Hyper-Catalan Numbers in Combinatorial Analysis

• Hyper-Catalan numbers are a multivariate sequence counting roofed subdivided polygons by recording specific face profiles, such as triangles, quadrilaterals, and higher polygons.
• They feature an explicit closed form using factorial expressions that generalizes classical Catalan formulas and serves as the coefficient in a generating series zero of a geometric polynomial.
• The associated Geode factorization and recurrence relations offer computational methods and link hyper-Catalan numbers to classical convolution paradigms in combinatorial geometry.

Searching arXiv for papers on hyper-Catalan numbers and closely related Catalan generalizations. The hyper-Catalan number is the multivariate quantity

$C[m_2,m_3,m_4,\ldots],$

defined as the number of subdivisions of a roofed polygon into $m_2$ triangles, $m_3$ quadrilaterals, $m_4$ pentagons, and so on. In the recent combinatorial treatment based on subdivided roofed polygons, or subdigons, the index vector records the polygonal type of the subdivision; trailing zeros do not change the type, the all-zero type $[\,]$ corresponds to the null subdigon, and $C[\,]=1$ (Rubine, 6 Jul 2025). The topic has acquired renewed attention because the generating series of these numbers is a formal-series zero of a general geometric polynomial, which in turn yields recurrences, factorization phenomena, and new explicit formulas for associated coefficients called the Geode (Rubine, 6 Jul 2025).

1. Definition and combinatorial interpretation

The basic object counted by hyper-Catalan numbers is a roofed subdivided polygon. In the later finite-interpretation formulation, a subdigon is described as a convex planar polygon subdivided by noncrossing diagonals, with a distinguished roof edge; its type is the infinite vector

$[m_2,m_3,m_4,\ldots],$

where $m_k$ records the number of $(k+1)$-gons in the subdivision (Rubine et al., 8 Aug 2025).

The coefficient

$C[m_2,m_3,m_4,\ldots]$

therefore counts exactly those subdigons having $m_2$0 triangles, $m_2$1 quadrilaterals, $m_2$2 pentagons, and so forth. The indexing is intrinsically multivariate: the data do not collapse to a single size parameter, but instead retain the full face-type profile of the dissection. This distinguishes the hyper-Catalan array from one-parameter Catalan generalizations such as Fuss-Catalan or higher-order Catalan sequences.

The null type plays the role of the empty object. In the notation used for the 2025 recurrence paper,

$m_2$3

This normalization is consistent with the generating-series identity and with the recursive decomposition of subdigons into smaller subdigons (Rubine, 6 Jul 2025).

2. Closed form and its historical position

The closed form for hyper-Catalan numbers is attributed to Erdelyi and Etherington, 1940. It is

$m_2$4

This formula is explicitly described as the hyper-Catalan analogue of the classical Catalan and Fuss-Catalan formulas (Rubine, 6 Jul 2025).

The same coefficient formula is written in the finite-interpretation literature as the coefficient of the formal solution of the geometric polynomial,

$m_2$5

thereby identifying the coefficient directly with $m_2$6 (Mukewar, 26 Jul 2025).

A later presentation rewrites the same structure using explicit vertex, edge, and face counts: $m_2$7 with

$m_2$8

satisfying

$m_2$9

This formulation packages the same enumerative data in terms of subdigon statistics (Rubine et al., 8 Aug 2025).

3. Generating series and the geometric polynomial

The central generating object is the multivariate series

$m_3$0

Wildberger and Rubine’s 2025 result, used in the recurrence paper, states that this generating sum is a formal-series zero of the general geometric univariate polynomial

$m_3$1

so that

$m_3$2

In this sense, the hyper-Catalan generating series is characterized as a distinguished root of the polynomial determined by the variables $m_3$3 (Rubine, 6 Jul 2025).

The combinatorial reason for this identity is a recursive grammar for subdigons: $m_3$4 Here $m_3$5 is the null subdigon, and $m_3$6 merges $m_3$7 subdigons around a central $m_3$8-gon. Under the monomial map $m_3$9,

$m_4$0

so the combinatorial recursion becomes precisely the polynomial equation for the generating series (Mukewar, 26 Jul 2025).

This perspective is structurally important. It does not merely encode the coefficients; it explains why the hyper-Catalan array is naturally tied to polynomial equations of the form $m_4$1. The recent literature treats this link as the basis for both recurrence derivations and finite truncation results.

4. Recurrences and the Catalan convolution paradigm

Using the multinomial expansion of powers of $m_4$2, the 2025 recurrence paper derives a recurrence for hyper-Catalans that expresses each coefficient in terms of other hyper-Catalans with smaller indices (Rubine, 6 Jul 2025). The paper explicitly states that this generalizes the familiar Catalan convolution

$m_4$3

The multivariate recurrence is presented in the paper as the direct consequence of the identity $m_4$4. Its purpose is algorithmic as well as structural: hyper-Catalan coefficients can be recovered recursively from smaller types rather than only from the closed form.

A concrete example given in the paper is

$m_4$5

This example illustrates the multishape character of the recurrence: the type $m_4$6 simultaneously records one triangle and one quadrilateral (Rubine, 6 Jul 2025).

The recurrence viewpoint situates hyper-Catalan numbers near the classical Catalan tradition, but with the crucial difference that the index set is a type vector rather than a single integer. The convolution is therefore distributed across all admissible polygonal face profiles.

5. The Geode factorization and proved conjectures

A further development is the factorization

$m_4$7

or, in the shorthand used in the paper,

$m_4$8

The factor $m_4$9 is called the Geode, and the paper proves that there is a unique polyseries $[\,]$0 satisfying this factorization (Rubine, 6 Jul 2025).

The same paper derives a recurrence for Geode coefficients, expressing them in terms of other hyper-Catalan and Geode coefficients, and ultimately in terms of hyper-Catalans alone. It emphasizes that this gives an effective recursive method for computing Geode coefficients and that these coefficients can be expanded as integer combinations of hyper-Catalan numbers.

Three conjectural closed forms of Wildberger are proved.

First, for a type with only one nonzero component,

$[\,]$1

These Geode elements are therefore exactly Fuss-Catalan numbers.

Second, for the Bi-Tri case,

$[\,]$2

where

$[\,]$3

Third, for two consecutive shapes in general,

$[\,]$4

with

$[\,]$5

The Bi-Tri formula is the specialization $[\,]$6 of this general expression (Rubine, 6 Jul 2025).

The Geode is thus partly explicit and partly unresolved. The paper’s conclusion is that, although its coefficients can be recursively expanded into finite integer combinations of hyper-Catalan numbers, a closed form for the general Geode coefficient remains unknown, and so does its combinatorial meaning.

6. Finite-level interpretations, powers, and related usages of the term

The 2025 finite-interpretation papers shift attention from the infinite formal series to bounded truncations. By introducing layering variables $[\,]$7, $[\,]$8, and $[\,]$9 for vertices, edges, and faces, one obtains a layered polynomial

$C[\,]=1$0

together with layered versions of the hyper-Catalan generating series (Mukewar, 26 Jul 2025). The resulting statement is that the formal identity remains valid at each truncation level: $C[\,]=1$1 With an additional bound $C[\,]=1$2 on polygon degree, the face-truncated version becomes a genuinely finite identity: $C[\,]=1$3 The point of these results is that the formal series zero may be interpreted as a family of finite polynomial identities indexed by truncation level (Mukewar, 26 Jul 2025).

The follow-up paper on powers of the zero series studies $C[\,]=1$4. Its coefficients are called hyper-Catalan power numbers and are interpreted combinatorially in terms of subdigons with a specified central $C[\,]=1$5-gon (Rubine et al., 8 Aug 2025). The paper also derives the explicit formula

$C[\,]=1$6

presented as the hyper-Catalan analogue of the classical coefficient formula for powers of the Catalan generating function. In the same discussion, the classical warm-up identity is

$C[\,]=1$7

The terminology is not uniform across the Catalan literature. Several papers study generalized Catalan families without using the name hyper-Catalan. One paper defines generalized Catalan numbers

$C[\,]=1$8

recovering ordinary Catalans from the specialization $C[\,]=1$9 via

$[m_2,m_3,m_4,\ldots],$0

but states explicitly that it does not use the term “hyper-Catalan numbers” (Richardson, 2020). Another paper defines super Catalan numbers

$[m_2,m_3,m_4,\ldots],$1

and likewise does not define hyper-Catalan numbers (Georgiadis et al., 2011). In a different direction, hypergraph-based generalizations introduce $[m_2,m_3,m_4,\ldots],$2 (Gunnells, 2021) and the two-parameter Hypergraph Fuss-Catalan numbers $[m_2,m_3,m_4,\ldots],$3, with

$[m_2,m_3,m_4,\ldots],$4

(Chavan et al., 2022).

Accordingly, the current roofed-polygon usage of “hyper-Catalan number” is specific: it refers to the multivariate coefficients $[m_2,m_3,m_4,\ldots],$5 attached to polygonal type vectors. Within that framework, the coefficients themselves are well understood, the generating series is a formal zero of the geometric polynomial, several important Geode families have closed forms, and the outstanding unresolved point is the general Geode coefficient, whose closed form and combinatorial interpretation remain unknown (Rubine, 6 Jul 2025).