Quantum error-correcting codes from aperiodic monotiles: the Hat and the Spectre

arXiv:2607.15326 · quant-ph, math.MG · Submitted 2026-07-16 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Quantum Radio. Generated commentary on the latest quantum physics and condensed matter papers.

Kai: We're looking at the paper: "Quantum error-correcting codes from aperiodic monotiles".

Mira: Here's the gist of the paper. Quantum error-correcting codes from aperiodic monotiles: the Hat and the Spectre investigates how quantum information can be protected using superpositions of tilings derived from aperiodic monotiles like the Hat and Spectre, extending previous constructions to these new shapes The paper demonstrates that these tilings support hybrid quantum–classical memory structures, where a robust classical bit about the long-range order is stored alongside protected quantum sectors Foundational Framework The Li and Boyle construction provides the canonical framework for this study, defining a candidate code space based on tiling states averaged over isometry orbits The candidate code is defined as C = span(Ψ

T: ⟩), where Ψ

T: ⟩ is the group-averaged state of a tiling T.

Kai: Erasure of a bounded region K is correctable if and only if the Knill–Laflamme conditions hold, which require strong local indistinguishability (H1) and local recoverability (H2).

Mira: Hat Tiling Analysis The Hat tiling exhibits two chirality classes, THat = T+Hat ⊔ T-Hat, distinguished by which handedness forms the majority.

Kai: Theorem 2 establishes strong local indistinguishability within each class, meaning patch frequencies are uniform within T+Hat and within T-Hat.

Title and authors: Kai: So we're looking at this paper, "Quantum error-correcting codes from aperiodic monotiles: the Hat and the Spectre," and it’s about how you can use these specific shapes to protect quantum information against errors.

Mira: Yeah, it's interesting because they are taking these aperiodic tilings—things that aren't periodic like Penrose tilings—and turning them into quantum error-correcting codes. It suggests a new way to build these codes beyond just the classic substitution rules we’ve seen before.

Lev: From an engineering standpoint, I'm curious how robust this is for real hardware. We’re dealing with noisy qubits, and if you have a code based on something complex like this Hat tiling, the overhead has to be manageable.

Kai: The paper sets up the Li and Boyle construction as the foundation, defining the candidate code space based on group-averaged states of these tilings, where erasure correction depends on strong local indistinguishability and local recoverability conditions.

Mira: And what they found is that for the Hat tiling, they prove that strong local indistinguishability holds unconditionally within each chirality class, meaning the patch frequencies are uniform in those classes.

Lev: That sounds promising for error correction, but how does this relate to what we can actually measure on a quantum computer? Does this construction translate directly into a usable logical qubit?

Kai: Well, they found that the Hat code splits into two sectors carrying a superselected classical label based on chirality—which is essentially the global handedness of the long-range order.

Mira: And that’s key because under the physically natural gauge group SE(two), this chirality bit survives, which means it's robust against position and orientation errors <ref:2607.15326#pg1>.

Lev: So, if we use SE(two) gauging, we can protect a classical bit about the global handedness of the tiling <ref:2607.15326#pg1>. But what happens when you look at a different group?

Kai: That’s where the Spectre tiling comes in. The Spectre is strictly chiral, and it has two local indistinguishability classes related by a rotation.

Mira: And when you gauge it with SE(two), those two sectors actually merge, so that sector label doesn't survive the gauging at all <ref:2607.15326#pg1>.

Lev: That’s a big difference then. So the Hat code stores one bit under SE(two), but the Spectre code is different under that same group <ref:2607.15326#pg1>.

Kai: Right, and Lev, what do you think about those scaling parameters they mentioned? The correctable-radius bookkeeping for the Hat tilings scales as r sub n equals n times square root five over three divided by the tile size.

Lev: That tells us the required region size grows linearly with number of tiles, which is good for scaling, but we need to know how much physical area that actually translates to on a chip.

Mira: And for the Spectre code, they show it has a C six statistical rotational symmetry per class and no reflection component when you consider its correction radius scaling as n times four plus the square root fifteen divided by two sigma squared steps.

Title and authors: Kai: So, we have these two different structures here: one where chirality survives SE(two), and one where the sector structure depends on a smaller gauge group G six which relates to a rotation of thirty degrees <ref:2607.15326#pg1>.

Lev: It seems like orientation parity isn't an intrinsic property of the tile itself, because you can swap those sectors by a rotation, not necessarily by gauging with SE(two) <ref:2607.15326#pg1>.

Kai: Exactly. The paper highlights that the Hat is the unique member of this family whose long-range order stores a frame-independent classical bit when gauged under SE(two) <ref:2607.15326#pg1>.

Mira: That’s an important distinction because it suggests a level of protection against local orientation changes that the Spectre doesn't offer in that specific gauge.

Lev: So, to summarize, we have the Hat code giving us a robust classical bit about global handedness when you observe it with SE(two), while the Spectre shows how sector structure depends on choosing your reference frame group G six <ref:2607.15326#pg1>.

Kai: And looking ahead, they pointed out some open problems regarding the singular-pair question for both tilings and trying to relate the rank four first cohomology data to the code structure.

Mira: That means we still don't fully know if there are any tiling configurations that can be tiled in two distinct ways outside a bounded region for either shape.

Lev: And I think until we nail down those singular-pair questions, it’s hard to predict the actual error threshold for running this on a real quantum processor.

Kai: It’s exciting because it shows how we can extend previous constructions, like the one by Li and Boyle with Penrose tilings, to these new shapes like the Hat and Spectre.

Mira: The core idea is that these tilings support hybrid quantum classical memory structures where a robust classical bit about the long-range order is stored alongside protected quantum sectors.

Lev: That hybrid nature is what makes this interesting for practical applications, because it’s not just a pure quantum error correction scheme.

Kai: So, to wrap up on the Hat and Spectre paper, we see that while both monotiles support superselected bits under G six, only the Hat's class label survives when gauged by SE(two) because its class-swapping isometry is improper <ref:2607.15326#pg1>.

Mira: That means for orientation protection specifically under position and orientation unobservability, the Hat code stands out as unique among these two examples.

Lev: I just think we need more concrete results on those scaling parameters before we can really say this translates into a practical error correction scheme that runs well on current superconducting hardware.

Kai: Right, so the next step is figuring out how to construct explicit finite toroidal approximants for these codes and seeing what happens to that chirality bit at finite size.

The paper's summary: Kai: So we're looking at how these specific shapes, like the Hat and the Spectre—these are called aperiodic monotiles—can be used to build quantum error-correcting codes.

Mira: Basically, they’re using these tilings to protect quantum information by creating superpositions of them. It’s extending older ideas about codes from simpler geometric patterns.

Lev: The core idea here is that you can store a robust classical bit about the long-range order right alongside protected quantum sectors in the same system. That sounds like a hybrid memory structure we could actually build in hardware, right?

Kai: Exactly. The paper shows that for the Hat tiling, you get two erasure-correcting sectors based on chirality—that’s whether it's handed left or right.

Mira: And under a specific mathematical framework called SE two gauging, that chirality bit survives. It means you can store a classical label about the global handedness of the pattern even if you mess with its orientation slightly.

Lev: That robustness is interesting because it means the correction doesn't have to be perfectly aligned in space, which makes it more compatible with physical constraints.

Kai: Then there’s the Spectre tiling, which is strictly chiral, and its two sectors merge when you try to gauge it under SE two. That sector label gets gauged away by a mere thirty degree rotation instead of disappearing entirely.

Mira: So the distinction is that for the Hat, you get a frame-independent bit under SE two, but for the Spectre, that sector structure depends on a smaller group G six related to those rotational symmetries.

Lev: It’s like having two different ways to protect information based on how you choose to look at the geometry of the pattern.

Kai: The paper points out that only the Hat is truly unique in this family because its class-swapping isometry isn't improper under SE two gauging, which is what makes it frame independent.

Mira: So, while both tilings offer error correction, the Hat seems to give us a more stable classical reference point when we look at position and orientation errors.

Lev: If we think about running this on real hardware, the scaling parameters are important. The Hat code needs a correction radius that grows with the number of tiles as n times square root five over three divided by tile size.

Kai: That means for every new tile you add, your error correction region has to grow in a specific way. It’s not just linear scaling, it’s something more complex.

Mira: And for the Spectre code, the correction radius scales with n times four plus the square root fifteen divided by two sigma squared steps. That’s a different kind of scaling behavior entirely.

Lev: The authors also flag some open problems about singular pairs—whether these tilings can be tiled in two distinct ways outside a bounded region—and they want to connect that to cohomology data for finite torus approximations.

Kai: And the main takeaway is that these shapes allow us to extend previous work, like the Li and Boyle construction, into new geometric territory with these aperiodic monotiles.

Mira: It shows that hybrid quantum classical memory structures are possible using these complex tiling systems, where the classical bit about long-range order stays safe while the quantum part gets protected.

Lev: The next thing we need to figure out is how to actually construct those explicit finite torus approximants and see what happens to that chirality bit when you stop thinking about an infinite system.

The paper's improvements: Tom: So the paper doesn't just stop there; they suggest some ways to make these codes better than what they currently have.

Kai: Right, they’re looking at how to refine the construction itself, especially for those scaling parameters we talked about earlier.

Mira: They’re probing the singular pair question again, asking if either shape can be tiled in two different ways outside a bounded region. That's a deep structural question for any tiling code.

Lev: Because if you can find those two distinct tilings, it affects how robust the error correction is when you try to put these codes on real hardware and have errors happen in unexpected places.

Kai: And they’re also trying to link the rank four first cohomology data to the actual structure of these codes. That’s a way for them to get more concrete structural information from the tiling geometry itself.

Mira: From a theorist's view, that connection would give us more leverage over whether these tilings are truly capturing all the necessary constraints for error correction.

Lev: For me, it means we need to know if we can construct explicit finite torus approximants for these codes. That’s the only way I can actually predict how much overhead those codes will impose on our current superconducting qubit setups.

Kai: If they manage to figure out the fate of that chirality bit at finite size, that would be a huge piece of information for experimentalists like me, because we need to know if we can actually measure that classical bit reliably.

Mira: It suggests the next step in this research is moving from pure construction proofs to understanding the structural limits and finite system behavior.

Lev: So basically, they’re trying to figure out the hard limits of what these geometric shapes can do as error correction tools, beyond just showing that they *can* be done.

Conclusion: Kai: So to wrap up, this paper on "Quantum error-correcting codes from aperiodic monotiles: the Hat and the Spectre" shows that we can build hybrid memory structures using these new shapes.

Mira: It confirms that by averaging over tiling states, we can store a robust classical bit about the long-range order alongside our protected quantum sectors.

Lev: The main implication is that it opens up a new family of codes based on non-periodic geometries, which could be useful if we ever want to move beyond the standard Penrose tilings.

Kai: I think what stands out is the distinction between the Hat and Spectre under different gauges, especially how the SE two gauging affects the classical bit in each case.

Mira: Exactly, because it tells us that orientation parity isn't an intrinsic property of a tile; for instance, only the Hat stores that frame-independent bit when you look at position and orientation unobservability with SE two.

Lev: From an engineering standpoint, it means we have to be very careful about which gauge group we use to protect our system because the protection changes depending on what physical constraints we impose.

Kai: And the scaling numbers for both codes are pretty specific, showing how the required correction radius grows with n in ways that are different for each shape.

Mira: It’s a lot of data to parse, but it grounds this abstract math in measurable parameters like those correction radii and the statistical rotational symmetry.

Lev: I just think we need to get more concrete results on those singular pair questions before we can really say this translates into a practical error correction scheme that runs well on current superconducting hardware.

Kai: Right, so the next thing we need to see is how those finite size effects manifest when they try to construct explicit finite torus approximants for these codes.

Josep Batle, Adam Bednorz

CRISP – Centre de Recerca Independent de sa Pobla · Departament de Física, Universitat de les Illes Balears · Faculty of Physics, University of Warsaw

quant-ph, math.MG

Submitted: 2026-07-16

Updated: 2026-10-30

Comments: Extended updated version

Code: https://github.com/isohedral/hatviz

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 92/100

The gist: Quantum error-correcting codes from aperiodic monotiles: the Hat and the Spectre investigates how quantum information can be protected using superpositions of tilings derived from aperiodic monotiles…

Key concepts

Li and Boyle Construction
This is the main framework used to define a candidate code space based on tiling states averaged over isometry orbits. It sets up the mathematical foundation for determining if a given tiling can serve as an error-correcting code, requiring strong local indistinguishability and recoverability conditions.
Chirality Classes
The Hat tiling has two classes based on which handedness forms the majority. The Spectre also has two local-indistinguishability classes related by a 30-degree rotation. These classes define separate sectors in the code space, each potentially carrying different information.
Gauge Group Dependence
The way information is protected depends on the symmetry group (gauge group) used to analyze the tiling. The Hat's classical bit survives under SE(2) gauging because its class-swapping isometry is improper, whereas the Spectre's analogous bit is gauged away by a proper rotation (R30).
Hybrid Quantum–Classical Memory
The resulting code structure allows for storing both quantum information and a classical bit. This classical bit represents the global handedness of the tiling's long-range order, providing a robust piece of information that is protected alongside the quantum sectors.

Terminology

Summary

Quantum error-correcting codes from aperiodic monotiles: the Hat and the Spectre investigates how quantum information can be protected using superpositions of tilings derived from aperiodic monotiles like the Hat and Spectre, extending previous constructions to these new shapes The paper demonstrates that these tilings support hybrid quantum–classical memory structures, where a robust classical bit about the long-range order is stored alongside protected quantum sectors

Foundational Framework

The Li and Boyle construction provides the canonical framework for this study, defining a candidate code space based on tiling states averaged over isometry orbits The candidate code is defined as C = span(Ψ[T]⟩), where Ψ[T]⟩ is the group-averaged state of a tiling T. Erasure of a bounded region K is correctable if and only if the Knill–Laflamme conditions hold, which require strong local indistinguishability (H1) and local recoverability (H2). p. 1

Where the paper says it

The candidate code is C = span{|Ψ[T]⟩}. Erasure of a bounded region K ⊂ R 2 is correctable if and only if the Knill– Laflamme conditions hold: for every operator OK supported in K, ⟨Ψ[T1] |OK|Ψ[T2]⟩ = c(OK) δ[T1],[T2] .

Hat Tiling Analysis

The Hat tiling exhibits two chirality classes, THat = T+Hat ⊔ T-Hat, distinguished by which handedness forms the majority. Theorem 2 establishes strong local indistinguishability within each class, meaning patch frequencies are uniform within T+Hat and within T-Hat. p. 3

Where the paper says it

Its correct scope, however, is per chirality class: patch frequencies are uniform within T + Hat and within T − Hat, and the two classes are separated by them. Theorem 2 (Strong local indistinguishability, Hat, within each class).

The code space splits into two erasure-correcting sectors carrying a superselected classical label based on chirality. p. 1
Where the paper says it

For the Spectre they are the two rotational subclasses of Baake, Gähler, Mazáč and Sadun, exchanged by a 30◦ rotation [5]. Each class separately supports the Li–Boyle construction, while a coarse local observable distinguishes the classes: the code space splits into two erasure-correcting sectors carrying a superselected classical label.

Under the physically natural gauge group SE(2), the Hat code stores one robust classical bit, which is the global handedness of its long-range order. p. 0
Where the paper says it

Whether the label survives depends on which isometries are gauged: the Spectre’s classes are exchanged by a 30◦ rotation and merge once all proper isometries are gauged, whereas the Hat’s are exchanged only by reflections. Under the physically natural gauge group SE(2), the Hat code therefore stores one robust classical bit — the handedness of its long-range order, readable in any bounded window with separation ∆ = |K| √ 5/3 — alongside its protected quantum sectors.

Spectre Tiling Analysis

The Spectre, a strictly chiral monotile, possesses two local-indistinguishability classes (LI1 and LI2) related by a 30◦ rotation. The isometry exchanging these classes is a proper rotation R30, which is gauged by SE(2). p. 8

Where the paper says it

1; the decisive difference, developed in §4. 4, is that the isometry exchanging the Spectre’s classes is a proper rotation, so the superselected label that survives SE(2)-gauging for the Hat does not survive it for the Spectre.

Consequently, the Spectre’s two-sector structure merges under SE(2)-gauging, meaning its sector label does not survive gauging. p. 1
Where the paper says it

Whether the label survives depends on which isometries are gauged, and here the two monotiles part ways. The Spectre’s classes are exchanged by a proper isometry, so gauging SE(2) — position and orientation unobservable, parity not gauged — merges its sectors; the Hat’s classes are exchanged only by reflections, so its label survives.

The Spectre code is a hybrid memory of exactly the type of Theorem 9 when gauged by G6 = R2 ⋊ C6. p. 8
Where the paper says it

In a toric code the ground sectors are locally indistinguishable, which is why relative phases between them are protected; here the sectors are locally distinguishable, which is why the label is robust and classical. CHat is a hybrid quantum–classical memory: one superselected bit — the global handedness of the tiling’s long-range order — alongside two erasure-correcting quantum sectors.

Code Parameters and Comparison

The Hat code is characterized by a chirality bit that survives SE(2)-gauging, while the Spectre’s analogous bit is gauged away by a mere rotation. The correctable-radius bookkeeping for the Hat tilings scales as rn = Ktiles √5/3. p. 8

Where the paper says it

The correctable-radius bookkeeping follows the deflation ladder: an erasure of radius r (in fundamental hat units a) closes after ⌈logφ2 (r/a)⌉ deflation steps; equivalently, the correctable radius after n steps scales as rn = a φ2n . (10) The Hat tilings have C6 statistical rotational symmetry, enhanced to D6 when reflections are counted, and first Čech cohomology Hˇ 1 (ΩHat, Z) = Z 4 [4]; these data enter the discussion of Section 5.

The Spectre code exhibits C6 statistical rotational symmetry per class with no reflection component, and its correction radius scales as rn = a(4 + √15)n per σ2 step. p. 8
Where the paper says it

The correctable-radius bookkeeping follows the deflation ladder: an erasure of radius r (in fundamental hat units a) closes after ⌈logφ2 (r/a)⌉ deflation steps; equivalently, the correctable radius after n steps scales as rn = a φ2n . (10) The Hat tilings have C6 statistical rotational symmetry, enhanced to D6 when reflections are counted, and first Čech cohomology Hˇ 1 (ΩHat, Z) = Z 4 [4]; these data enter the discussion of Section 5.

Open Problems

The primary open problem is to settle the singular-pair question (Lemma 7) for both the Hat and the Spectre, asking whether either admits two distinct tilings agreeing outside a bounded region. Additionally, researchers aim to relate the rank-4 first cohomology data to the code structure and construct explicit finite toroidal approximants for these codes. p. 8

Where the paper says it

(10) The Hat tilings have C6 statistical rotational symmetry, enhanced to D6 when reflections are counted, and first Čech cohomology Hˇ 1 (ΩHat, Z) = Z 4 [4]; these data enter the discussion of Section 5. 4 The Spectre code The Spectre is a strictly chiral aperiodic monotile discovered by Smith, Myers, Kaplan and GoodmanStrauss shortly after the Hat [3].

The fate of the Hat chirality bit at finite size remains an open question. p. 1
Where the paper says it

Section 2 recalls the Li–Boyle framework and fixes the vocabulary of both fields. Section 3 develops the Hat: we first establish the structural inputs the framework needs, then prove recoverability in its correct scope and isolate the single question that remains open, and finally construct the code and explain why the Hat unexpectedly stores an additional classical bit.

Conclusion on Gauge Dependence

The analysis concludes that under G6, both monotiles carry a superselected bit, but under SE(2), only the Hat’s class label survives because its class-swapping isometry is improper. This makes the Hat the unique member of the family whose long-range order stores a frame-independent classical bit. p. 0

Where the paper says it

Whether the label survives depends on which isometries are gauged: the Spectre’s classes are exchanged by a 30◦ rotation and merge once all proper isometries are gauged, whereas the Hat’s are exchanged only by reflections. Under the physically natural gauge group SE(2), the Hat code therefore stores one robust classical bit — the handedness of its long-range order, readable in any bounded window with separation ∆ = |K| √ 5/3 — alongside its protected quantum sectors.

The Spectre’s sectors exist relative to a smaller gauge group G6, confirming that orientation parity is not an intrinsic property of a tile. p. 1
Where the paper says it

The Spectre’s classes are exchanged by a proper isometry, so gauging SE(2) — position and orientation unobservable, parity not gauged — merges its sectors; the Hat’s classes are exchanged only by reflections, so its label survives. Under the physically natural gauge group, an infinite Hat tiling superselects one robust classical bit: the global handedness of its long-range order, readable in any bounded window through the reflected-tile fraction, with separation ∆ = |K| √ 5/3.

The paper identifies the Hat as the unique member of the family whose long-range order stores a frame-independent classical bit. p. 0
Where the paper says it

Whether the label survives depends on which isometries are gauged: the Spectre’s classes are exchanged by a 30◦ rotation and merge once all proper isometries are gauged, whereas the Hat’s are exchanged only by reflections. Under the physically natural gauge group SE(2), the Hat code therefore stores one robust classical bit — the handedness of its long-range order, readable in any bounded window with separation ∆ = |K| √ 5/3 — alongside its protected quantum sectors.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Strong local indistinguishability in Quantum Error Correction: The system can store logical information in degrees of freedom that no local observer can read by utilizing tilings whose patch frequencies are uniform within each chirality class, allowing for a hybrid quantum–classical memory.

  2. Robust Classical Bit Storage: The AI system can store a robust classical bit representing the global handedness of its long-range order, which is readable in any bounded window with a separation of approximately 0.745 tile diameters.

  3. Sector-Specific Error Correction: The system can operate two erasure-correcting quantum sectors, one for each chirality class, where the sector label serves as a superselected classical bit that is preserved under erasure of any bounded region K and readable via the reflected-tile counter Nrefl K.

  4. Gauge Group Dependent Information Protection: The system's ability to store this classical bit depends on the gauge group used; it can be protected under SE(2) (position and orientation unobservable) but is gauged away by a proper rotation (R30) when using the full E(2).

Abstract

Li and Boyle showed that the Penrose tiling defines a quantum error-correcting code: superpositions of tilings over isometry orbits protect quantum information against erasure of any bounded region. We extend the construction to the aperiodic monotiles discovered by Smith, Myers, Kaplan and Goodman-Strauss. For the Hat, we prove strong local indistinguishability for all Hat tilings, and we prove local recoverability unconditionally for all nonsingular Hat tilings via the torus parametrization of the underlying cut-and-project scheme. The remaining singular case reduces to one sharply posed geometric question -- can a region that is a union of hats be retiled a second way? -- which we verify computationally has no counterexample up to a substantial scale: a certified 2490-tile patch admits precisely one tiling by hats, so all 2 2490 of its tile-subregions retile uniquely. Unlike the Penrose, Ammann-Beenker and Fibonacci tilings, both monotiles form two local-indistinguishability classes, so their code spaces split into two erasure-correcting sectors carrying a superselected classical label. Whether the label survives depends on which isometries are gauged: the Spectre's classes are exchanged by a 30 rotation and merge once all proper isometries are gauged, whereas the Hat's are exchanged only by reflections. Under the physically natural gauge group SE(2), the Hat code therefore stores one robust classical bit -- the handedness of its long-range order, readable in any bounded window with separation Δ= K sqrt 5 /3 -- alongside its protected quantum sectors. It is the reflexible monotile, not the chiral one, that carries the chirality bit. We give the exact Perron-Frobenius data of both codes, including the per-class reflected-Hat frequencies (3 minus or plus sqrt 5)/6 and the Spectre orientation-class frequencies (5 plus or minus sqrt 15)/10.

Sources

Related papers