or the secant line S01; these give the Chow hypersurfaces. The proof then analyzes the singular stratification via incidence varieties of secant and osculant planes, that is, higher-order tangent planes to the rational normal curve, using tangent-space descriptions of secant varieties. The key mechanism for proving the residual arrangement empty is Lemma 4.5, which constructs families of k-spaces spanned by rows of Z(I_n) so that, as partitions refine, the osculant planes at the endpoints are spanned in the limit; this ensures every stratum is Zariski dense in the amplituhedron.

What would settle it

Take k=3 and compute, symbolically or numerically, the 1-dimensional strata listed in Proposition 4.10 together with the Euclidean closure of A∞_3; if any point of those strata lies outside the closure, the residual arrangement is nonempty and the claimed canonical form fails. A positive check would verify that the residue of Ω(A∞_3) at every such 1-dimensional stratum is a one-form with simple poles at the vertices, with residues ±1.

Watch

Extended reading notes

Core claim

The central discovery is Theorem 1.1: the limit amplituhedron A∞_k is a positive geometry on Gr(k,k+2), with algebraic boundary ∂_a A∞_k = CH(C_{k+1}) ∪ CH(S01). Here C_{k+1} is the rational normal curve of degree k+1, S01 is the secant line through the images of 0 and 1, and CH(X) denotes the Chow hypersurface of k-spaces meeting X. The authors prove that the residual arrangement, meaning the part of the singular locus lying outside the Euclidean boundary, is empty, so the unique adjoint is the constant 1. Consequently the canonical form is Ω(A∞_k) = 1/(CF(C_{k+1}) CF(S01)) dx_1 ∧ ... ∧ dx_{2k} in local coordinates, with simple poles along exactly the two boundary components.

Load-bearing premise

The load-bearing premise is that, as partitions of [0,1] are refined, the k-spaces built from sampled points on the curve exactly span the higher-order tangent planes at the two endpoints; if that spanning fails, some singular stratum is missed, the residual arrangement is nonempty, and the canonical form would gain extra poles.

Editorial extensions

If this is right

For every k and m=2, the infinite-particle limit has a unique canonical form Ω(A∞_k)=1/(CF(C_{k+1}) CF(S01)) times the volume form, with simple poles only along the algebraic boundary.
The algebraic boundary of the limit amplituhedron is exactly the union of the two Chow hypersurfaces, so the canonical denominator is determined by a rational normal curve and a secant line.
The singular stratification of the boundary is completely described by higher secant and osculant varieties; iterated singular loci are themselves secant varieties, generalizing the matroid of a polytope's facet arrangement.
The k=1 case recovers the pizza slice, a positive geometry bounded by a parabola segment and its secant line, so the theorem unifies the previously known examples.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial: the same limiting construction with m>2 would likely replace the two Chow hypersurfaces by Chow forms of higher-dimensional varieties attached to the rational normal curve, but the paper does not treat that case.
Editorial: the paper's emphasis on the interval [0,1] suggests the endpoint secant line is what repairs positivity; without it, the convex hull of the full rational normal curve would be a disk-like set with no rational canonical form, so endpoint data may be a general mechanism.
Editorial: an explicit testable extension is to compute the canonical form for k=3 from the stratification and check the residue conditions at all vertices, extending the k=2 symbolic computation reported in the paper.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The n→∞ limit amplituhedron is a real idea, and the boundary description in Theorem 2.7 is pretty. The paper takes the m=2 case, defines A∞_k as the union of amplituhedra over all partitions of [0,1], and shows its algebraic boundary is the union of the Chow hypersurfaces of the rational normal curve and one secant line. That is a clean and plausible statement, and the stratification in Section 3 via secant and osculant planes is the right tool; it gives a full picture for every k, not just k=2. If the main result holds, it is a new infinite family of positive geometries with explicit canonical forms.

The soft spot is real and specific. Lemma 4.5 claims that A∞_k is Zariski dense in the strata O^ℓ_{0,j} and O^ℓ_{1,j}. The proof constructs Y(In) with ℓ+1 rows: in (4.1) there are γ(0),...,γ(t_j) (j+1 rows), γ(t_{i1}),...,γ(t_{ic}) (c = ℓ−j−1 rows), and γ(1). For any fixed n these are distinct points on the rational normal curve, so they span ℓ+1 dimensions, not ℓ. Then V_Y has dimension 2k−2ℓ−2, not 2k−2ℓ, and the union over s has dimension 2k−ℓ−j−3, two short of dim O^ℓ_{0,j}. The density conclusion does not follow. The likely fix is to drop the endpoint γ(1) (or γ(0) in the symmetric case); the osculant plane alone should give the line intersection. That looks plausible, but the printed argument has a hole, and this is the only step that puts A∞_k inside the higher strata, so Theorem 4.1 and the positive-geometry conclusion ride on it.

Smaller issues: Proposition 2.3 invokes compactness before closedness of A∞_k is established (Proposition 2.4), and the Zariski-density step in Theorem 2.7 is compressed. Both look fixable. The reliance on [RST24] is legitimate prior work, though a referee may want the dependence stated more explicitly.

Bottom line: the idea is good, the main theorem is probably right, and the flaw is addressable. It deserves a serious referee and a request for revision, not a desk rejection.

Referee Report

4 major / 4 minor

Summary. The paper defines the limit amplituhedron A^∞_k as the union, over all partitions I of the interval [0,1], of the finite amplituhedra A^I_k constructed from totally positive matrices whose rows lie on the rational normal curve C_{k+1} ⊂ P^{k+1}, with m=2. The main theorem (Theorem 1.1) states that A^∞_k is a positive geometry in Gr(k,k+2) whose algebraic boundary is the union of the Chow hypersurface of C_{k+1} and the Chow hypersurface of the secant line S_{01}. Section 2 identifies this algebraic boundary, Section 3 stratifies it by varieties of secants and osculant planes, and Section 4 claims the residual arrangement is empty, yielding a canonical form equal to the reciprocal of the product of the two Chow forms.

Significance. If the proof can be made rigorous, this is a significant contribution: it provides the first positive-geometry statement for an infinite-particle limit of amplituhedra in the m=2 case, identifies the boundary components explicitly with Chow hypersurfaces, and gives a stratification by higher secants and osculant planes that is of independent interest. The paper also contains a useful explicit computation for k=2 in Example 2.9. However, several load-bearing arguments in the current version have concrete dimension-count or foundational gaps, so the stated results are not yet established as written.

major comments (4)

[Lemma 4.5, Eqs. (4.1) and (4.2)] The displayed ℓ-space Y(I_n) in (4.1) has ℓ+1 rows: the j+1 rows γ(0),…,γ(t_j), the c=ℓ−j−1 rows γ(t_i1),…,γ(t_ic), and the additional row γ(1). For distinct nodes on the rational normal curve these rows are linearly independent, so Y(I_n) has dimension ℓ+1, not ℓ. Consequently the fiber V_{Y(I_n)} = {V ∈ Gr(k,k+2) : Y(I_n) ⊂ V} has dimension 2(k−ℓ−1), not 2k−2ℓ as asserted in the proof. The union over the parameters s_1,…,s_c therefore has dimension at most 2k−ℓ−j−3, whereas dim(O^ℓ_{0,j}) = 2k−ℓ−j−1; when k=ℓ the fiber is empty. The same problem occurs in (4.2) for O^ℓ_{1,j}. Since Lemma 4.5 is the only step that places A^∞_k inside the higher strata O^ℓ_{0,j} and O^ℓ_{1,j}, Theorem 4.1 and Corollary 4.9 do not follow as written. A likely repair is to delete the endpoint γ(1) from (4.1) and γ(0) from (4.2) so that Y(I_n) is genuinely ℓ-dimensional, but this must be stated explicitly and all subsequent dimension counts redone.
[Lemmas 4.3 and 4.4] The dimension counts attributed to the fibers of V_{Y(s)} are off by ℓ. In Lemma 4.3 the text says the fiber of V_{Y(s)} over s has dimension 2k−ℓ, which equals dim(Sec^ℓ(C_{k+1})); however, for an ℓ-dimensional linear subspace Y, V_{Y(s)} is a Grassmannian of dimension 2(k−ℓ). The missing 2ℓ dimensions come from varying the ℓ points on the rational normal curve, so the intended conclusion may be repairable, but the written equality is false. The same type of error appears in Lemma 4.4, where the fiber dimension is written as 2k−ℓ−1 although dim(V_{Y(0,s)}) = 2(k−ℓ). These are used to conclude Zariski density of A^∞_k in the secant strata, so the proofs of Lemmas 4.3 and 4.4 are not valid as written.
[Proposition 2.3] The proof invokes compactness of A^∞_k : 'A^∞_k is compact by Tychonoﬀ’s theorem'. This is unjustified, since A^∞_k is a union over partitions rather than a product, and its closedness in the ambient Grassmannian is not known at that point. Proposition 2.4, which establishes closedness, in turn uses Proposition 2.3, so the interior formula and the boundary description rest on a circular compactness claim. The argument needs to be reworked, for instance by proving closedness of A^∞_k directly from the boundary description in Lemma 2.12, or by establishing Proposition 2.4 independently and then deriving Proposition 2.3.
[Proof of Theorem 2.7] The map φ: [0,1] → A^∞_k, t ↦ Y_t is not a well-defined map to the Grassmannian, because Y_t is a subvariety of V_t rather than a point of Gr(k,k+2). Consequently, the sentence 'the image φ([0,1]) is Zariski dense in CH(C_{k+1})' is not meaningful as written; what is needed is an argument that the union ∪_{t∈[0,1]} Y_t is Zariski dense in CH(C_{k+1}). The proof also does not establish irreducibility or the dimension of this union. In addition, the statement in Lemma 2.13 that codim(̊Z(Y)) = 2 is inconsistent with the dimension count for the family Y from Lemma 2.12, which has dimension 2k−1 and hence has image of codimension at most 1 in Gr(k,k+2). These issues must be clarified for the algebraic-boundary theorem to be rigorous.

minor comments (4)

[Abstract] The abstract contains a typo: 'negative helcity particles' should read 'negative helicity particles'.
[Example 2.9, Table 1] The notation 'Tγ3(0)(C3)' should be written as T_{γ3(0)}C3 or similar, and the table could indicate more clearly that the same point S01 appears in several strata as the secant line itself rather than as a hypersurface component.
[Proposition 2.2] The direct-limit construction sets v'_i = 0 for newly inserted indices; the text should explicitly state that the resulting inclusion maps are well-defined because the images in Gr(k,k+2) depend only on the product with the rows of Z(I), not on the ambient n of the original nonnegative Grassmannian.
[Section 2, notation] The symbols ∂a and ∂ for algebraic and Euclidean boundary are introduced but used somewhat interchangeably in a few places; for instance, the definition '∂aAn,k is the Zariski closure ∂An,k of the boundary' would be clearer if the two boundary notions were distinguished consistently throughout.

Circularity Check

1 steps flagged · score 2.0 of 10

Self-citation supplies a key boundary input, but the central limit construction remains independent; no definitional or fitted-input circularity is present.

self citation load bearing [Section 2, Lemma 2.12, and its use in Proposition 2.4 and Theorem 2.7]
"Similarly there is an immediate generalization of [RST24, Theorem 3.1]. ... Proof. We follow [RST24]. ... It remains to show that there are no other boundary strata ... This is immediate from the proof of [RST24, Theorem 3.1] and Lemma 2.11."

The classification of the finite amplituhedron boundary is a load-bearing input: Proposition 2.4 uses Lemma 2.12 to identify all Euclidean boundary strata of the finite amplituhedra, and Theorem 2.7 then derives the limit boundary from that. The lemma is not independently reproved here; it is imported from [RST24], whose authors include Sinn, an author of the present paper. Thus a key premise in the derivation chain is supported by a self-citation rather than by a self-contained proof in this text. The dependence is partial, however: the limit amplituhedron, the Chow-hypersurface description, the stratification, and the residual-arrangement argument are new content that does not reduce to [RST24].

full rationale

No definitional circularity was found: A∞_k is defined as a union of finite amplituhedra over partitions, and the algebraic boundary components CH(C_{k+1}) and CH(S01) are computed from Euclidean boundary strata, not built into the definition. No fitted-parameter-then-predicted relationship appears. The main theorem is not a renaming of a known result: although the k=1 case is the known 'pizza slice', the general-k limit amplituhedron and its Chow-hypersurface boundary are new. The only noteworthy circularity-adjacent issue is the reliance on [RST24] in Lemma 2.12 for the finite boundary description. Since one current author is also an author of [RST24], this is a self-citation; it is load-bearing for Proposition 2.4 and Theorem 2.7, but it is a prior, independently stated result, and the paper's central contribution—the limit construction, stratification, and positivity—does not collapse into it. A separate review concern is that Y(In) in (4.1)/(4.2) appears to have ℓ+1 rows, so the stated dimension 2k−2ℓ may be wrong; that is a correctness matter, not circularity, and under the hard rules it does not raise the circularity score. Overall: one minor self-citation with independent central content, score 2.

Assumptions & free parameters 1 free parameters · 6 assumptions · 1 invented entities

The main construction depends on a hand-chosen interval [0,1] on the rational normal curve, a black-box boundary theorem from [RST24], and standard tools in algebraic geometry and positive geometry. No numerical parameters are fitted to data. The only invented entity is the limit amplituhedron itself, which is explicitly defined rather than assumed.

free parameters (1)

Interval [0,1] on the rational normal curve = [0,1]
The limit amplituhedron is defined only for partitions of [0,1]; choosing a different interval changes the algebraic boundary and can destroy the positive geometry (using the full curve gives the unit disk, which has no rational canonical form). This is a hand-chosen input to the construction, not a number fitted to data.

assumptions (6)

standard math Standard facts about rational normal curves, secant varieties, and Grassmannian tangent spaces
Used throughout Sections 2 and 3, for example in the definitions of Chow hypersurfaces and in the tangent-space computations in Section 3.1.
standard math Generalized Vandermonde determinant formula for osculant planes (Remark 3.3)
This formula, cited from [Cha18], guarantees that osculant planes at distinct points of the rational normal curve are transversal; it is used in Proposition 3.4 and Proposition 3.8.
domain assumption Boundary description of finite m=2 amplituhedra (Lemma 2.12, from [RST24, Theorem 3.1])
The algebraic boundary of the finite amplituhedron A_{n,k,2} is used as a black box for all k and n, extended by a circular-shift argument. This is a heavy cited premise for Theorem 2.7.
domain assumption The restriction to totally positive matrices Z(I) whose rows lie on the rational normal curve (Remark 2.1)
The paper only treats these special realizations of cyclic polytopes; the authors acknowledge that not every cyclic polytope is affinely equivalent to such a realization.
ad hoc to paper Choice of the interval [0,1] on the rational normal curve
The interval is explicitly chosen so that the secant line S01 between the endpoints raises the degree of the denominator to k+2, which is needed for a rational canonical form. The paper states this is necessary for the positive-geometry conclusion.
domain assumption Positive geometry and adjoint theory from [ABL17, Lam22, Koh+24]
The recursive definition of positive geometry, the adjoint interpolation condition, and the role of the residual arrangement are imported from the cited literature and used in Section 4.

invented entities (1)

Limit amplituhedron A^∞_k
purpose: A new semialgebraic set in Gr(k,k+2) defined as the direct limit of finite amplituhedra as the number of particles n goes to infinity; the main object of study.
This is a newly defined mathematical object. It is constructed explicitly and used to derive the boundary and positive-geometry theorems, not postulated to explain an otherwise unconnected observation.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Taking the amplituhedron to the limit." pith.science (2026). https://pith.science/paper/3LNXQ4NL

@misc{pith2026250108221,
  author       = {Pith},
  title        = {Pith review of: Taking the amplituhedron to the limit},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3LNXQ4NL}},
  note         = {Machine review of arXiv:2501.08221}
}

abstract

The amplituhedron is a semialgebraic set given as the image of the non-negative Grassmannian under a linear map subject to a choice of additional parameters. We define the limit amplituhedron as the limit of amplituhedra by sending one of the parameters, namely the number of particles $n$, to infinity. We study this limit amplituhedron for $m = 2$ and any $k$, relating to the number of negative helcity particles. We determine its algebraic boundary in terms of Chow hypersurfaces. This hypersurface in the Grassmannian is stratified by singularities in terms of higher order secants of the rational normal curve. In conclusion, we show that the limit amplituhedron is a positive geometry with a residual arrangement that is empty.

Figures

Figures reproduced from arXiv: 2501.08221 by the authors.

**Figure 1.** The limit amplituhedron A∞ 1 with its algebraic boundary. would be the convex hull of the rational normal curve, so up to a projective change of coordinates the unit disk. This is, however, not a positive geometry in P2 because it cannot have a rational canonical form (for instance for degree reasons: there is no rational section of the canonical divisor of P2 with a denominator of degree 2), see [ABL17, Section 10… view at source ↗

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Geometric Landau Analysis and Symbol Bootstrap
hep-th 2025-08 unverdicted novelty 7.0 of 10

Boundary structure of negative geometries, combined with Landau analysis, determines physical singularities and yields symbol alphabets for six-point two-loop and five-point three-loop ladder integrals in planar N=4 s...

Reference graph

Works this paper leans on

6 extracted references · 2 canonical work pages · cited by 1 Pith paper

[1]

Pos itive geometries and canonical forms

[ABL17] Nima Arkani-Hamed, Yuntao Bai, and Thomas Lam. “Pos itive geometries and canonical forms”. In: Journal of High Energy Physics (Nov. 2017). issn: 1029-

work page 2017
[6]

Totally positive matrices and cy clic polytopes

arXiv: 2402.06527. [Stu88] Bernd Sturmfels. “Totally positive matrices and cy clic polytopes”. In: Linear Algebra and its Applications 107 (1988), pp. 275–281. issn: 0024-3795. url: https://www.sciencedirect.com/science/article/pii/0024379588902509. Authors’ addresses: Joris Koeﬂer, MPI-MiS Leipzig joris.koefler@mis.mpg.de Rainer Sinn, University of Leipz...

arXiv 1988
[1992]

url: http://dx.doi.org/10.1007/978-1-4757-2189-8

isbn: 9781475721898. url: http://dx.doi.org/10.1007/978-1-4757-2189-8 . [Koh+24] Kathlén Kohn et al. Adjoints and Canonical Forms of Polypols

doi:10.1007/978-1-4757-2189-8
[2022]

21 [RST24] Kristian Ranestad, Rainer Sinn, and Simon Telen

url: https://arxiv.org/abs/2208.05407. 21 [RST24] Kristian Ranestad, Rainer Sinn, and Simon Telen. Adjoints and Canonical Forms of Tree Amplituhedra

arXiv
[2024]

url: https://arxiv.org/abs/2108.11747

arXiv: 2108.11747. url: https://arxiv.org/abs/2108.11747. [Lam22] Thomas Lam. An invitation to positive geometries

arXiv
[8479]

The Amplituh edron

url: http://dx.doi.org/10.1007/JHEP11(2017)039. [AT14] Nima Arkani-Hamed and Jaroslav Trnka. “The Amplituh edron”. In: Jour- nal of High Energy Physics 10 (Oct. 2014). issn: 1029-8479. url: http://dx.doi.org/10.1007/JHEP10(2014)030. [Cha18] Sungkon Chang. “Generalized Vandermonde Matrix”. In: Pi Mu Epsilon Journal 14 (2018), pp. 549–556. issn: 0031952X. u...

work page arXiv 2017

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [1]

Pos itive geometries and canonical forms

[ABL17] Nima Arkani-Hamed, Yuntao Bai, and Thomas Lam. “Pos itive geometries and canonical forms”. In: Journal of High Energy Physics (Nov. 2017). issn: 1029-

work page 2017

[2] [6]

Totally positive matrices and cy clic polytopes

arXiv: 2402.06527. [Stu88] Bernd Sturmfels. “Totally positive matrices and cy clic polytopes”. In: Linear Algebra and its Applications 107 (1988), pp. 275–281. issn: 0024-3795. url: https://www.sciencedirect.com/science/article/pii/0024379588902509. Authors’ addresses: Joris Koeﬂer, MPI-MiS Leipzig joris.koefler@mis.mpg.de Rainer Sinn, University of Leipz...

arXiv 1988

[3] [1992]

url: http://dx.doi.org/10.1007/978-1-4757-2189-8

isbn: 9781475721898. url: http://dx.doi.org/10.1007/978-1-4757-2189-8 . [Koh+24] Kathlén Kohn et al. Adjoints and Canonical Forms of Polypols

doi:10.1007/978-1-4757-2189-8

[4] [2022]

21 [RST24] Kristian Ranestad, Rainer Sinn, and Simon Telen

url: https://arxiv.org/abs/2208.05407. 21 [RST24] Kristian Ranestad, Rainer Sinn, and Simon Telen. Adjoints and Canonical Forms of Tree Amplituhedra

arXiv

[5] [2024]

url: https://arxiv.org/abs/2108.11747

arXiv: 2108.11747. url: https://arxiv.org/abs/2108.11747. [Lam22] Thomas Lam. An invitation to positive geometries

arXiv

[6] [8479]

The Amplituh edron

url: http://dx.doi.org/10.1007/JHEP11(2017)039. [AT14] Nima Arkani-Hamed and Jaroslav Trnka. “The Amplituh edron”. In: Jour- nal of High Energy Physics 10 (Oct. 2014). issn: 1029-8479. url: http://dx.doi.org/10.1007/JHEP10(2014)030. [Cha18] Sungkon Chang. “Generalized Vandermonde Matrix”. In: Pi Mu Epsilon Journal 14 (2018), pp. 549–556. issn: 0031952X. u...

work page arXiv 2017