⚛️ high-energy theory

Beyond Calabrese-Cardy Scaling: Exceptional-Point Sensitivity from the de Sitter RT Surface

This paper demonstrates that non-Hermitian critical chains near exceptional points exhibit a unique entanglement entropy scaling with an additional $\log(\Delta L)$ term, which arises from the de Sitter geometry of the renormalization flow and signifies that the system's entanglement remains sensitive to small energy gaps even at sub-finite scales, unlike their Hermitian counterparts.

Original authors: Kuang-Hung Chou

Published 2026-07-24

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Kuang-Hung Chou

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ 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

Imagine the universe as a giant, invisible web of connections. In the world of quantum physics, these connections are called "entanglement," a spooky phenomenon where particles remain linked no matter how far apart they are. Scientists often use a measuring stick called "entanglement entropy" to see how strong these links are. Usually, when they look at a chain of particles that is perfectly balanced and critical (like a tightrope walker right on the edge of falling), the amount of entanglement follows a very predictable, famous rule known as the Calabrese–Cardy scaling. It's like a musical score that always plays the same melody, no matter how long the song is.

However, there is a strange, exotic corner of physics called "non-Hermitian" systems. Think of these as worlds where energy can leak in or out, or where the rules of the game are slightly "off-kilter" compared to our everyday reality. In these systems, particles can hit a special "exceptional point," a place where two different states of the system merge into one, creating a unique kind of chaos. For a long time, physicists assumed that if you made the energy gap (the difference between two states) very tiny—so tiny it was smaller than the size of the system itself—the entanglement would just ignore it, acting as if the gap didn't exist. But a new study suggests that in these weird, off-kilter worlds, the entanglement is much more sensitive than we thought. It's as if the tightrope walker can feel a breeze that is too faint for anyone else to notice.

The Paper's Discovery: A Hidden "Residual" Whisper

In this paper, Kuang-Hung Chou explores what happens to entanglement in these non-Hermitian chains when they are near an exceptional point. The main finding is surprising: even when the energy gap ( $\Delta$ ) is incredibly small—so small that it is less than $1/L$ (where $L$ is the total size of the system)—the entanglement entropy still "remembers" it.

Usually, in normal (Hermitian) systems, if a gap is smaller than the system's size, it's effectively invisible to entanglement. It's like trying to hear a whisper in a stadium; the noise of the crowd drowns it out. But in these non-Hermitian chains, the entanglement doesn't get drowned out. Instead, the total entanglement ( $S_A$ ) splits into two parts. The first part is the usual, predictable Calabrese–Cardy melody that depends on the size of the piece of the chain you are looking at. The second part is a new, "residual" term, $S_{res} = \log(\Delta L)$ .

This residual term is fascinating because it doesn't care how big the piece of the chain is. Whether you look at a huge chunk of the system or just a single atom, this extra "whisper" of entanglement is there. It acts like a constant vertical offset on a graph, shifting the whole result up or down depending on the size of the gap and the system. The paper shows through simulations that this term is detectable even for a one-site subsystem, proving that the system retains a sensitivity to the gap that normal systems lose.

The Geometry of the Invisible: A De Sitter Twist

To understand why this happens, the author uses a clever analogy involving geometry and holography (the idea that a 3D world can be described by a 2D surface). In normal physics, we often imagine the entanglement as a rubber band stretched between two points on the surface of a bowl-shaped universe (Anti-de Sitter space). This rubber band dips down but turns around before hitting the bottom, so it never feels the very deepest, most singular part of the bowl.

However, in these non-Hermitian systems, the geometry changes into something called "de Sitter" space. Imagine the bowl turning inside out or becoming a different shape entirely. In this new geometry, the rubber band (the entanglement surface) doesn't just dip and turn; it stretches all the way down to the very bottom of the universe, the "infrared endpoint." Because it reaches the bottom, it gets stuck on a "singular core"—a point of infinite density that represents the tiny energy gap.

Because the rubber band touches this singular core, it can't just ignore the gap. The paper argues that you can't fully "disentangle" (untangle) the system down to a single, simple point without hitting a mathematical wall. Instead, the process must stop at a "two-site" state—a tiny pair of entangled atoms at the very end of the chain. When the author calculates the entanglement of this final two-site pair, it perfectly matches the mysterious $\log(\Delta L)$ term found in the simulations.

What This Means

The paper suggests that this extra entanglement is a "residual" memory left behind because the system couldn't be fully untangled to a simple state. It's the long-range entanglement that survives even after you've tried to simplify the system as much as possible.

The authors are careful to note that this result comes from simulations of free fermions (particles that don't interact with each other) and specific mathematical models. They suggest that this might be a general feature of these types of critical systems, but they don't claim to have proven it for all interacting systems yet. They also point out that this behavior has no equivalent in normal, Hermitian physics. In short, the paper reveals that in the strange, off-kilter world of non-Hermitian physics, the universe is more sensitive to tiny details than we previously believed, and this sensitivity is written into the very geometry of how entanglement flows.

Technical Summary: Beyond Calabrese–Cardy Scaling: Exceptional-Point Sensitivity from the de Sitter RT Surface

Problem Statement
In one-dimensional quantum critical systems, entanglement entropy typically follows the Calabrese–Cardy scaling law, characterized by a logarithmic dependence on subsystem size. This behavior is well-understood in Hermitian unitary systems via Conformal Field Theory (CFT) and holographic Ryu–Takayanagi (RT) surfaces in Anti-de Sitter (AdS) space. However, non-Hermitian (nH) critical chains, particularly those near exceptional points (EPs) associated with logarithmic CFT (LCFT), exhibit unique properties. While previous studies have shown that nH systems can retain the Calabrese–Cardy form with a non-unitary central charge, the sensitivity of entanglement entropy to small energy gaps ( $\Delta$ ) in the presence of EP singularities has been largely overlooked. Specifically, it is unclear how entanglement behaves when the gap $\Delta$ is smaller than the finite-size scale ( $1/L$ ), a regime where such gaps are effectively invisible to entanglement in standard unitary critical chains.

Methodology
The authors investigate this phenomenon using non-Hermitian free fermion models, specifically the two-legged non-Hermitian Su–Schrieffer–Heeger (SSH) model with an EP regulated by a small energy gap $\Delta$ . The study employs a multi-faceted approach:

Lattice Calculation: Exact computation of biorthogonal entanglement entropy for finite systems on a ring of length $L$ with a subsystem of length $\ell$ .
Holographic Interpretation: Utilizing the continuous Multiscale Entanglement Renormalization Ansatz (cMERA) to map the entanglement structure of nH states to an emergent spacetime geometry. The authors analyze the metric generated by the non-unitary RG flow, identifying it as de Sitter (dS) spacetime rather than AdS.
Geometric Analysis: Examining the behavior of extremal surfaces (RT surfaces) in the emergent dS geometry compared to AdS, specifically focusing on how these surfaces interact with the infrared (IR) endpoint of the RG circuit.
Finite-Depth Circuit Analysis: Modeling the RG flow on a finite ring to determine the state of the system after a finite number of disentangling layers, identifying the residual entangled state at the IR endpoint.

Key Contributions and Results
The paper establishes that non-Hermitian critical systems near exceptional points exhibit a distinct scaling behavior for entanglement entropy that differs fundamentally from Hermitian counterparts.

Residual Entropy Term: The entanglement entropy $S_A(\ell)$ for a subsystem of length $\ell$ on a ring of length $L$ is found to follow the form:
$S_A(\ell) = \frac{c}{3} \log \left[ \frac{L}{\pi} \sin \left( \frac{\pi \ell}{L} \right) \right] + \log(\Delta L) + s_0$
where $c = -2$ is the non-unitary central charge and $s_0$ is a non-universal constant. The term $S_{\text{res}} \equiv \log(\Delta L)$ is an interval-independent contribution termed "residual entropy."
Sub-Finite-Size Gap Sensitivity: Unlike unitary critical chains where a gap $\Delta < 1/L$ is invisible to entanglement, the non-Hermitian systems studied here retain sensitivity to $\Delta$ even when $\Delta < 1/L$ . This is evidenced by the fact that the entropy depends on the product $\Delta L$ , causing a vertical shift in the Calabrese–Cardy profile that is detectable even for a one-site subsystem.
Geometric Origin in de Sitter Spacetime: The authors interpret this result through the lens of emergent geometry. In the cMERA construction for nH free fermions, the scale direction acquires a Lorentzian sign, generating a de Sitter (dS) spacetime.
- In AdS (Hermitian case), the RT surface connects boundary endpoints and can bypass the IR singularity.
- In dS (non-Hermitian case), the timelike extremal geodesics extend toward past infinity (the IR endpoint). Consequently, the RT surface must reach the IR endpoint.
IR State Entanglement: A finite-depth RG circuit on a ring cannot terminate in a fully disentangled one-site product state without encountering a singularity in the metric at the IR endpoint. Instead, the circuit must terminate in a nontrivial, entangled two-site IR state. The authors compute the entropy of this specific two-site IR state and find it exactly reproduces the $\log(\Delta L)$ term. This identifies the residual entropy as the long-range entanglement remaining after finite-depth disentangling.

Significance
The paper claims to identify a new class of long-range entanglement in non-Hermitian critical systems that is absent in Hermitian theories. The significance lies in:

Breaking the "Invisibility" of Small Gaps: Demonstrating that non-unitary criticality allows entanglement to resolve energy gaps below the finite-size scale, a phenomenon with no Hermitian analogue.
Holographic Mechanism: Providing a geometric explanation for this behavior via the dS RT surface, which necessitates reaching the IR endpoint and thus inheriting the contribution of the IR state.
Refining LCFT Understanding: Suggesting that logarithmic non-Hermitian criticality carries an additional entanglement contribution beyond standard non-unitary CFT results, rooted in the specific nature of the dS extremal surface.

The authors note that while this behavior is derived for free-fermion models, it raises questions about whether this residual entropy is a general feature of logarithmic non-unitary criticality or specific to free models. They suggest future work should aim to derive this term directly from logarithmic CFT and test its robustness against interactions.

Technical Summary: Beyond Calabrese–Cardy Scaling: Exceptional-Point Sensitivity from the de Sitter RT Surface

More like this