Bootstrapping Symmetries in Quantum Many-Body Systems from the Cross Spectral Form Factor

This paper introduces a bootstrap framework that utilizes the cross spectral form factor and spectral correlations to systematically reconstruct the full representation theory and identify hidden finite group symmetries in quantum many-body systems, including their character tables and fusion rules, without prior knowledge of the symmetry group.

The Gist

Imagine you are a detective trying to solve a mystery inside a locked room. The room is a quantum machine (a complex system of particles), and it has a secret rulebook—a symmetry—that dictates how the particles behave.

Usually, to find the rulebook, you need to see the gears and levers (the mathematical equations) that make the machine work. But sometimes, the rulebook is hidden. The gears look messy, but the machine still follows a hidden pattern.

This paper introduces a new detective tool called "Bootstrapping Symmetries." Instead of looking at the gears, this tool listens to the machine's "heartbeat" (its energy levels) to figure out the secret rulebook.

Here is how it works, broken down into simple concepts:

1. The Mystery: Hidden Rules

In physics, symmetries are like rules that say, "If you do this, the system looks the same."

• Obvious Symmetry: Like a snowflake. You can rotate it, and it looks the same. You can see this easily.
• Hidden Symmetry: Imagine a magic trick where a deck of cards seems shuffled randomly, but there's actually a secret order. In quantum machines, these hidden rules are often so complex or "non-local" (spread out across the whole machine) that we can't see them just by looking at the parts.

2. The Clue: The Cross Spectral Form Factor (xSFF)

To find the hidden rule, the authors listen to the machine's "heartbeat."

• The Heartbeat (SFF): If you tap a bell, it rings. The pattern of the ringing tells you about the bell's shape. In quantum physics, the "ringing" is the pattern of energy levels.
• The New Twist (xSFF): Usually, scientists listen to one part of the machine at a time. This paper introduces a new way to listen: Cross-correlation. Imagine listening to two different rooms in a house at the same time. If the sound in Room A and Room B are perfectly synchronized in a weird way, it proves there is a hidden pipe connecting them, even if you can't see the pipe.
• The Plateau: When they listen for a long time, the "heartbeat" settles into a flat, steady hum (a plateau). The height and pattern of this hum are like a fingerprint. It doesn't matter if the machine is chaotic or calm; this fingerprint reveals the hidden symmetry's structure.

3. The Method: The Bootstrap

"Bootstrapping" is like pulling yourself up by your own bootstraps. You start with a little bit of known information and use logic to pull out the rest.

• The Known Subgroup: The detective starts with a small, known piece of the puzzle. Maybe they know the machine has a simple "flip" symmetry (like a light switch).
• The Algebraic Rules: The detective knows the rules of the game (mathematics). For example, if you combine two types of symmetry, you must get a third type. You can't just make up new rules; they have to fit together like a puzzle.
• The Process:
  1. Measure: Use the xSFF to get the fingerprint (the plateau heights).
  2. Guess: The computer tries to build a "candidate" rulebook that fits the known piece and the fingerprint.
  3. Test: It checks if the candidate rulebook follows all the algebraic rules (like a puzzle piece fitting into a slot).
  4. Repeat: If it doesn't fit, it tries a different candidate. Eventually, only one rulebook fits perfectly.

4. The Results: Solving the Case

The authors tested this on several "locked rooms" (quantum models):

• The S3 Chain: They found a hidden symmetry group called $S_3$ (like the symmetries of a triangle) just by listening to the energy levels, even though they only knew about a smaller $Z_3$ part.
• The Kennedy-Tasaki Chain: This was a tricky one where the symmetry was hidden by a "non-local" transformation (like a magic spell that rearranges the whole room at once). The bootstrap method found the hidden $D_4$ symmetry without needing to know the spell.
• The Fermi-Hubbard Model: They rediscovered a famous, complex symmetry called $\eta$-pairing (related to superconductivity) purely from spectral data.

5. Why This Matters

Before this, finding hidden symmetries was like trying to guess the shape of a shadow by looking at the object casting it. If the object was too complex, you were stuck.
Now, this method allows physicists to:

• Identify the "Shape" of the Shadow: They can reconstruct the entire group theory (the number of rules, how they combine, and their dimensions) just from the energy data.
• Work with Black Boxes: You don't need to know the internal wiring of the quantum computer. You just need to measure its output.
• Future Applications: This could help experimentalists on real quantum computers find new phases of matter or hidden laws of nature that we haven't even written down yet.

The Analogy in a Nutshell

Imagine you are in a dark room with a complex machine. You can't see the machine, but you can hear the sounds it makes when you turn the dials.

• Old Way: You try to guess the machine's design by listening to one dial at a time. It's hard to tell if the machine is a toaster or a radio.
• New Way (This Paper): You listen to how the sounds of two different dials interact with each other. The way they "dance" together reveals the exact blueprint of the machine, even if the machine is hidden in a box. The "bootstrap" is the logic that says, "If the dance looks like this, the machine must be built this way, because no other design could produce that specific dance."

This paper essentially gives us a new pair of ears to hear the hidden music of the quantum world.

Technical Summary

1. Problem Statement

Symmetries are fundamental to organizing quantum many-body physics, governing phase transitions, and classifying phases of matter. However, identifying hidden symmetries—those not manifest in the Hamiltonian's local operators or arising from non-local generators, emergent algebraic structures, or Hilbert space fragmentation—remains a significant challenge.

Existing methods fall into two categories, both with limitations:

• Algebraic Methods: Construct the commutant algebra of the Hamiltonian. While exact in principle, they require explicit access to wavefunctions and suffer from exponential growth in complexity with system size.
• Spectral Methods: Use level-spacing statistics (e.g., gap ratios) to detect the presence of hidden symmetries or estimate the number of symmetry sectors. However, these are "coarse" diagnostics; they cannot identify the specific symmetry group, enumerate irreducible representations (irreps), or determine the algebraic structure (fusion rules, character tables).

The Gap: There is no systematic method to reconstruct the full representation-theoretic data of a hidden symmetry group $G$ using only spectral data, without prior knowledge of the group's generators.

2. Methodology: The Bootstrap Framework

The authors propose a bootstrap framework that combines numerical spectral data with algebraic consistency conditions to reconstruct the representation theory of a hidden finite group $G$.

A. Physical Setup and Inputs

• Known Subgroup ($N$): The method assumes partial knowledge: a subgroup $N \subset G$ is known (its generators and irreps are accessible). The Hilbert space can be resolved into $N$-symmetry sectors.
• Unknowns: The full group $G$, its irreps, and crucially, the branching multiplicities ($b_{\lambda, \alpha}$), which describe how each irrep $\alpha$ of $G$ decomposes into irreps $\lambda$ of $N$ upon restriction.

B. The Cross Spectral Form Factor (xSFF)

To extract constraints on the branching multiplicities, the authors introduce a novel observable: the Cross Spectral Form Factor (xSFF).

• Definition: A generalization of the standard Spectral Form Factor (SFF) that measures cross-correlations between distinct symmetry sectors of the known subgroup $N$.
  $$K^{(N)}_{\lambda_a, \lambda_b}(t) = \text{Re} \left\langle \frac{\text{tr}[P_{\lambda_a} U(t)] \text{tr}[P_{\lambda_b} U^\dagger(t)]}{d_{\lambda_a} d_{\lambda_b}} \right\rangle$$
  where $P_\lambda$ projects onto the $N$-irrep sector $\lambda$.
• Physical Insight: While the intermediate-time ramp of the SFF depends on chaos vs. integrability, the late-time plateau is universal. The height of the xSFF plateau is determined strictly by the branching rules and the dimensions of the multiplicity spaces, independent of the system's dynamical nature (chaotic or integrable).
• Constraints Derived:
  1. Non-vanishing Plateaus: If $K^{(N)}_{\lambda_a, \lambda_b} > 0$, the corresponding Gram matrix element $G_{\lambda_a \lambda_b} = \sum_\alpha b_{\lambda_a, \alpha} b_{\lambda_b, \alpha}$ must be positive.
  2. Degenerate Plateaus: If $K^{(N)}_{\lambda_a, \lambda_a} = K^{(N)}_{\lambda_b, \lambda_b} = K^{(N)}_{\lambda_a, \lambda_b}$, then the branching vectors are identical ($\vec{b}_{\lambda_a} = \vec{b}_{\lambda_b}$).
  3. Multiplicity-Free Check: Comparing diagonal plateaus to theoretical benchmark lines determines if branching multiplicities are restricted to $\{0, 1\}$.

C. The Bootstrap Algorithm

The algorithm iteratively searches for a candidate group $G$ (parameterized by its rank $r$, the number of irreps) that satisfies:

1. Numerical Constraints: Derived from the xSFF plateaus (branching vector types and degeneracies).
2. Algebraic Constraints:
   • Monoidality: The fusion of restricted representations must match the restriction of fused representations (Eq. 21).
   • Dimensionality: $\sum N_{\alpha_i \alpha_j}^{\alpha_k} d_{\alpha_k} = d_{\alpha_i} d_{\alpha_j}$.
   • Associativity & Commutativity: Standard fusion algebra properties.
   • Rigidity: Existence of dual representations and unity conditions.
3. Output: The algorithm outputs the branching matrix $B_{\lambda, \alpha}$, fusion coefficients $N_{\alpha_i \alpha_j}^{\alpha_k}$, and the character table of $G$.

The algorithm handles both linear representations (standard groups) and corepresentations (magnetic groups with anti-unitary symmetries).

3. Key Contributions

1. Introduction of xSFF: A new spectral observable that encodes branching structure information, bridging the gap between coarse spectral statistics and detailed algebraic identification.
2. Systematic Reconstruction: A computational protocol that uniquely identifies hidden symmetry groups and their full representation theory (irreps, dimensions, fusion rules, character tables) using only spectral data and a known subgroup.
3. Handling Complexity: The framework successfully addresses:
   • Non-local hidden symmetries (Kennedy-Tasaki transformation).
   • Higher branching multiplicities (multiplicity-free vs. non-free).
   • Anti-unitary symmetries and non-normal subgroups.
   • Projective representations and compact Lie groups (extensions).

4. Results and Examples

The authors validate the framework on several quantum lattice models:

• Example I: $S_3$ Symmetry (O'Brien-Fendley Model):
  • Input: Known $Z_3$ subgroup.
  • Result: Successfully reconstructed the hidden $S_3$ symmetry, recovering the correct branching rules, fusion ring, and character table.
• Example II: Non-local Hidden Symmetry (Kennedy-Tasaki Transformed Spin-1 Chain):
  • Input: Known on-site $V_4 \cong Z_2 \times Z_2$ symmetry. The true symmetry $D_4$ is hidden by a non-local unitary transformation.
  • Result: The bootstrap identified $D_4$ without needing to find the explicit non-local transformation, demonstrating the method's ability to bypass complex operator analysis.
• Example III: Higher Branching Multiplicities (Extended Ashkin-Teller Model):
  • Input: Known $V_4$ subgroup.
  • Result: Successfully handled cases where branching multiplicities $b_{\lambda, \alpha} > 1$, reconstructing the $S_4$ symmetry group.
• Example IV: Anti-unitary & Non-normal Subgroups (Three-state Quantum Torus Chain):
  • Input: Known unitary subgroup $N \cong Z_3^2 \rtimes Z_2$.
  • Result:
    • At generic points, correctly identified the magnetic group structure ($S_3^2$) via corepresentation theory.
    • At the self-dual point, identified a hidden $Z_3^2 \rtimes Z_4$ symmetry where the known subgroup is not normal, a scenario previously difficult to handle.
• Extensions:
  • Projective Representations: Applied to the driven Bose-Hubbard model, detecting signatures of projective representations of $D_8 \times Z_2$.
  • Lie Groups: Applied to the 1D Fermi-Hubbard model, recovering the famous $\eta$-pairing $SO(4)$ symmetry from $U(1) \times U(1)$ data.

5. Significance and Outlook

• Practical Route to Symmetry Discovery: This work provides a practical, data-driven route to identify symmetries directly from dynamical spectral observables, bypassing the need for explicit wavefunction analysis or guessing non-local operators.
• Universality: The method works for both chaotic and integrable systems, as the late-time plateau is a universal feature determined by symmetry structure.
• Experimental Feasibility: The authors propose a protocol to measure xSFF using randomized measurements on quantum simulators (similar to recent SFF experiments), making this a viable tool for near-term quantum devices.
• Future Directions: The framework opens avenues for studying generalized symmetries (non-invertible, higher-form), extracting F-symbols via higher-moment xSFF, and applying similar bootstrap techniques to quantum states (wavefunctions) rather than just Hamiltonians.

In summary, this paper establishes a powerful "spectral bootstrap" that transforms raw energy-level correlations into a complete algebraic description of hidden symmetries, fundamentally advancing the toolkit for analyzing quantum many-body systems.