Onsiteability of Higher-Form Symmetries

✨

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

The Big Picture: Can We "Localize" a Symmetry?

Imagine you have a giant, complex machine (a quantum system) with many moving parts. In physics, we often look for symmetries—rules that say, "If I do this specific change to the machine, it looks exactly the same."

Usually, we want these changes to be onsite. This means the rule is simple: "Change this specific gear, and that specific gear stays alone." You don't need to reach across the whole machine to fix it; you just tweak one local part.

However, some symmetries are "higher-form." Instead of acting on a single gear (a point), they act on a whole string of gears or a sheet of metal (lines or surfaces). The big question this paper asks is: Can we take these complex, "spread-out" symmetry rules and simplify them into simple, local "onsite" rules?

The authors say: Yes, but only if the machine isn't "glitched" in a specific way.

The Old Rule vs. The New Discovery

The Old Rule (For Simple Symmetries):
For a long time, physicists believed in a simple "Golden Rule":

If a symmetry has a "glitch" (called an anomaly), it cannot be made local (onsite).
If it has no glitch, it can be made local.
Analogy: Think of a glitch as a tangled knot in a rope. If the rope is knotted, you can't straighten it out just by pulling on the ends (local moves). You have to untie the knot first.

The New Discovery (For Higher-Form Symmetries):
The authors found that for "higher-form" symmetries (those acting on lines or surfaces), this Golden Rule is broken.

A symmetry can have a glitch (an anomaly) and still be made local.
Analogy: Imagine a rope that looks knotted from the outside (anomalous), but if you look closely at the weave, you realize the knot is actually just a pattern that can be untangled by adding a little extra string (ancillas) and rearranging the weave (a circuit).

So, the paper asks: What is the real rule for when we can untangle these knots?

The Real Rule: The "Transgression" Test

The authors propose a new test called Transgression. Think of this as a "stress test" for the symmetry.

The Setup: You have a symmetry acting on a 3D space (like a block of ice).
The Test: Imagine slicing a thin sheet out of that ice. Now, look at the symmetry acting only on that 2D sheet.
The Result:
- If the symmetry on the sheet is perfectly clean (no glitches), then the original 3D symmetry can be made local (onsite).
- If the symmetry on the sheet is still glitched, then the original 3D symmetry cannot be made local.

The Metaphor:
Imagine you are trying to organize a messy library (the 3D system).

The "Old Rule" said: "If the library is messy, you can't organize it."
The "New Rule" says: "Even if the whole library is messy, you might still be able to organize it, unless the mess gets worse when you look at just the fiction section (the 2D sheet)."
If the fiction section is still a disaster, you can't organize the whole library. If the fiction section is tidy, you can organize the whole thing.

The "Semion" Example: A Failed Test

The paper uses a specific example called the Semion to show this.

The Semion is a type of particle in a 2D world that has a "twist" in its behavior (a topological spin of 1/4).
When the authors apply their "Transgression Test" (looking at the 1D line inside the 2D world), they find a glitch.
Conclusion: Because the test failed, the Semion's symmetry cannot be made local. It is "un-onsiteable." You cannot simplify its rules to act on individual points, no matter how much you rearrange the system.

The "Fermion" Example: A Passed Test

In contrast, they look at a Fermion (a type of particle like an electron).

It also has a glitch in the 2D world.
However, when they apply the "Transgression Test" to the 1D line, the glitch disappears! The line is clean.
Conclusion: Even though the 2D world is glitched, the 1D line is fine. Therefore, the Fermion's symmetry can be made local.

The "Pauli" Payoff

The paper goes one step further. They prove that if a symmetry can be made local, it can be transformed into something very simple and familiar: Pauli Operators.

Analogy: Think of a complex, custom-built robot arm. The authors show that if the robot is "fixable," you can actually replace its complex joints with simple, standard Lego blocks (Pauli operators).
This is huge for quantum computing. It means that if a symmetry passes their test, we can build it using standard, reliable quantum computer parts (like the ones used in error-correcting codes).

Summary of the Paper's Claims

The Problem: We want to know if complex, "spread-out" symmetry rules can be simplified into simple, local rules.
The Breakthrough: The old rule (No Glitch = Local) is wrong for these complex symmetries. A system can be glitched and still be local.
The Solution: The authors introduce a new test called Transgression.
- If the symmetry looks clean when you slice it down to a lower dimension, it is onsiteable (can be simplified).
- If the slice is still glitched, it is not onsiteable.
The Result: If a symmetry passes this test, it can be built using simple, standard quantum building blocks (Pauli operators).
The Limit: They do not claim this applies to medical treatments or future technologies outside of quantum physics. They strictly define the mathematical conditions for when these symmetries can be simplified in lattice models.

In short: You can't always tell if a system is "fixable" by looking at the whole mess. You have to slice it open and check the layers. If the inner layers are clean, the whole thing can be organized.

1. Problem Statement

In quantum many-body systems and lattice models, a symmetry is termed onsite if it acts independently on local degrees of freedom (site-by-site). A central question in the field is determining when a global symmetry, which may act non-locally (e.g., on extended objects like loops or surfaces), can be transformed into an onsite form.

Standard Lore: For ordinary 0-form symmetries in (1+1) dimensions, a symmetry is onsiteable if and only if it is anomaly-free (specifically, its 't Hooft anomaly vanishes). This equivalence links onsiteability to the possibility of gauging the symmetry.
The Gap: For higher-form symmetries (which act on extended objects), the relationship between onsiteability and anomalies is subtler. A higher-form symmetry can possess a non-trivial 't Hooft anomaly (obstructing gauging in the continuum QFT sense) yet still admit an onsite realization on a lattice.
Core Question: What is the correct, general criterion for the onsiteability of higher-form symmetries on lattices? The authors seek to clarify the conditions under which a higher-form symmetry can be "disentangled" into an onsite action using ancillas and finite-depth quantum circuits (FDQCs).

2. Methodology

The authors employ a combination of algebraic topology, lattice gauge theory, and quantum information theory:

Definitions:
- Onsiteability: A symmetry is onsiteable if it can be transformed into a strictly local (onsite) form via tensoring with finite-dimensional ancillary Hilbert spaces and conjugation by a finite-depth quantum circuit.
- Transgression ( $\Phi$ ): A cohomological map $\Phi: H^{d+2}(B^{p+1}G, U(1)) \to H^{d+1}(B^p G, U(1))$ . This map relates the anomaly of a $p$ -form symmetry in $(d+1)$ dimensions to the anomaly of a $(p-1)$ -form symmetry restricted to a codimension-1 submanifold.
- Higher Gauging: The process of gauging a symmetry within a submanifold. The authors propose that onsiteability is equivalent to the possibility of "1-gauging" (gauging within codimension-1).
Analytical Tools:
- Group Cohomology: Used to classify anomalies ( $H^*(BG, U(1))$ ).
- Lattice Anomaly Indices: Introduction of indices valued in groups of Quantum Cellular Automata (QCA), specifically $H^{d+2-q}(BG, \text{QCA}_{q-1})$ , which capture obstructions unique to lattice models not present in continuum QFT.
- Else-Nayak Reduction: A technique to analyze the associativity of symmetry operators restricted to boundaries or submanifolds to extract anomaly indices.
- Pauli Stabilizer Models: Utilizing the structure of toric codes and stabilizer formalism to demonstrate explicit onsite realizations.

3. Key Contributions and Results

A. Equivalence between Onsiteability and Higher Gauging

The paper establishes a fundamental equivalence: A higher-form symmetry is onsiteable if and only if it admits "higher gauging" (specifically 1-gauging) on the lattice.

This generalizes the 0-form result where onsiteability $\iff$ anomaly-free.
For higher-form symmetries, the obstruction is not the bulk 't Hooft anomaly itself, but the transgression of that anomaly.

B. (2+1)D Finite 1-Form Symmetries

For a finite 1-form symmetry group $G$ in (2+1)D:

Anomaly Index: Characterized by $[\omega_4] \in H^4(B^2G, U(1))$ .
Onsiteability Condition: The symmetry is onsiteable if and only if the transgression of its anomaly vanishes:
$\Phi([\omega_4]) = 0 \in H^3(BG, U(1))$
Physical Interpretation: $\Phi([\omega_4])$ represents the 't Hooft anomaly of the 0-form symmetry generated by restricting the 1-form symmetry operators to a 1D line. If this restricted anomaly is non-trivial, the 1-form symmetry cannot be disentangled into an onsite form.
Pauli Realization: The authors prove that if a 1-form symmetry in (2+1)D is onsiteable, it can be transformed (via ancillas and circuits) into transversal Pauli operators. This implies that onsiteable 1-form symmetries have a structure similar to stabilizer codes (e.g., Toric code).
Example (Semion vs. Fermion):
- Semion ( $\mathbb{Z}_2$ 1-form): Has a non-trivial transgression ( $\Phi(\omega_4) \neq 0$ ). It is not onsiteable.
- Fermion ( $\mathbb{Z}_2$ 1-form): Has a trivial transgression ( $\Phi(\omega_4) = 0$ ) despite having a non-trivial bulk anomaly. It is onsiteable (realized in the Toric code).

C. (3+1)D and Generic Dimensions: Lattice Anomalies

The authors extend the analysis to generic dimensions $(d+1)$ and higher-form symmetries ( $p$ -form).

Lattice Anomaly Indices: They identify new obstructions to onsiteability that exist only on lattices, characterized by indices in $H^{d+2-q}(B^{p+1}G, \text{QCA}_{q-1})$ $H^{d + 2 - q} (B^{p + 1} G, QCA_{q - 1})$ .
- For (3+1)D 1-form symmetry, they define an index $[\omega_3] \in H^3(B^2G, \text{QCA}_1)$ (where $\text{QCA}_1 \cong \mathbb{Q}^+$ ).
Transgression of Lattice Anomalies: The obstruction to onsiteability is determined by the transgression of these lattice indices.
- Condition: $\Phi([\omega_3]) \in H^2(BG, \text{QCA}_1)$ must vanish.
General Conjecture: A finite $p$ -form symmetry in $(d+1)$ D is onsiteable if and only if all successive transgressions of the sequence of "lattice anomaly" indices vanish:
$\Phi_p([\omega_{d+2-q}]) = 0 \in H^{d+2-p-q}(BG, \text{QCA}_{q-1}), \quad \forall 0 \le q \le d+1$
This includes the continuum 't Hooft anomaly (case $q=0$ ) and higher-order lattice obstructions.

4. Significance and Implications

Refinement of Anomaly Theory: The paper resolves the apparent contradiction where a symmetry is anomalous in QFT but onsiteable on a lattice. It clarifies that the relevant anomaly for onsiteability is the transgressed anomaly (obstruction to 1-gauging), not necessarily the bulk anomaly.
Quantum Error Correction & Fault Tolerance: The result that onsiteable 1-form symmetries can be realized as transversal Pauli operators is crucial for quantum computing. Transversal gates are fault-tolerant; this work provides a systematic classification of which symmetries in topological phases can support such gates.
Entanglement Entropy Bounds: The authors note that onsiteability imposes constraints on entanglement entropy. If a symmetry is onsiteable, the entanglement entropy of a region scales differently than if it were obstructed, providing a diagnostic tool for identifying phases with specific symmetry structures.
Lattice vs. Continuum Distinction: By introducing "lattice anomaly" indices involving QCAs, the work highlights that lattice models possess richer structural constraints than continuum field theories, particularly regarding the realization of symmetries in short-range entangled phases.

Summary Conclusion

The paper establishes that onsiteability of higher-form symmetries is equivalent to the triviality of their transgressed anomalies (obstructions to 1-gauging). While continuum 't Hooft anomalies are necessary conditions, they are not sufficient for lattice models; additional "lattice anomaly" indices must also vanish. This framework unifies the understanding of symmetry realization, topological phases, and fault-tolerant quantum computation, demonstrating that onsiteable higher-form symmetries are fundamentally linked to stabilizer-like structures.