🔢 mathematics

Non-invertible symmetries and modular invariance in lattice models

This paper presents a generic algorithm to decompose the transfer-matrix space of classical 2d lattice models with Temperley-Lieb face interactions into simple modules, enabling the analysis of topological operator actions and the computation of modular transformations for irreducible characters at primitive roots of unity.

Original authors: Yacine Ikhlef

Published 2026-07-01

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yacine Ikhlef

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are looking at a giant, intricate quilt made of thousands of tiny squares. This isn't just any quilt; it's a mathematical model of how particles or "spins" interact in a two-dimensional world. This paper, written by Yacine Ikhlef, is like a master pattern book that explains how to take apart this quilt, understand its hidden symmetries, and predict how it behaves when you twist or turn it.

Here is a breakdown of the paper's main ideas using everyday analogies:

1. The Quilt and the Rules (The Lattice Model)

Think of the "lattice model" as a grid of tiles. Each tile has a specific color or type (like a red square or a blue circle). The rules of the game say that certain colors can only sit next to certain other colors.

The Fusion Category: This is the "rulebook" that tells you which colors can touch. It's like a social club where some members are friends and others aren't.
The Temperley-Lieb (TL) Algebra: This is the specific set of mathematical rules that governs how these tiles interact. The author focuses on a special type of interaction where two neighboring tiles can "merge" or "loop" together, much like two people shaking hands and then letting go, leaving a loop behind.

2. The Invisible Threads (Topological Operators)

The paper introduces "topological operators." Imagine you have a piece of string. If you lay this string on your quilt, it doesn't matter exactly where the string is placed, as long as it goes around the same obstacles in the same way.

The Analogy: Think of a hula hoop. If you slide the hoop around a pole, it doesn't matter if you wiggle the hoop left or right; as long as it stays around the pole, the result is the same.
The Paper's Claim: The author shows how to insert these "strings" (operators) into the grid. These strings change the rules of the game locally, but because they are "topological," the overall result depends only on the shape of the path the string takes, not the tiny details of the path.

3. Taking the Quilt Apart (Decomposition)

The core achievement of the paper is a "recipe" for taking the massive space of all possible quilt patterns and breaking it down into smaller, simpler, and indestructible building blocks.

The Analogy: Imagine you have a giant, complex Lego castle. The author provides an algorithm to take it apart and say, "Okay, this section is made of 5 red bricks, this section is 3 blue bricks, and this section is a special 'seed' brick that can grow into a whole tower."
The "Seed States": The paper identifies special "seed" states. These are like the foundation stones. If you apply the rules of the game (the TL algebra) to a seed, it grows into a specific, simple module (a small, self-contained Lego structure). The author proves that the entire system is just a collection of these simple structures glued together.

4. The Examples (The Fibonacci, Ising, and Potts Models)

The author tests this recipe on three famous "quilt" designs:

The Fibonacci Model: A system where tiles follow rules similar to the Fibonacci sequence (1, 1, 2, 3, 5...). It's a bit like a pattern where every new step is the sum of the two before it.
The Ising Model: A classic model for magnetism. Think of it as a grid of tiny magnets that can point up or down.
The Three-State Potts Model: A version of the Ising model where the magnets can point in three different directions instead of two.

For each of these, the author calculates exactly how the "seeds" fit together and how the "invisible strings" (topological operators) act on them. It's like mapping out exactly how a specific twist of the string changes the pattern of the Ising magnets.

5. The Magic Mirror (Modular Invariance)

Finally, the paper looks at what happens when you look at the quilt through a "magic mirror" that twists and turns the entire grid (a modular transformation).

The Analogy: Imagine taking your quilt, folding it into a donut shape, and then stretching it into a different shape. The paper asks: "If I do this, how do the patterns change?"
The Result: The author derives a formula (a transformation matrix) that predicts exactly how the "characters" (the mathematical descriptions of the patterns) change when the grid is twisted. This is crucial because it connects the discrete grid (the quilt) to the continuous world of Conformal Field Theory (the smooth, flowing physics that emerges at large scales).

Summary

In short, this paper provides a universal toolkit for:

Building complex 2D models based on simple fusion rules.
Deconstructing these models into their simplest, irreducible parts (the "seeds").
Predicting how these parts react to "invisible strings" (symmetries).
Calculating how the whole system transforms when the geometry of the world is twisted.

The author does this without needing to guess; they use a step-by-step algorithm based on the fundamental data of the "rulebook" (the fusion category) to solve these problems for a wide variety of models, including some that are critical (like phase transitions) and some that are not.

Problem Statement
The paper addresses the structural analysis of classical two-dimensional lattice statistical models defined by face interactions derived from a fusion category $\mathcal{C}$ . Specifically, it focuses on models where the local interactions satisfy the Temperley–Lieb (TL) algebra relations. The central problem is to systematically determine the decomposition of the transfer-matrix space of states into a direct sum of simple TL (or affine TL) modules for various boundary conditions (fixed, periodic, and twisted periodic). Furthermore, the work seeks to characterize the algebra of topological operators acting on these spaces and to derive the modular transformation properties of the associated irreducible TL characters, particularly at primitive roots of unity.

Methodology
The author employs a framework based on fusion categories and the representation theory of the Temperley–Lieb algebra.

Model Definition: The study considers face models on a square lattice where sites carry simple objects of a fusion category $\mathcal{C}$ . The Boltzmann weights are constructed using quantum dimensions and F-symbols. The local face operators satisfy the TL relations with a loop weight $\beta = d(a)$ , where $a$ is a specific simple object.
Topological Operators: Following the approach of Petkova and Zuber for Conformal Field Theories (CFTs), lattice topological operators are defined as morphisms of TL modules. These operators correspond to inserting modified conditions (seams) along contours. The paper constructs families of operators $Y_p$ (associated with simple objects $p$ ) and $Q_L$ (associated with category automorphisms $L$ ) for both open and closed boundary conditions.
Recursive Decomposition Algorithm: A generic algorithm is presented to decompose the state space $V(N)$ $V (N)$ into simple modules.
- Seed States: The method relies on identifying "seed states"—specific states of fixed size that are annihilated by certain lowering operators ( $c_j$ ) and generate simple modules under the action of TL diagrams.
- Fixed Boundary Conditions: For open boundaries, the decomposition is shown to be a direct sum of simple TL modules $I_k$ . The multiplicities are encoded in matrices $n_k$ derived from Chebyshev polynomials of the adjacency matrix, analogous to fused adjacency matrices in boundary CFT.
- Twisted Periodic Boundary Conditions: For closed boundaries, the decomposition involves simple affine TL modules $I_{k,z}$ . The algorithm utilizes the spectrum of the cyclic translation operator $\Omega$ and the loop operator $f$ to determine the parameters $z$ and the multiplicities.
Modular Analysis: For loop weights of the form $\beta = 2\cos(\pi/h)$ (corresponding to unitary RSOS models), the paper defines modular characters as traces over affine TL simple modules. By analyzing the decomposition of twisted RSOS models and the eigenvalues of topological operators, the author reverses the logic of previous CFT results to derive the modular S-matrix transformation for these TL characters.

Key Contributions and Results

General Decomposition Algorithm: The paper provides a constructive, recursive algorithm to analytically decompose the transfer-matrix space of states for generic 2d lattice models with TL interactions into simple TL modules. This applies to both fixed and twisted periodic boundary conditions.
Topological Operator Algebra: The action of topological operators on the decomposed state spaces is explicitly calculated. In the examples of Fibonacci, Ising, and three-state Potts models, the algebra of TL endomorphisms is shown to be generated by the topological operators associated with simple objects and category automorphisms.
Explicit Examples:
- Fibonacci Model: The decomposition and topological operator actions are detailed for fixed, periodic, and twisted boundary conditions. The scaling limit is identified with the minimal model $\mathcal{M}(5,4)$ , with specific conformal spins arising from the shift operator.
- Ising Model: The face model is mapped to the critical quantum Ising Hamiltonian. The topological operators are identified with spin-flip and Kramers–Wannier duality operations. The scaling limit corresponds to $\mathcal{M}(4,3)$ .
- Three-state Potts Model: Using the Tambara–Yamagami category $TY(\mathbb{Z}_3)$ , the paper analyzes the decomposition and operator actions, identifying the scaling limit with $\mathcal{M}(6,5)$ .
- Non-Critical Model: A model based on $psu(2)_5$ fusion rules (where $d(a) > 2$ ) is analyzed, showing the state space as an infinite direct sum of simple modules.
Modular Invariance of TL Characters: For the $A_n$ series (unitary RSOS models), the paper derives the modular transformation of irreducible TL characters. It establishes that these characters form a representation of the modular group, with the S-matrix given by a product of the fusion category S-matrix and a sign factor. The Verlinde formula is applied to suggest fusion rules for the simple affine TL modules.

Significance and Claims
The paper claims to provide a "systematic way of computing this decomposition from first principles" for any face model with TL interactions associated with a fusion category. This decomposition is described as a "cornerstone for the symmetry analysis" of such models.

Theoretical Unification: The work unifies the treatment of various exactly solved models (Ising, Potts, RSOS) under the language of fusion categories and TL algebras, providing a generic algorithm where previous methods were often model-specific.
Modular Representation Theory: A fundamental result claimed is the derivation of the modular transformation of irreducible TL characters. This extends the understanding of modular invariance beyond standard CFT characters to the algebraic structures of lattice models.
Limitations and Future Directions: The author notes that for non-critical models (where $\beta \ge 2$ ), the algebra of endomorphisms is infinite-dimensional, while the topological operators generate only a finite-dimensional subalgebra; the concrete form of the remaining endomorphisms remains an open question. Additionally, while the Verlinde formula suggests fusion rules for affine TL modules, the paper states that the validity of these rules relies on conditions that require further verification. The work does not propose new experimental applications but rather offers a rigorous algebraic framework for understanding the symmetries and scaling limits of lattice models.