Scaling Dimensions and Two-Point Amplitudes from Operator Covariance Matrices

arXiv:2610.24943 · cond-mat.stat-mech · Submitted 2026-10-18 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Condensed Matter. The day in condensed matter: materials, phases and the measurements under them.

Kai: We're looking at the paper: "Scaling Dimensions and Two-Point Amplitudes from Operator Covariance Matrices".

Mira: Here's the gist of the paper. The long distance behavior of two-point functions of critical lattice statistical mechanics models depend on detailed information of the underlying quantum field theory.

Kai: Scaling dimensions and two-point amplitudes can be extracted from the eigendecomposition of the operator covariance matrix, which allows for a complete determination of the structure of the continuum two-point functions from the lattice covariance matrix.

Mira: Pruning redundant reasoning tokens The paper provides an explicit proof that under the assumption of scale invariance, the leading decay of the eigenvalues are controlled by the smallest Nl scaling dimensions, even after taking irrelevant perturbations into account.

Kai: This understanding allows for a new method where eigenvectors of the covariance matrix can be used to construct amplitudes of continuum two-point correlation functions, enabling a complete determination of the universal structure from the lattice covariance matrix.

Paper summary: Kai: So, to wrap up on "Scaling Dimensions and Two-Point Amplitudes from Operator Covariance Matrices," the authors essentially showed that we can use the eigendecomposition of an operator covariance matrix built from cross correlations of finite lattice operators to determine the continuum two-point function structure.

Mira: The core claim is that this approach provides an explicit proof that the leading decay of eigenvalues is governed by the smallest N-ell scaling dimensions, even when you include irrelevant perturbations. This lets them use those eigenvectors to build amplitudes for continuum correlation functions.

Lev: It means they've proven a way to go beyond just confirming results with Monte Carlo data; they have a method to extract the universal structure from the lattice covariance matrix itself, which is powerful for testing hypotheses.

Kai: The paper by Chen and Kaul focuses on proving that this diagonalization uncovers the hierarchy of scaling dimensions and allows for the extraction of universal amplitudes like k ab from cross ratios, with concrete results on the 2D Ising model matching CFT predictions <ref:2610.24943#pg1>.

Mira: The title suggests a deep connection between these spectral properties and the fundamental scaling behavior of critical models. It connects what we measure on a lattice to what happens in the continuum limit through this matrix method.

Lev: The implication is that for our field, it suggests a systematic way to probe the underlying quantum field theory structure using finite operator sets and their correlations.

Kai: That's it for this piece on how this paper by Chen and Kaul uses operator covariance matrices to extract scaling dimensions from lattice data.

Conclusion: Kai: So, this paper by Chen and Kaul is about using operator covariance matrices to figure out the two-point functions of critical lattice models, basically what you see in these statistical mechanics simulations.

Mira: Right, they're taking all those finite lattice operators and making a covariance matrix, which then lets them extract the scaling dimensions from its eigendecomposition.

Lev: From my side, I’m thinking about how much overhead that calculation would actually take on real hardware if we tried to run it at scale.

Kai: It seems the main takeaway is that this method gives us a way to construct those universal continuum correlation functions directly from the lattice covariance matrix without needing an infinite number of operators.

Mira: The authors prove that under scale invariance, the leading decay of these eigenvalues is controlled by the smallest scaling dimensions, even when you throw in some irrelevant perturbations.

Lev: That means we can use this spectral information to get a solid handle on how fast things decay, which is crucial for any error correction scheme we build.

Kai: And they show this works on synthetic data based on the scaling hypothesis and also numerically exact data from the two dimensional Ising model.

Mira: The results on the 2D Ising model actually match what we expect from conformal field theory, with the extracted amplitudes aligning with their CFT values to within one percent <ref:2610.24943#pg1>.

Lev: That level of agreement is good, but it doesn't tell us much about how this method would scale when we move beyond simple synthetic data sets to more complex physical systems.

Kai: Exactly, the paper shows that these two-point amplitudes are universal and don't depend on the specific lattice geometry you choose.

Mira: So we can test things like orthogonality between different conformal towers just by looking at those cross ratios, which is a neat way to see structure in the theory.

Lev: It opens up a new avenue for testing theory without having to simulate every single possible configuration on a computer.

Department of Physics, The Pennsylvania State University · Center for Theory of Emergent Quantum Matter

cond-mat.stat-mech

Submitted: 2026-10-18

Updated: 2026-10-29

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 79/100

The gist: The long distance behavior of two-point functions of critical lattice statistical mechanics models depend on detailed information of the underlying quantum field theory.

Key concepts

Operator Covariance Matrix: This matrix is built by calculating cross-correlations between a finite set of lattice operators. It depends only on the distance (separation 'r') between the operators due to translation symmetry. Analyzing its eigenvalues reveals crucial information about the underlying quantum field theory, specifically scaling dimensions.
Scaling Dimensions: These dimensions characterize how different types of lattice operators behave under scale transformations at a critical point. The paper proves that the leading decay rate of eigenvalues is controlled by the smallest scaling dimensions, even when considering small irrelevant perturbations, allowing for robust extraction from data.
Two-Point Amplitudes (kab): These are specific amplitudes that describe the correlation functions between two operators in a continuum field theory. The paper shows how to extract these universal values from the eigenvectors of the covariance matrix by analyzing normalized cross ratios, confirming their universality across different lattice geometries like square and triangular lattices.

Terminology

Scaling Dimensions — 42× in this paper · explained in 2 episodes
Degeneracies — 3× in this paper · explained in 2 episodes
Fixed Point — 3× in this paper · explained in 2 episodes
Universality — 3× in this paper · explained in 2 episodes
Clarity — 2× in this paper · explained in 3 episodes
Core — 2× in this paper · explained in 2 episodes
Verification — 2× in this paper · explained in 2 episodes
Adam — 1× in this paper · explained in 3 episodes
Conformal Field Theory (CFT) — 1× in this paper · explained in 3 episodes
Consistency — 1× in this paper · explained in 3 episodes
Continuum Limit — 1× in this paper · explained in 2 episodes
Degeneracy — 1× in this paper · explained in 4 episodes
Generator — 1× in this paper · explained in 2 episodes
Phase Transition — 1× in this paper · explained in 3 episodes
Phase Transitions — 1× in this paper · explained in 2 episodes
Pruning — 1× in this paper · explained in 4 episodes
Scaling Law — 1× in this paper · explained in 2 episodes
Scaling Laws — 1× in this paper · explained in 4 episodes
Thermodynamic Limit — 1× in this paper · explained in 2 episodes

Summary

The long distance behavior of two-point functions of critical lattice statistical mechanics models depend on detailed information of the underlying quantum field theory.

Scaling dimensions and two-point amplitudes can be extracted from the eigendecomposition of the operator covariance matrix, which allows for a complete determination of the structure of the continuum two-point functions from the lattice covariance matrix.

Pruning redundant reasoning tokens

The paper provides an explicit proof that under the assumption of scale invariance, the leading decay of the eigenvalues are controlled by the smallest Nl scaling dimensions, even after taking irrelevant perturbations into account. This understanding allows for a new method where eigenvectors of the covariance matrix can be used to construct amplitudes of continuum two-point correlation functions, enabling a complete determination of the universal structure from the lattice covariance matrix. The method is demonstrated on synthetic data sets drawn from the scaling hypothesis and numerically exact data for the two dimensional Ising model.

How the training signal is built

The core of the analysis involves constructing an Nl × Nl operator covariance matrix Cij (x1, x2) by taking cross correlations of a finite set of Nl lattice operators. This matrix depends only on the separation r = x2 − x1 due to translation symmetry. The eigenvalues λa(r) are shown to have an asymptotic behavior of λa(r) ∼ r − 2∆a.

Where the gains come from

The authors prove that the leading asymptotic behavior of the eigenvalues is controlled by the smallest scaling dimensions, even after taking irrelevant perturbations into account. This proof relies on scale invariance at the critical point. The eigenvectors of the covariance matrix, U(r), are then used to construct orthogonal lattice operators Oea(r; R) which can be analyzed for cross correlation Cab(r; R).

Extraction of kab for the triangular lattice

The two-point amplitudes kab in Eq. (1) can be extracted from the covariance matrix eigenvectors by analyzing the normalized cross ratio ρab(r; R). It is shown that given non-degenerate scaling dimensions, lim r→+∞ lim R→+∞ ρab(r; R) = kab. This extraction is tested numerically on the 2D Ising model, where the predicted amplitudes are k12 = 1/2 in the even sector and k12 = 1/9 in the odd sector.

The analysis reveals that orthogonality between scaling operators belonging to distinct conformal towers can be tested by examining ρab(r; R). When fields belong to distinct conformal towers, their plane two-point amplitude vanishes, and the rescaled quantity Rgρab(r; R) can instead collapse to a nonzero function of r/R determined by the first surviving correction. This demonstrates how orthogonality between different conformal towers can be examined with the operator covariance method.

The numerical verification using synthetic data confirms that eigenvalues decay with an exponent equal to the leading scaling dimensions, as predicted by Eq. (10). Furthermore, the study of synthetic data with an irrelevant perturbation shows that while it changes finite-distance amplitudes, it does not change the asymptotic powers of the eigenvalues. The analysis on the 2D Ising model confirms that the extracted scaling dimensions agree with their assigned CFT values to within one percent.

The application to the triangular lattice shows that for both even and odd sectors, |k12| is invariant under different lattice geometry, indicating a universal property of the two-point amplitudes. This universality is confirmed by showing that the values |k12| for both sectors are invariant under different lattice geometry. The results from the square lattice also show that the values |k12| for both the even and the odd sectors are invariant under different lattice geometry.

The analysis of operator covariance spectra on the infinite triangular lattice shows that for non-degenerate branches, k13 = 0 because fields belong to distinct conformal towers. This demonstrates how orthogonality between different conformal towers can be examined with the operator covariance method. The analysis on the square lattice also shows that the orthogonality imposed by diagonalization leads to a common zero at r/R = 1 for ρ12(r; R). This confirms the universal property of the two-point amplitudes.

The analysis on the square lattice shows that diagonalization demonstrates that the values |k12| for both sectors are invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the square lattice also show that the values |k12| for both sectors are invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the triangular lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the square lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the triangular lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the square lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the triangular lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the square lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the triangular lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the square lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes. The results from the triangular lattice also show that for both even and odd sectors, |k12| is invariant under different lattice geometry. This indicates the universal property of the two-point amplitudes.

Improvements for AI systems

Automated extraction of scaling dimensions from lattice operator covariance matrices: The system can utilize Eq (10) to predict that the leading asymptotic behavior of the eigenvalues is lambda i(r) c r - 2 Omega i and then use numerical data to fit the exponents Omega i, effectively determining the scaling dimensions of continuum operators.
Reconstruction of continuum two-point function amplitudes: The system can implement Section III's method by using eigenvectors to construct the amplitudes kab in Eq (1) from the normalized cross ratio rho ab(r; R) by analyzing its limit as r/R approaches zero and R approaches infinity, allowing for the extraction of universal data.
Verification of conformal invariance: The system can test for orthogonality between scaling operators belonging to distinct conformal towers by checking if the normalized cross ratio converges to zero in the ordered limits of Eq 25, which serves as a check of conformal invariance at the critical point.
Handling finite-distance corrections: The system can analyze how subleading corrections are captured by examining the rescaled correlation function Rg rho ab(r; R) against r/R, allowing it to determine if the "leading non-vanishing correction is of order R−g with g > 0" and extract these finite-distance effects.
Cross-geometry invariance testing: The system can compare the extracted two-point amplitudes |k12| for both the triangular and square lattices to demonstrate the values |k12| for both the even and the odd sectors are invariant under different lattice geometry, confirming the universal property of the two-point amplitudes.