⚛️ high-energy theory

Fixed-point tensor network for compactified boson conformal field theory

This paper constructs fixed-point tensor networks for the 2D compactified boson conformal field theory using boundary data, demonstrating their ability to accurately reproduce the closed-string spectrum, generate stable renormalization-group flows, and enable continuous movement along the $c=1$ moduli space via exactly marginal deformations.

Original authors: Gong Cheng, Dong-Yu Bao, Zheng-Cheng Gu

Published 2026-07-07

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Gong Cheng, Dong-Yu Bao, Zheng-Cheng Gu

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 the universe as a giant, complex tapestry. Physicists have long known that at its most fundamental level, this tapestry is woven from patterns of energy and symmetry described by something called "Conformal Field Theory" (CFT). Think of CFTs as the master blueprints for how these patterns behave when you zoom in or out, or when you stretch and twist the fabric of space.

For a long time, scientists could only perfectly map out the blueprints for very simple, "rational" patterns—like a grid with perfect, repeating squares. But the real universe often involves "irrational" patterns: complex, non-repeating designs that are much harder to pin down. One of the most famous of these difficult patterns is the "compactified boson," a theory that describes how particles behave when they are trapped on a tiny, circular track (like a bead on a wire loop).

The Big Breakthrough
This paper introduces a new way to map these tricky, irrational patterns using a tool called a Fixed-Point (FP) Tensor Network.

To understand this, imagine you are trying to describe a massive, intricate 3D sculpture. Instead of trying to describe the whole thing at once, you break it down into small, standard Lego blocks. In physics, these "blocks" are called tensors. Usually, when you try to build a model of these complex circular-track theories, the Lego blocks get messy and infinite in number, making the model impossible to build on a computer.

The authors of this paper found a clever trick. They realized that even though the theory is complex, you can still build it using a finite, manageable set of Lego blocks if you look at the problem from the "edge" of the system (using what they call "open-string" data).

How They Did It: The "Edge" Strategy
Think of the circular track theory as a drum. Usually, it's hard to understand the sound of the whole drum by looking at the skin. But if you look at the rim (the boundary), you can figure out the whole song.

The Lego Blocks: They created special tensors (the Lego blocks) based on the rules of how the "edge" of the system behaves.
The Discretization: Since computers can't handle infinite numbers, they replaced the smooth, continuous edge with a grid of points (like pixels on a screen). They proved that if you have enough pixels (a high enough "cutoff"), the computer model becomes incredibly accurate, almost indistinguishable from the real, smooth theory.
The Result: When they put these blocks together, the resulting model perfectly reproduced the "closed-string" spectrum. In plain English, this means the model correctly predicted the energy levels and vibrations of the entire system, just as if they had solved the impossible math equations directly.

The "Magic Knob"
One of the most exciting parts of their discovery is a "controllable exactly marginal deformation."

Imagine you have a radio dial that tunes between different stations. In this theory, there is a "moduli space" (a landscape of possibilities) where you can smoothly change the size of the circular track without breaking the physics. The authors found a way to encode this "dial" directly into a single Lego block.

By slightly adjusting the geometry of their tensor (the shape of the block), they could turn a "knob" (a parameter they call $\tau$ ) that smoothly shifts the system from one size of the circle to another. This allows them to simulate a continuous flow of physics, moving from one state to another without the simulation crashing or becoming unstable.

Why This Matters
The authors didn't just build a model for one specific case; they built a framework that works for a whole family of these complex theories.

Stability: They tested their model using a sophisticated computer algorithm (Tensor Complex Renormalization). The model didn't fall apart; it stayed stable and accurately predicted the physics, even when they simplified the blocks to their most basic forms.
A New Route: This provides a concrete, step-by-step "lattice" (grid-based) route to understanding a broad class of irrational theories that were previously too messy to handle with standard computer methods.

In Summary
The authors took a notoriously difficult, "irrational" theory of physics (particles on a tiny circle) and built a stable, computer-friendly Lego model for it. They showed that by looking at the edges and using a clever grid system, you can perfectly reconstruct the complex interior. Furthermore, they found a way to build a "dial" into the model that lets you smoothly tune the physics, offering a powerful new tool to explore the deep structure of the universe's fundamental laws.

Technical Summary: Fixed-point tensor network for compactified boson conformal field theory

Problem Statement
Fixed-point (FP) tensor networks offer a discrete spacetime representation of conformal field theories (CFTs), providing a framework to understand holographic duality, generalized symmetries, and quantum gravity. While recent advances have successfully constructed exact FP tensors for rational CFTs (RCFTs) using boundary changing operators (BCOs), extending this framework to irrational CFTs remains a significant challenge. The 2D compactified boson theory is a prototypical irrational CFT, characterized by a continuous family of theories parametrized by the compactification radius $R$ , which is generically irrational. In such theories, the fusion of representations generates infinitely many states, and boundary states are expressed as infinite sums or integrals rather than finite discrete combinations. This complexity renders the algebraic structure less tractable and prevents a straightforward discrete formulation. The central problem addressed in this work is whether a fully discrete spacetime-lattice description can be constructed for this irrational CFT and whether it can reproduce the exact closed-string spectrum and stable renormalization group (RG) flows.

Methodology
The authors construct FP tensors for the 2D compactified boson CFT ( $c=1$ ) at a generic radius $R$ by adapting the RCFT construction to boundary data involving open-string vertex operators.

Tensor Construction from Boundary Data:
- The FP tensor is built from correlation functions of BCOs that change conformal boundary conditions (CBCs).
- The authors utilize two families of CBCs: Dirichlet (labeled by D-brane positions $x_a$ ) and Neumann (labeled by Wilson lines $\theta_a$ ).
- For a rank-3 tensor (isosceles right triangle), the legs carry open-string vertex operators with $U(1)$ charges determined by the boundary conditions. The tensor elements are derived from the boundary three-point functions of these operators, mapped to the triangle geometry via conformal transformations.
- To make the tensor suitable for numerical calculation, the continuous families of boundary conditions are discretized into finite grids ( $x_m$ and $\theta_l$ ) with sizes $\Lambda_D$ and $\Lambda_N$ .
Handling Irrationality and Marginal Deformations:
- A key difficulty is that a finite discretization of a single boundary family (e.g., only Dirichlet) introduces an effective marginal deformation due to the presence of level- $(1,1)$ descendant states ( $J_{-1}\bar{J}_{-1}|0\rangle$ ) in the superposition of Cardy states.
- To address this, the authors propose a Refined FP Tensor using a Mixed Shrinkable Boundary Condition (MSBC). This involves a weighted linear combination of discretized Dirichlet and Neumann Cardy states. In this symmetric combination, the level- $(1,1)$ contributions from the two sectors enter with opposite signs and cancel each other, eliminating the unwanted marginal perturbation.
- Additionally, the authors construct a Rank-4 FP Tensor from a boundary four-point function. They find that the rank-4 construction naturally suppresses the residual marginal perturbation, often yielding accurate results even without the MSBC, provided only primary components are retained.
Marginal Flow Control:
- The authors demonstrate that the geometry of the rank-4 tensor can be parametrized by a variable $\tau$ (related to the corner size). Modifying this geometric factor allows for the active generation of an exactly marginal flow, enabling continuous movement along the $c=1$ moduli space (changing the radius $R$ ).
Numerical Verification:
- The validity of the constructed tensors is tested by extracting the closed-string spectrum from the transfer matrix.
- The Tensor Complex Renormalization (TCR) algorithm is employed to simulate RG flows. The extracted central charges and conformal dimensions are compared against exact analytical results for specific radii (e.g., $R=1$ , $R=1/\sqrt{2}$ , $R=\sqrt{2}$ ).

Key Contributions and Results

Discrete Formulation of Irrational CFT: The paper successfully extends the FP tensor construction beyond RCFTs to the compactified boson, a generic irrational CFT. It demonstrates that a fully discrete lattice description is possible despite the continuous nature of the theory's parameters.
High-Accuracy Spectrum Reproduction: Numerical results show that the constructed tensors reproduce the closed-string spectrum with high accuracy. As the number of descendant states retained in the tensor increases, the numerical spectrum systematically converges to the exact theoretical spectrum.
Stable RG Flows: The tensors exhibit stable RG flows under the TCR algorithm. The refined rank-3 tensor (using MSBC) and the rank-4 tensor both successfully recover the expected conformal spectrum and central charge ( $c \approx 1$ ) with correct degeneracy structures.
Controllable Marginal Deformation: The work identifies a mechanism to control exactly marginal deformations at the single-tensor level. By tuning the geometric parameter $\tau$ in the rank-4 tensor, the authors generate flows that move continuously along the $c=1$ moduli space, accurately reproducing the spectra of compactified boson CFTs at different radii.
Cutoff Conditions: The authors derive specific cutoff conditions for the grid sizes ( $\Lambda_D > 2\sqrt{2}R$ and $\Lambda_N > \sqrt{2}/R$ ) that ensure the discretized boundary state superposition selects the vacuum Ishibashi state while suppressing irrelevant higher-charge primaries.

Significance and Claims
The authors claim that this framework establishes a concrete, lattice-level route toward describing a broad class of 2D irrational CFTs. By constructing exact FP tensors from generalized symmetry and boundary data, the work provides a non-perturbative method to explore CFT moduli spaces. The ability to encode marginal deformations directly at the tensor level offers a new tool for studying flows between different compactified boson theories. The paper positions this work as a step toward understanding the algebraic reformulation of continuous quantum field theories in discrete spacetime, potentially aiding in the exploration of holographic duality and generalized symmetries in irrational settings. The authors note that while the construction is currently limited to the compactified boson, the methodology suggests a pathway to more complex irrational theories. Future directions mentioned include constructing ground-state wave functions directly from the FP tensor and exploring broader moduli spaces, including circle and orbifold branches.

Technical Summary: Fixed-point tensor network for compactified boson conformal field theory

More like this