Kinematic Associahedron in Scattering Amplitudes

• Kinematic associahedron is a polytope in kinematic space that uses planar Mandelstam invariants to represent bracketings of scattering processes.
• The construction provides a unique canonical form with logarithmic singularities that mirrors the recursive factorization of tree-level Feynman diagrams.
• It bridges positive geometry, cluster algebras, and worldsheet formulations, with extensions to deformed realizations for multi-scalar and gauge theories.

The kinematic associahedron is a realization of the associahedron—a classical simple polytope encoding bracketings or non-crossing partitions—in the space of kinematic invariants describing scattering processes in quantum field theory. This construction manifests the combinatorial, geometric, and algebraic structures underlying tree-level scattering amplitudes, particularly those of planar bi-adjoint scalar $\phi^3$ theory, and bridges positive geometry, canonical forms, and cluster algebras with physical notions such as locality and unitarity. The associahedron in kinematic space provides a unique, positive-geometric basis for understanding amplitude singularities, factorization, and connections to worldsheet formulations such as the CHY (Cachazo-He-Yuan) formalism. Recent works extend this framework to deformed realizations relevant for multi-scalar cubic theories and relate the kinematic associahedron to positive geometries in gauge theory, notably the momentum amplituhedron.

1. Kinematic Space and the Associahedron Construction

Kinematic space $\mathcal{K}_n$ is spanned by planar Mandelstam invariants $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$ for $1 \leq i < j \leq n$, subject to linear momentum conservation constraints $\sum_{j \neq i} s_{i,j} = 0$, leading to $\dim \mathcal{K}_n = n(n-3)/2$ independent variables for $n$ massless cyclically ordered particles (Damgaard et al., 2020, Arkani-Hamed et al., 2017, Torres, 2017).

For a planar ordering, variables $$X_{i,j} := s_{i,i+1,\dots,j-1} = (p_i + ... + p_{j-1})^2$$ naturally label the diagonals of an $n$-gon. One defines a simplicial positive region $\Delta_n \subset \mathcal{K}_n$ via $\mathcal{K}_n$0, imposing positivity of all planar channels. An $\mathcal{K}_n$1-dimensional affine subspace $\mathcal{K}_n$2 is carved out by linear constraints:

$\mathcal{K}_n$3

for positive constants $\mathcal{K}_n$4 and all non-adjacent $\mathcal{K}_n$5. The intersection $\mathcal{K}_n$6 yields a simple $\mathcal{K}_n$7-dimensional polytope whose face structure coincides with the classical Stasheff associahedron (Torres, 2017).

The facets $\mathcal{K}_n$8 correspond to propagators going on-shell, or equivalently to the vanishing of certain planar Mandelstam variables, matching tree-level factorization channels of the amplitude. Each co-dimension-1 face can be identified with the factorization of the associahedron into lower-point associahedra, encapsulating the recursive structure of tree-level Feynman diagrams (Damgaard et al., 2020, Arkani-Hamed et al., 2017).

2. Canonical Form, Amplitude, and Singularities

On a simple polytope $\mathcal{K}_n$9 of dimension $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$0, there exists a unique log-canonical top-form ("canonical form") $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$1 with logarithmic singularities on each facet and unit residues. For the kinematic associahedron $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$2, the canonical form is

$s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$3

where $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$4 runs over all planar binary trees (equivalently, all triangulations of the $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$5-gon) (Arkani-Hamed et al., 2017, Torres, 2017, Damgaard et al., 2020). Alternatively, the canonical rational function for the $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$6 amplitude is

$s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$7

This function encodes all physical singularities: simple poles on the facets $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$8, corresponding to physical factorization channels.

When evaluated on a face where $s_{i,j} = (p_i + p_{i+1} + \cdots + p_j)^2$9, the canonical form factorizes:

$1 \leq i < j \leq n$0

This property directly mirrors the factorization of the associated scattering amplitude, providing a geometric origin for both locality and unitarity (Arkani-Hamed et al., 2017, Damgaard et al., 2020).

3. Cluster Algebraic Structure

The kinematic associahedron realizes a cluster algebra of type $1 \leq i < j \leq n$1 in kinematic space (Torres, 2017). The variables $1 \leq i < j \leq n$2 correspond to cluster variables associated with diagonals of an $1 \leq i < j \leq n$3-gon. Each triangulation gives rise to a cluster, and mutations (diagonal flips) correspond to facet adjacency in the polytope:

$1 \leq i < j \leq n$4

for $1 \leq i < j \leq n$5, representing the tropicalization of cluster algebraic exchange (Torres, 2017). The collection of all $1 \leq i < j \leq n$6 clusters produces the full vertex set of the associahedron.

A hypercube necklace of $1 \leq i < j \leq n$7 subalgebras (snake clusters) is embedded within the polytope, and their adjacency graph forms a cycle that reflects certain symmetry properties and subcluster structures. For low $1 \leq i < j \leq n$8, explicit realization of the associahedron in cluster terms directly yields known dilogarithmic identities and cluster polylogarithms (e.g., the Abel pentagon identity for $1 \leq i < j \leq n$9).

4. Deformed Realizations and Multi-Scalar Generalizations

The construction generalizes to "deformed associahedra" relevant for massive or mixed-scalar theories (Jagadale et al., 2022, Mahato et al., 19 Jul 2025). Here, kinematic variables are shifted and scaled:

$\sum_{j \neq i} s_{i,j} = 0$0

with deformation parameters $\sum_{j \neq i} s_{i,j} = 0$1 and shifts $\sum_{j \neq i} s_{i,j} = 0$2. The ABHY constraints become

$\sum_{j \neq i} s_{i,j} = 0$3

while positivity becomes $\sum_{j \neq i} s_{i,j} = 0$4. These deformed polytopes encode amplitudes for multi-scalar cubic couplings, with canonical forms weighted by $\sum_{j \neq i} s_{i,j} = 0$5 that reflect the strengths of cubic interactions (Jagadale et al., 2022, Mahato et al., 19 Jul 2025).

Tree-level amplitudes in theories with mixed couplings are expressible as weighted sums of canonical forms over multiple deformed associahedra. Extension to loop level is achieved via D-type cluster polytopes (halohedra), and BCFW-like recursion constructs the canonical form via projective triangulations of the polytope, with residues corresponding to physical factorization (Mahato et al., 19 Jul 2025).

5. Worldsheet Image and CHY Formalism

There exists a diffeomorphism between the moduli space of real ordered points $\sum_{j \neq i} s_{i,j} = 0$6 (the "worldsheet associahedron") and the kinematic associahedron (Arkani-Hamed et al., 2017, Jagadale et al., 2022). The Parke–Taylor form on $\sum_{j \neq i} s_{i,j} = 0$7,

$\sum_{j \neq i} s_{i,j} = 0$8

pushes forward to $\sum_{j \neq i} s_{i,j} = 0$9 under the scattering equations

$\dim \mathcal{K}_n = n(n-3)/2$0

Similarly, deformed scattering equations generate diffeomorphic images of the deformed associahedra in kinematic space. In this sense, the universality of the Parke–Taylor form in the CHY formalism is a direct manifestation of the geometry of the kinematic associahedron (Jagadale et al., 2022).

6. Relation to Other Positive Geometries

A central outcome is the precise relation between the canonical form of the kinematic associahedron in $\dim \mathcal{K}_n = n(n-3)/2$1 theory and the sum of reduced canonical forms of momentum amplituhedra in $\dim \mathcal{K}_n = n(n-3)/2$2 SYM:

$\dim \mathcal{K}_n = n(n-3)/2$3

after removing GL$\dim \mathcal{K}_n = n(n-3)/2$4 redundancy and pulling back to kinematic space (Damgaard et al., 2020). Both structures share the same singularity loci, dictated by $\dim \mathcal{K}_n = n(n-3)/2$5, and thus the same factorization channels.

Higher-$\dim \mathcal{K}_n = n(n-3)/2$6 generalizations via matroid subdivisions and weak separation yield the generalized kinematic associahedron, intimately connected with generalized Feynman diagrams, and encapsulate compatibility conditions for poles in broader families of amplitude geometries (Early, 2019).

7. Illustrative Examples and Applications

Explicit realizations for $\dim \mathcal{K}_n = n(n-3)/2$7 (pentagon, two-dimensional) and $\dim \mathcal{K}_n = n(n-3)/2$8 (hexagon/cyclohedron, three-dimensional) clarify the geometric and combinatorial organization of tree-level amplitudes (Arkani-Hamed et al., 2017, Jagadale et al., 2022, Mahato et al., 19 Jul 2025). For example, the $\dim \mathcal{K}_n = n(n-3)/2$9 canonical form decomposes as a sum over all triangulations, each term corresponding to a planar cubic diagram:

$n$0

In the context of effective field theory limits, certain deformations correspond geometrically to projections onto lower-dimensional "accordiohedra" or "Stokes polytopes," yielding higher-point contact interactions and encoding the geometric origin of kinematic singularities in the EFT amplitude (Mahato et al., 19 Jul 2025).

---

References (arXiv IDs):

• (Arkani-Hamed et al., 2017) Scattering Forms and the Positive Geometry of Kinematics, Color and the Worldsheet
• (Torres, 2017) Cluster Algebras in Kinematic Space of Scattering Amplitudes
• (Early, 2019) Planar kinematic invariants, matroid subdivisions and generalized Feynman diagrams
• (Damgaard et al., 2020) Momentum Amplituhedron meets Kinematic Associahedron
• (Jagadale et al., 2022) Towards Positive Geometries of Massive Scalar field theories
• (Mahato et al., 19 Jul 2025) BCFW like recursion for Deformed Associahedron