Measurement-Induced Entanglement in Conformal Field Theory

This paper demonstrates that measurement-induced entanglement in Tomonaga-Luttinger liquids is a universal, conformally invariant phenomenon determined by the CFT's operator content, which can be exactly computed via a replica trick and fundamentally differs from entanglement induced by forcing specific measurement outcomes.

The Gist

Imagine a quantum system as a giant, invisible web of connections holding a group of particles together. In a special state called a "quantum critical state," these particles are deeply entangled, meaning their fates are linked across vast distances, like a choir singing in perfect harmony even when separated by miles.

This paper explores what happens to that harmony when you start "listening in" on specific parts of the choir. In the quantum world, "listening" means performing a measurement.

The Big Question: Listening vs. Forcing a Note

Usually, when scientists study what happens when you measure a quantum system, they use a shortcut. They pretend that the measurement always forces the system to pick a specific, pre-determined outcome (like forcing the choir to sing a specific note, say "C"). The authors call this MIEF (Measurement-Induced Entanglement with Forced outcomes).

However, in the real world, measurements are random. When you measure a quantum particle, it doesn't just pick a note you told it to; it picks a note based on probability (like a coin flip). The authors call the real-world scenario MIE (Measurement-Induced Entanglement).

The paper asks: Is the result of a real, random measurement the same as the result of a forced, pre-determined measurement?

The Discovery: They Are Totally Different

The authors found that no, they are not the same.

• The Forced Scenario (MIEF): If you force the system to pick a specific outcome, the remaining particles (the ones you didn't measure) end up with a certain amount of connection. This is like telling the choir to sing "C" and seeing how the rest of the song changes.
• The Real Scenario (MIE): When you let the system choose randomly (following the "Born rule," which is nature's way of deciding probabilities), the remaining particles end up with a different amount of connection.

The authors calculated exactly how much connection remains in the real scenario for a broad class of quantum systems (called Tomonaga-Luttinger liquids). They found that the "real" entanglement is fundamentally different from the "forced" version.

How They Solved the Puzzle: The "Copycat" Trick

Calculating the average of all possible random outcomes is incredibly hard because there are infinite possibilities. To solve this, the authors used a mathematical tool called the replica trick.

Think of it like this:

1. Imagine you have a messy room (the quantum system) and you want to know how messy it is on average after you clean a few spots randomly.
2. Instead of trying to calculate the average of one messy room, you create copies of the room.
3. You clean the spots in all the copies, but you do it in a way that links the copies together mathematically.
4. By looking at how these linked copies interact, you can figure out the average messiness of the single real room without having to simulate every single random outcome.

In the paper, they used this trick to handle the randomness of the measurements. They discovered that the key to the answer lies in something called "winding numbers."

The "Winding" Analogy

Imagine the quantum field is like a piece of elastic string wrapped around a cylinder.

• Forced Measurement: You pin the string down at specific points. The string can only wiggle in a limited way.
• Real Measurement: You pin the string down, but you don't know exactly where it lands. It could be pinned at point A, point B, or anywhere in between, and it might wrap around the cylinder a different number of times (winding) each time.

The authors found that to get the correct answer for real measurements, you have to average over all the possible ways the string could wrap around the cylinder, weighted by how likely each way is to happen.

The "Born Averaging" Insight

The paper concludes with a beautiful interpretation: The entanglement you get from real measurements is simply the average of all possible "forced" scenarios, weighted by how likely each scenario is to occur.

It's like saying: "If you want to know the average temperature of a room, you don't just measure it once. You imagine every possible temperature the room could be, calculate the result for each, and then take a weighted average based on how likely each temperature is."

The Results

The authors didn't just guess; they did the math exactly and checked it with computer simulations (using a model called the XXZ spin chain).

• They found that the "real" entanglement follows a specific, universal pattern that depends on the distance between the unmeasured regions.
• They discovered a surprising mathematical feature: At a certain point (related to a number called $n=1/2$), the behavior of the entanglement changes its character, which is different from the "forced" scenario.
• They confirmed that for real measurements, the system actually gains new long-range connections that wouldn't exist if you just forced a specific outcome.

Summary

In short, this paper shows that randomness matters. You cannot replace the messy, probabilistic nature of real quantum measurements with a clean, forced outcome and expect the same result. The "noise" of the measurement actually creates a unique type of long-distance connection between particles, which the authors have now calculated exactly for a wide class of quantum systems.

Technical Summary

1. Problem Statement

The paper addresses the challenge of analytically calculating Measurement-Induced Entanglement (MIE) in long-range entangled quantum critical states.

• Context: Local measurements can drastically reshape many-body entanglement patterns. While numerical studies have shown that MIE can induce long-range entanglement and criticality, analytical results are scarce.
• The Gap: Previous analytical work often relied on Forced MIE (MIEF), where the measurement outcome is post-selected to a specific state (e.g., $|1010\dots\rangle$). This ignores the intrinsic randomness of quantum measurements. The true MIE requires averaging over all possible measurement outcomes weighted by their Born probabilities ($p_m$).
• The Challenge: The definition of MIE involves a non-linear average of entanglement entropy ($\sum p_m S_m$), which is difficult to compute analytically in field theory because the averaging prevents the use of standard Boundary Conformal Field Theory (BCFT) techniques that rely on fixed boundary conditions.
• Target System: The authors focus on Tomonaga-Luttinger Liquids (TLL), a broad class of 1+1D quantum critical states described at low energies by a compact free boson Conformal Field Theory (CFT).

2. Methodology

The authors develop a novel analytical framework combining the replica trick with free-boson specific techniques to handle the Born-weighted averaging.

• Replica Trick for Non-Linearity:
  To compute the average $\sum p_m S_m$, they employ a double replica limit. They introduce two replica indices:

  1. $n$: The standard Rényi index for entanglement entropy.
  2. $k$: An auxiliary index introduced to handle the non-linearity of the Born average.
     The MIE is expressed as:
     $$\text{MIE}(A) = \lim_{n\to 1} \lim_{k\to 0} \frac{1}{1-n} \frac{d}{dk} \log \left( \frac{Z_A}{Z_0} \right)$$
     where $Z_A = \sum_m p_m (\text{tr} \rho_{m,A}^n)^k$ and $Z_0 = \sum_m p_m (\text{tr} \rho_m)^{nk}$.

• Path Integral Representation:
  The partition functions $Z_A$ and $Z_0$ are mapped to Euclidean path integrals on replicated Riemann surfaces ($M_n$).

  • Measurement as Defects: The measurement region $C$ acts as a slit or defect where the bosonic field $\phi$ is pinned to specific values $m(x)$.
  • Quantum-Classical Split: The free boson action allows the partition function to be factorized into a "quantum" fluctuation part (independent of measurement outcomes) and a "classical" part dependent on the boundary conditions (measurement outcomes).

• Handling Winding Sectors (The Key Innovation):
  In compact boson CFTs, the field $\phi$ has a compactification radius, leading to topological winding sectors ($w \in \mathbb{Z}$).

  • In the Forced (MIEF) case, the boundary conditions are fixed, and the classical contribution is a simple sum over windings.
  • In the True MIE case, the Born averaging couples different replicas. The authors utilize a basis rotation technique (inspired by Fradkin and Moore) to decouple the replicas. They rotate the $Q = nk+1$ replica fields such that $Q-1$ fields have vanishing boundary conditions (modulo winding), while a single "center-of-mass" field retains the explicit dependence on the measurement outcome $m$.
  • Crucially, they account for winding sums in this rotated basis, which were previously neglected in similar contexts. This leads to a "winding function" $W_{n,k}$ that captures the essential physics of the Born averaging.

• Conformal Mapping:
  The problem is mapped from the infinite cylinder to a finite cylinder $C(n)$ of height $h(\zeta)$ and circumference $\beta=2\pi$, where $\zeta$ is the conformal cross-ratio of the intervals. This allows the extraction of universal, scale-invariant results.

3. Key Contributions

1. First Exact Analytic Result for MIE: The paper provides the first closed-form analytical expression for the MIE in a class of long-range entangled ground states (compact free boson CFT).
2. Distinction Between MIE and MIEF: The authors rigorously demonstrate that physical MIE is fundamentally different from post-selected MIEF. The difference arises entirely from the winding contributions coupled by the Born averaging.
3. Born-Averaging Interpretation: They derive a physical interpretation where the MIE is equivalent to a Born average over Dirichlet boundary conditions. The measurement outcomes effectively sample different Dirichlet boundary conditions $(\phi_1, \phi_2)$ at the measurement regions, weighted by the partition function of the system with those specific boundary conditions.
4. Universal Scaling Laws: They derive exact scaling behaviors for the MIE in the limit of large separation (small cross-ratio $\zeta \ll 1$).

4. Key Results

A. Analytical Expression

The MIE for Rényi index $n$ is given by:
$$\text{MIE}^{(n)}(A) = \frac{1}{1-n} \left[ W'_n - n W'_1 - \log \frac{\eta(q_n)}{\eta(q_1)^n} \right]$$
where $W'_n$ is the derivative of the winding function with respect to the replica index $k$ (evaluated at $k=0$), and $\eta$ is the Dedekind eta function representing quantum fluctuations.

B. Asymptotic Behavior ($\zeta \ll 1$)

The scaling of MIE with the cross-ratio $\zeta$ exhibits a phase transition at $n=1/2$, distinct from the forced case:

• For $n < 1/2$: $\text{MIE} \sim \zeta^{2n(1-n)g}$
• For $n \ge 1/2$: $\text{MIE} \sim \zeta^{g/2} \frac{1}{\sqrt{\log(1/\zeta)}}$

Contrast with Forced MIE (MIEF):

• MIEF scales as $\zeta^{2ng}$ for $n<1$ and $\zeta^{2g}$ for $n>1$.
• The true MIE has a smaller exponent ($g/2$ vs $2g$) for $n > 1/2$, indicating that physical measurements induce more long-range entanglement than post-selected measurements.
• The appearance of the $1/\sqrt{\log(1/\zeta)}$ prefactor is a novel feature not predicted by standard Operator Product Expansion (OPE) analyses.

C. Measurement-Induced Information (MII)

The authors analyze the Measurement-Induced Information (MII), defined as the change in mutual information due to measurement.

• Real Measurements: MII is positive, meaning measurements increase correlations between distant regions.
• Forced Measurements: MII is negative, meaning post-selecting a specific outcome (like the antiferromagnetic state) actually decreases correlations compared to the pre-measurement state. This highlights the inadequacy of MIEF as a proxy for physical measurement protocols.

D. Numerical Verification

The authors performed numerical simulations on the XXZ spin-1/2 chain (which maps to a TLL) using:

• Free fermions ($\Delta=0$): Exact diagonalization.
• Interacting cases ($\Delta \neq 0$): Density Matrix Renormalization Group (DMRG) with Matrix Product States (MPS).
• Results: The numerical data for MIE (averaged over Born probabilities) collapses perfectly onto the theoretical curves derived from the CFT, validating the analytical predictions across different interaction strengths ($\Delta$) and Rényi indices ($n$).

5. Significance and Implications

• Fundamental Physics: The work establishes that the randomness of quantum measurements is not just a technical nuisance but a fundamental driver of entanglement structure. The "Born averaging" over boundary conditions is a universal mechanism in CFTs.
• Quantum Computation: Since MIE is a resource for Measurement-Based Quantum Computation (MBQC), these results clarify the resources available in physical systems versus idealized post-selected scenarios.
• Phase Diagnostics: The distinct scaling of MIE provides a new tool to diagnose symmetry-enriched CFTs and gapless symmetry-protected topological (SPT) phases.
• Methodological Advance: The technique of using a double replica trick to handle Born-weighted sums of non-linear observables in compact boson theories opens the door for analyzing other measurement-induced phenomena in higher dimensions and different CFTs.

In summary, this paper bridges the gap between theoretical CFT calculations and the physical reality of quantum measurements, proving that the "Born rule" averaging leads to a unique, universal, and enhanced form of long-range entanglement that cannot be captured by fixing measurement outcomes.