Papers
Topics
Authors
Recent
Search
2000 character limit reached

T-POP in Tamari Lattices

Updated 14 July 2026
  • T-POP is the specialization of Defant’s pop-stack-sorting operator applied to Tamari lattices, characterized by Dyck paths and 312-avoiding permutations.
  • It yields a rational generating function for t-Pop sortable elements and identifies the image with a Motzkin-number family.
  • The analysis leverages combinatorial models, lattice congruence techniques, and pattern avoidance to extend classical pop-stack sorting to structured lattices.

“T-POP” (Editor’s term) denotes the specialization of Defant’s pop-stack-sorting operator to Tamari lattices, that is, the dynamics of PopTamn\mathsf{Pop}_{\mathrm{Tam}_n} on the nn-th Tamari lattice. Defant defined, for each complete meet-semilattice MM, the operator

PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),

where ⋖\lessdot is the covering relation. In the Tamari setting, Tamn\mathrm{Tam}_n is the set of Dyck paths of semilength nn, or equivalently, 312-avoiding permutations under weak order, and the resulting dynamics connect Catalan structures, lattice-theoretic covers, pattern avoidance, and orbit enumeration. The central results are an explicit rational generating function for tt-Pop\mathsf{Pop}-sortable elements and the identification of the image of PopTamn\mathsf{Pop}_{\mathrm{Tam}_n} with a Motzkin-number family (Hong, 2022).

1. Definition in the Tamari setting

For nn0, the operator nn1 is the meet of nn2 together with all elements it covers. The minimal element of nn3 is denoted nn4. An element nn5 is called nn6-nn7-sortable if

nn8

and nn9 denotes the number of MM0-MM1-sortable elements in MM2 (Hong, 2022).

This formulation places T-POP in the broader program of extending classical pop-stack sorting from the symmetric group to lattice-theoretic and Coxeter-theoretic settings. In the Tamari case, the operator is especially tractable because the lattice admits several interchangeable models—Dyck paths, 312-avoiding permutations, and bracket-vector encodings—which support explicit structural and enumerative analysis. A later generalization to ornamentation lattices shows that when the underlying rooted plane tree is an MM3-element chain, the ornamentation lattice is the MM4-th Tamari lattice, so the Tamari case is the basic linear instance of a larger family (Ajran et al., 17 Jan 2025).

2. Enumeration of MM5-MM6-sortable elements

The principal enumerative theorem gives an explicit rational generating function: MM7 where MM8 is the MM9-th Catalan number. This verifies Defant’s conjecture that for fixed PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),0, the generating function is rational (Hong, 2022).

The significance of this formula is twofold. First, it identifies a closed-form rational structure in an iterated lattice-dynamical process that a priori need not have any such regularity. Second, it shows that the dependence on the iteration depth PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),1 is controlled by Catalan coefficients, reinforcing that the Tamari case is governed not merely by generic lattice theory but by the specific algebra of Catalan combinatorics. The theorem is not only an existence result: it gives a concrete series for the full family of counting sequences PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),2.

A further structural ingredient is the decomposition of PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),3-PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),4-sortable bracket vectors into irreducible components. If PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),5 is the generating function for all PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),6-PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),7-sortable elements and PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),8 the generating function for irreducible ones, then

PopM(x)=⋀({y∈M:y⋖x}∪{x}),\mathsf{Pop}_M(x)=\bigwedge(\{y\in M: y\lessdot x\}\cup \{x\}),9

This relation exposes the enumerative mechanism behind the rational formula: irreducibles behave as the primitive combinatorial units, and general elements assemble from them by a free-sequence type construction (Hong, 2022).

3. Image of the operator and Motzkin enumeration

A second main theorem concerns the image ⋖\lessdot0. The refined generating function is

⋖\lessdot1

At ⋖\lessdot2, the image size is a Motzkin number: ⋖\lessdot3 This settles a conjecture of Defant and Williams identifying the image size with a fundamental integer sequence (Hong, 2022).

The structural characterization of the image is equally important. Under the isomorphism between ⋖\lessdot4 and the lattice of 312-avoiding permutations under the right weak order, the image consists exactly of 312-avoiding permutations ending with ⋖\lessdot5 and with no double descents. The counting argument then passes through a reflection to 231-avoiding permutations and uses a known formula, attributed in the summary to Petersen, for permutations with ⋖\lessdot6 descents and ⋖\lessdot7 peaks: ⋖\lessdot8

This result clarifies that the image is not an opaque subset produced by an iterative meet operation. It has a direct pattern-avoidance characterization and a classical combinatorial enumeration. In later work on ornamentation lattices, the image of ⋖\lessdot9 and even the images of Tamn\mathrm{Tam}_n0 were further analyzed; for chains, this yields a complete characterization on Tamari lattices, and for Tamn\mathrm{Tam}_n1 the resulting generating function gives Motzkin numbers, recovering and generalizing Hong’s Tamari result (Ajran et al., 17 Jan 2025).

4. Proof architecture and combinatorial models

The original Tamari analysis proceeds by encoding Dyck paths as Tamn\mathrm{Tam}_n2-bracket vectors and describing the pop-stack operator explicitly through updates to these vectors. The argument then decomposes bracket vectors into irreducible components, derives generating-function recurrences for irreducible and general cases, and solves the resulting functional equations to obtain the rational formula for Tamn\mathrm{Tam}_n3 (Hong, 2022).

For the image theorem, the proof uses the weak-order model of Tamari lattices together with lattice congruence machinery. The summary specifically invokes results of Reading and Björner–Wachs: 312-avoiding permutations form sublattices and can be described as minimal representatives of sylvester-congruence classes. This transfer from Tamari-lattice elements to pattern-avoiding permutations is what makes the image characterization tractable.

These methods illustrate a recurring feature of T-POP. Although the operator is defined purely in terms of meets and covers, the effective analysis is carried out in auxiliary combinatorial languages: bracket vectors for iteration, weak-order embeddings for image structure, and classical pattern-avoidance statistics for enumeration. The Tamari case is therefore a meeting point of semilattice dynamics and Catalan combinatorics rather than a purely order-theoretic curiosity.

5. Extensions beyond the original Tamari analysis

Subsequent work substantially broadened the T-POP perspective. The paper on ornamentation lattices studies the pop-stack operator Tamn\mathrm{Tam}_n4 for a rooted plane tree Tamn\mathrm{Tam}_n5, where the ornamentation lattice of an Tamn\mathrm{Tam}_n6-element chain is the Tamn\mathrm{Tam}_n7-th Tamari lattice. It computes the maximum size of a forward orbit, characterizes the image of Tamn\mathrm{Tam}_n8, provides necessary conditions for membership in the image of Tamn\mathrm{Tam}_n9, and completely characterizes the image of nn0 on a Tamari lattice (Ajran et al., 17 Jan 2025).

In that framework, the Tamari case appears as the linear prototype. The summary emphasizes several structural notions—minimal reductions, sections, beads, and hugs. A “no-hug” criterion generalizes Hong’s characterization of the image for Tamari lattices, while a beads condition controls iterated images. For chains, the paper gives a complete characterization and an explicit generating function for nn1; for nn2 this is the Catalan generating function, and for nn3 it gives Motzkin numbers (Ajran et al., 17 Jan 2025).

A complementary line of work studies the image of nn4 on other lattices, including the weak order of type nn5, the Tamari lattice of type nn6, and the lattices of order ideals of the root posets of types nn7 and nn8. In each case the focus is the generating function

nn9

and the paper settles four conjectures of Defant and Williams on these images (Choi et al., 2022). This suggests that the Tamari case is the most explicit member of a larger image-enumeration program for pop operators on structured lattices.

T-POP in the Tamari sense should be distinguished from the Coxeter pop-tsack torsing operator, denoted tt0, defined for a finite irreducible Coxeter group tt1 with fixed Coxeter element tt2 by

tt3

where tt4 is the join in the noncrossing partition lattice of the reflections lying weakly below tt5 in absolute order. This operator is a “Bessis dual” version of the Coxeter pop-stack sorting map rather than the Tamari-lattice operator itself (Defant et al., 2021).

Its dynamics differ sharply from those of T-POP on Tamari lattices. In coincidental types and type tt6, the identity element is the unique periodic point of tt7, the maximum size of a forward orbit is the Coxeter number tt8, and the forward orbit of tt9 has size Pop\mathsf{Pop}0 and is isolated in the sense that none of the non-identity elements of the orbit have preimages lying outside of the orbit (Defant et al., 2021). A later paper resolves conjectures on enumerating elements with near-maximal orbit length and gives complete classifications in types Pop\mathsf{Pop}1, Pop\mathsf{Pop}2, and Pop\mathsf{Pop}3 (Li, 2022).

There is also an unrelated modern use of the acronym in machine learning: “T-POP: Test-Time Personalization with Online Preference Feedback,” an algorithm for LLM personalization via test-time alignment and dueling bandits (Qu et al., 29 Sep 2025). That usage is entirely separate from the Tamari-lattice literature. In combinatorics and lattice dynamics, T-POP is best understood as shorthand for the Tamari specialization of Defant’s pop-stack operator, centered on rational Pop\mathsf{Pop}4-sortable enumeration, Motzkin-characterized images, and a growing web of generalizations across Catalan and Coxeter structures.

Topic to Video (Beta)

No one has generated a video about this topic yet.

Whiteboard

No one has generated a whiteboard explanation for this topic yet.

Follow Topic

Get notified by email when new papers are published related to T-POP.