⚛️ high-energy theory

TQFT gravity dual to the orbifold groupoid of Narain code CFTs over $\mathbb{Z}_4$

This paper establishes a topological field theory dual to a family of Narain code CFTs constructed from $A_3$ lattices, demonstrating that their seven inequivalent theories arise as distinct topological boundary conditions of a single abelian Chern-Simons theory and analyzing their interrelations via gauging and Siegel-Weil identities.

Original authors: Rajae Sammani, Abdellah Sebbar

Published 2026-10-23

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Rajae Sammani, Abdellah Sebbar

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

In the vast landscape of theoretical physics, there is a persistent effort to understand how the universe might be built from the ground up. Physicists often look for patterns that connect the microscopic world of quantum particles to the macroscopic world of gravity and space-time. One powerful tool for this exploration is a class of theories called conformal field theories, which describe systems that look the same at every scale. Within this family, there is a specific type known as Narain theories, which model strings moving on a torus, a shape like a doughnut. These theories are usually defined by a continuous set of parameters, making them difficult to catalog completely. To bring order to this chaos, researchers have developed a method using error-correcting codes, the same kind of mathematics used to protect data on your phone or in space communications. By translating the continuous physics into a finite set of discrete codes, they can count and classify these theories with precision, turning an infinite problem into a manageable list.

A team of researchers at the University of Ottawa has taken this approach further, focusing on a specific family of these theories that had remained partially mysterious. They investigated a group of twenty-two distinct theories that arise from a particular mathematical construction involving a lattice called $A_3$ and a specific type of code over the numbers zero to three. While a previous study had successfully listed and categorized these twenty-two theories, it left open the question of how they fit together physically. The new paper provides the missing map, revealing that all twenty-two theories are actually different faces of a single, underlying three-dimensional topological field theory. Think of this bulk theory as a single, solid object, and the twenty-two theories as different ways to wrap a boundary around it. Just as a single piece of clay can be molded into many different shapes, this single bulk theory supports twenty-two distinct boundary conditions, each corresponding to one of the theories in the list.

The researchers discovered that these twenty-two theories are not all equally connected. They fall into three distinct groups, or orbits, based on the internal structure of their codes. One group contains "free" theories, another holds "mixed" theories, and the third contains a single "torsion" theory. The key finding is that you cannot transform a theory from one group into another simply by applying standard symmetries or shifting parameters; you must perform a process called gauging. This is a specific physical operation that changes the theory's fundamental nature. The team mapped out every possible connection between these theories, showing exactly which pairs can be linked by gauging a symmetry group of size two or four. They found that the web of connections is highly intricate: the single torsion theory is just one step away from every other theory, while some pairs of free and mixed theories are separated by a much larger gap, requiring a more complex operation to bridge.

Beyond mapping the connections, the authors explored what happens to the physical spectrum of these theories when they are transformed. They tracked how the energy levels of the particles change, noting that gauging can either raise the energy gap, making the theory more stable, or lower it, introducing new light states. In one striking example, they showed how a theory with a large energy gap could be created from a simpler one by removing specific states, effectively filtering the spectrum. This process also affects the currents, which are the flows of charge within the theory. In some cases, the gauging operation generates entirely new currents that were not present before, appearing only in the twisted sectors created by the transformation. This detailed accounting allows physicists to predict exactly how the properties of a theory will shift as it moves through this network of connections.

The paper also tackles a deep question about how these theories relate to gravity. In the holographic view of physics, a theory on a boundary is dual to a theory of gravity in the bulk. The researchers calculated the sum of the partition functions over all possible shapes of the bulk space, a procedure known as the handlebody sum. For a simpler case with only one pair of dimensions, this sum perfectly matched the average of all possible theories. However, for the case studied here with two pairs, the result was more subtle. The sum over the standard shapes of space-time did not cover all twenty-two theories equally. Instead, it produced a uniform average over only the twelve "free" theories. To recover the average over all twenty-two theories, including the mixed and torsion ones, an additional contribution was required. This extra piece corresponds to a specific shape of space-time with a line defect running through its core, a configuration that had been overlooked in previous, simpler models.

This discovery has a significant implication for how we understand the relationship between discrete codes and continuous gravity. The researchers proved that for systems with three or more pairs of dimensions, the sum over standard space-time shapes cannot be written as a simple, positive combination of the boundary theories. In other words, the standard gravitational path integral fails to reproduce the full ensemble of theories in a straightforward way. This suggests that for more complex systems, the connection between the bulk gravity and the boundary codes is more fragile, requiring non-standard topological configurations to make the math work. The work confirms that while error-correcting codes provide a powerful finite framework for classifying these theories, the bridge to gravity is not always a simple average, and the structure of the universe at this level is richer and more constrained than previously thought.

Technical Summary: TQFT Gravity Dual to the Orbifold Groupoid of Narain Code CFTs over Z4

Problem and Context
This paper investigates the physical interpretation of a specific family of Narain conformal field theories (CFTs) constructed via "Construction A" from error-correcting codes over the ring $\mathbb{Z}_4$ . While previous work established a correspondence between self-dual codes and Narain lattices, and between the average of Narain theories and sums over handlebodies in $U(1)$ gravity, these results were primarily limited to prime or square-free alphabets. The authors focus on the $A_3$ root lattice family over $\mathbb{Z}_4$ , a case where the alphabet is not square-free and the codes are not all "free" (i.e., not all isomorphic to $\mathbb{Z}_4^d$ ). Specifically, for two hyperbolic pairs ( $d=2$ , central charge $c=\bar{c}=6$ ), the classification yields twenty-two distinct even self-dual gluing codes falling into seven inequivalent T-duality classes. The central problem is to understand the bulk topological field theory (TQFT) dual to this family, the structure of the orbifold groupoid connecting these theories, and the nature of the gravitational path integral (sum over handlebodies) in this non-square-free context.

Methodology
The authors employ a combination of algebraic coding theory, lattice constructions, and 3D topological field theory techniques:

Bulk-Boundary Correspondence: They identify the bulk theory as a single abelian Chern-Simons theory, denoted $Z_2$ (equivalent to an untwisted $\mathbb{Z}_4^2$ gauge theory), whose topological boundary conditions are labeled by the even self-dual gluing codes. This utilizes the SymTFT (Symmetry Topological Field Theory) or "sandwich" picture.
Orbifold Web Construction: Using the reduction of codes modulo 2, the authors derive the gauge groups connecting different boundary conditions. They explicitly construct twisted partition functions for gauging finite symmetries ( $\mathbb{Z}_2, \mathbb{Z}_4, \mathbb{Z}_2 \times \mathbb{Z}_2$ ) to map between the seven T-duality classes.
Handlebody Summation: The authors compute the partition function of the bulk Chern-Simons theory summed over genus-one handlebodies. This involves evaluating the Weil representation of $SL(2, \mathbb{Z}_4)$ acting on the vacuum state and projecting onto the space of code vectors.
Finite Siegel-Weil Identity: They prove an identity expressing the handlebody sum as a linear combination of code boundary states, analyzing the coefficients for different numbers of hyperbolic pairs ( $d$ ).

Key Contributions and Results

Bulk Theory and Symmetry Orbits: The twenty-two code theories at $d=2$ are identified as the topological boundary conditions of a single bulk Chern-Simons theory $Z_2$ . The invertible symmetries of the bulk (T-dualities and fractional B-field shifts) organize these boundaries into three distinct orbits based on the group type of the code:
- Free: 12 codes (Classes $F_0, F_1, F_2$ ).
- Mixed: 9 codes (Classes $M_0, M_1, M_2$ ).
- Torsion: 1 code (Class $T$ ).
  The authors demonstrate that theories in different orbits cannot be connected by invertible symmetries alone; they are related only by gauging.
The Orbifold Web: The paper constructs the complete web of gaugings between the seven classes.
- Gauging is determined by the intersection of codes modulo 2.
- Connections exist via groups of order 2, 4, or 8.
- Five of the seven classes are self-dual under specific gaugings (related to T-dualities).
- Notably, the torsion class $T$ is within one gauging step of every other class, while specific pairs (e.g., $F_1$ and $M_0$ ) require gauging of order 8 to connect.
- The authors explicitly calculate the effect of these gaugings on the CFT spectrum, including the gap, the degeneracy of light states, and the emergence of $\widehat{su}(2)$ current algebras. For instance, gauging the torsion class $T$ by $\mathbb{Z}_2$ yields the mixed class $M_2$ , raising the spectral gap from $1/2$ to $1$.
Finite Siegel-Weil Identity and Handlebody Sum:
- The authors prove a finite Siegel-Weil identity (Theorem 5.1) valid for any number of pairs $d$ . This identity expresses the handlebody sum $P_d[e_0]$ in terms of the uniform average of free codes ( $A_d$ ) and the torsion code vector ( $v_{T_d}$ ).
- Case $d=2$ : The coefficient of the torsion code vanishes. The handlebody sum is exactly the uniform average over the twelve free theories. The uniform average over all twenty-two theories is recovered by adding a specific correction term proportional to the torsion code vector: $Z_{all} = \frac{9}{11}Z_{free} + \frac{2}{11}Z_T$ .
- Case $d \ge 3$ : The coefficient of the torsion code becomes negative. Consequently, the handlebody state (the vector in the code space) cannot be expressed as a non-negative linear combination of code boundary states. However, the authors explicitly note that this does not by itself exclude a positive decomposition of the corresponding scalar partition function, as the map from bulk states to partition functions is not injective. Whether the scalar partition function admits a positive decomposition remains an open question.

Significance
The paper claims to extend the understanding of TQFT gravity duals to Narain CFTs beyond the square-free alphabet regime. Its primary significance lies in:

Clarifying Non-Square-Free Cases: It provides the first explicit treatment of the $\mathbb{Z}_4$ case for multiple hyperbolic pairs, revealing that while the handlebody sum alone reproduces the uniform average over the free theories for $d=2$ , recovering the uniform average over all twenty-two theories requires a correction term proportional to the torsion code. For $d \ge 3$ , the handlebody state has negative components, indicating it is not a non-negative combination of code boundary states; however, the existence of a positive ensemble for the scalar partition function remains open, suggesting that contributions from non-handlebody topologies (specifically solid tori with line defects) may be necessary to define a physical ensemble.
Orbifold Groupoid Structure: It maps the complete structure of the orbifold groupoid for the $A_3$ family, showing how free, mixed, and torsion theories are interconnected via gauging, a structure that was previously obscured in descriptions restricted to free codes.
Ensemble Positivity: It highlights that for $d \ge 3$ , the handlebody state fails to be a positive combination of code boundary states, raising questions about the existence of positive ensembles for higher $d$ or the necessity of more complex bulk topologies to achieve a positive decomposition of the scalar partition function.

The work serves as a concrete realization of the SymTFT framework for code CFTs and establishes a finite Siegel-Weil identity that connects bulk topological sums to boundary code averages, while delineating the limitations of this correspondence in non-square-free settings.

More like this