⚛️ high-energy theory

Krylov Complexity and $c$ -function along RG Flows

This paper investigates Krylov spread complexity along holographic renormalization-group flows, establishing that while complexity acceleration increases as the covariant central charge decreases in fixed-dimensional flows, the two quantities become co-monotonic in flows across dimensions due to the reorganization of degrees of freedom into lower-dimensional sectors.

Original authors: Carlos Nunez, Dibakar Roychowdhury

Published 2026-08-05

📖 8 min read🧠 Deep dive

CC BY 4.0

Original authors: Carlos Nunez, Dibakar Roychowdhury

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, cosmic library. Inside this library, every possible state of matter and energy is a unique book. Some books are simple, like a blank page; others are incredibly complex, filled with intricate stories of interacting particles. Physicists have long wondered: how do we measure the "difficulty" of writing a specific story? This is where Quantum Complexity comes in. Think of it not as how long a story is, but how many steps it takes to assemble the ingredients of a specific quantum state from scratch. It's like asking, "How many Lego bricks do I need to snap together, and in what order, to build a spaceship?"

Now, imagine you have a special tool called the Krylov Complexity. Instead of just counting bricks, this tool tracks how a quantum state "spreads out" through the library over time. It's like watching a drop of ink fall into a glass of water. At first, the ink is a tight, simple blob. As time passes, it spreads, mixing with more and more parts of the water, becoming harder and harder to describe. Scientists have recently discovered a surprising link between this spreading ink and the shape of space itself. In the world of Holography, a theory suggesting our 3D universe might be a projection of a higher-dimensional reality, the "spreading" of a quantum state is mathematically equivalent to a heavy particle falling through the gravity of a black hole. The faster the ink spreads, the faster the particle falls.

This brings us to the big question: What happens to this "spreading speed" when the universe itself changes? In the quantum world, we have something called Renormalization Group (RG) flows. Think of this as zooming a camera in and out. When you zoom out (looking at the big picture), you see fewer details and fewer active ingredients. When you zoom in, you see a chaotic mess of tiny interactions. Physicists have a rule of thumb called the c-function, which acts like a counter for how many "degrees of freedom" (or active ingredients) exist at any zoom level. Usually, as you zoom out toward the "infrared" (the low-energy, big-picture view), this counter goes down. The universe gets simpler.

But here is the mystery: Does the speed at which our quantum "ink" spreads (the complexity acceleration) speed up or slow down as the universe gets simpler? Does the ink spread faster when there are fewer ingredients, or does it slow down? This is the puzzle the authors of this paper set out to solve.

The Paper's Story: Falling Particles and Changing Dimensions

In this paper, the authors, Carlos Nuñez and Dibakar Roychowdhury, act like cosmic detectives. They use the "falling particle" trick to measure how fast quantum complexity grows (specifically, they look at the acceleration of this growth, which they call U) and compare it to the "ingredient counter" (the c-function) as the universe flows from high energy to low energy.

They found that the answer depends entirely on how the universe changes its shape. They discovered two very different stories:

Story 1: The Same-Sized Room (Same-Dimensional Flows)
Imagine you are in a room, and you slowly remove furniture. The room gets emptier, but it's still a 3D room. This is what happens in "Lorentz-invariant domain-wall flows." The universe stays the same size, but the number of active quantum ingredients decreases as you move toward the infrared.

In this scenario, the authors found a perfect inverse relationship. As the "ingredient counter" (c-function) goes down, the "spreading acceleration" (U) goes up.

The Analogy: Think of a crowded dance floor. When it's packed with people (high energy, many degrees of freedom), it's hard for anyone to move fast; the crowd is stiff. But as people leave the floor (flowing to the infrared), the remaining dancers have more space. Suddenly, they can spin and jump with much greater acceleration. The fewer the dancers, the faster they can move.
The Finding: The paper shows that for these flows, the product of the ingredient counter and the spreading acceleration follows a strict mathematical rule. As the universe simplifies, the complexity of the remaining state accelerates wildly.

Story 2: The Shifting Dimensions (Flows Across Dimensions)
Now, imagine a different scenario. Instead of just removing furniture, you suddenly fold the room. A 4D room folds down into a 2D hallway. This happens in "compactification," where extra dimensions of space curl up so tightly that they disappear from our view.

Here, the rules change completely. The authors found that in these cases, the "ingredient counter" and the "spreading acceleration" move together. They are co-monotonic.

The Analogy: Imagine a massive orchestra playing in a huge concert hall (6D). Suddenly, the musicians are forced to squeeze into a tiny, narrow tunnel (4D). Even though the tunnel is smaller, the music doesn't get simpler; it gets reorganized. The sound waves bounce off the walls in new, complex ways. The "complexity" of the sound doesn't slow down just because the space got smaller; in fact, the way the sound spreads might speed up because the new, tighter geometry forces the waves to interact differently.
The Finding: In these dimension-shifting flows, as the c-function increases (or behaves differently due to the new geometry), the complexity acceleration also increases. This is the opposite of the first story. The authors argue this proves that the universe isn't just "losing" ingredients; it's reorganizing them into new, lower-dimensional patterns.

What They Actually Did and Found

The authors didn't just guess; they did the heavy math. They took specific, well-known models of the universe (like those involving Dp-branes, which are like membranes in string theory, and wrapped M5-branes) and calculated exactly how a particle would fall through them.

They derived a new formula: They showed that for standard flows where the dimension stays the same, the c-function and the complexity acceleration are locked in an inverse relationship. They even found a universal rule for Dp-branes: the c-function multiplied by the acceleration to the power of 8 equals a constant. This means the relationship holds true regardless of the specific type of brane, as long as the dimension doesn't change.
They tested the "Dimension Change" hypothesis: They looked at flows where a 4D theory becomes 2D, and a 6D theory becomes 4D. In these cases, they calculated the exact values and found the relationship flipped. Instead of one going up while the other goes down, they both went up together.
They ruled out a simple explanation: The paper argues against the idea that complexity always slows down as the universe simplifies. The "dimension-changing" flows prove that the universe can get simpler in terms of dimensions but more complex in terms of how information spreads, because the ingredients are being rearranged, not just deleted.

How Sure Are They?

The authors are very confident in their mathematical derivations. They didn't run a computer simulation or guess; they solved the equations of motion for particles falling in these specific, theoretical universes. The results are proven within the context of their mathematical models.

However, they are careful to note that these are theoretical models based on "top-down" string theory constructions. While the math is solid, they acknowledge that in the real world, we can't easily test these specific high-energy flows. They suggest that their findings offer a new "diagnostic tool": by measuring how complexity accelerates, we might be able to tell if a physical system is just losing energy (same-dimensional flow) or fundamentally changing its structure (dimension-changing flow).

The Takeaway

The main message is a beautiful twist on how we view the universe's evolution.

If the universe just gets "emptier" but stays the same size, the remaining stuff gets more frantic (complexity acceleration goes up as degrees of freedom go down).
If the universe folds up and changes its shape, the remaining stuff gets more organized in a way that keeps the complexity acceleration high, even as the c-function behaves differently.

This suggests that "complexity acceleration" is a sensitive detector. It can tell us whether the universe is simply running out of steam or if it's undergoing a deep, structural reorganization. It's like having a speedometer that doesn't just tell you how fast a car is going, but also whether the car is driving on a straight road or turning a corner. The authors have built a new tool to read the map of the quantum universe.

Technical Summary: Krylov Complexity and C-function along RG Flows

Problem Statement
The paper investigates the relationship between Krylov spread complexity and the renormalisation-group (RG) flow of quantum field theories (QFTs) within the holographic framework. While Krylov complexity has been applied to various dynamical regimes (chaos, integrability, confinement), its behavior along RG flows remains poorly understood. Specifically, it is unclear how the reorganisation of degrees of freedom during an RG flow is encoded in the Lanczos coefficients or the probability current of the Krylov chain. The authors aim to determine whether the geometric evolution of spread complexity can be directly related to standard holographic measures of the number of field-theory degrees of freedom, specifically the covariant central charge ( $c_{cov}$ ).

Methodology
The authors employ the "proper-momentum proposal" established in prior works [5, 8–10], which posits that the growth rate of spread complexity for a locally excited state is equal to the proper radial momentum of an infalling massive probe in the dual bulk geometry.

Geometric Setup: The study considers generic holographic backgrounds of the form $ds^2 = bA(r, \vec{y})(-dt^2 + d\vec{x}^2) + bB(r, \vec{y})dr^2 + g_{ij}dy^idy^j$ .
Probe Dynamics: A massive point-like particle is released from rest at a UV radial position $r_{UV}$ and falls radially. The authors derive the trajectory $r(t)$ , the proper momentum, and the Krylov complexity $C_K(t)$ and its time derivatives.
Key Observable: The authors focus on the second time derivative of the complexity, $\ddot{C}_K(t)$ , defining a dimensionless "complexity acceleration" quantity:
$U(t) = \frac{2}{m\sqrt{A(r_{UV})}} \ddot{C}_K(t) = \frac{A'(r)}{\sqrt{A^3(r)B(r)}}$
This quantity is shown to be a local combination of the redshift ( $A$ ) and radial metric ( $B$ ) functions.
C-function: The study utilizes the covariant holographic c-function defined in Ref. [34], which is constructed from the extrinsic curvature of spacelike slices and remains valid in ten or eleven dimensions, retaining information about internal manifolds.
Comparative Analysis: The authors analyze the relationship between $U$ $U$ and $c_{cov}$ $c_{co v}$ across three distinct classes of flows:
- Lorentz-invariant domain-wall flows preserving spacetime dimension.
- Top-down Dp-brane flows (preserving dimension).
- Flows across dimensions via twisted compactification (e.g., 4D to 2D, 6D to 4D).

Key Results

Same-Dimensional Flows (Domain Walls):
For Lorentz-invariant domain-wall flows in $(d+1)$ dimensions with metric $ds^2 = e^{2a(r)}dx_{1,d-1}^2 + dr^2$ , the authors derive a new, remarkably simple inverse relation:
$c_{cov}(r) = \frac{1}{G_N^{(d+1)}} U(r)^{d-1}$
Under the weak energy condition, $c_{cov}$ decreases monotonically towards the infrared (IR), while $U$ increases monotonically along the falling trajectory. This implies an anti-correlation: as effective degrees of freedom are lost, the acceleration of the Krylov spread increases. This is illustrated using the GPPZ flow.
Same-Dimensional Flows (Dp-branes):
For the top-down Dp-brane family, the authors find a universal relation independent of the radial coordinate:
$c_{cov} \cdot U^8 = \Upsilon(p)$
Here, $\Upsilon(p)$ depends on the brane dimension $p$ , the Yang-Mills coupling, and the number of colors, but the exponent 8 is universal across the family. This generalizes the inverse relation found for domain walls, showing that the central function and complexity acceleration probe complementary aspects of the same radial evolution.
Flows Across Dimensions (Compactification):
The authors examine flows where the spacetime dimension changes due to twisted compactification (e.g., wrapped M5-branes from 6D to 4D, and twisted Klebanov-Witten from 4D to 2D).
- Result: In these cases, $c_{cov}$ and $U$ are co-monotonic. Both quantities increase (or decrease together) as the flow proceeds towards the IR.
- Specific Relations: For the 4D to 2D flow, an explicit algebraic relation is derived: $c_{cov} \propto (4-U)^{-3}$ . For the 6D to 4D wrapped M5 flow, a parametric relation is established with controlled UV and IR expansions.
- Interpretation: This reversal from anti-correlation to co-monotonicity reflects the fact that compactification reorganizes degrees of freedom into lower-dimensional sectors rather than simply depleting them. The breaking of the original Lorentz group and the presence of massless modes on the compact manifold alter the behavior of the c-function.

Significance and Claims
The paper claims to provide the first systematic comparison between a holographic c-function and the second time derivative of spread complexity along RG flows. The primary significance lies in identifying the complexity acceleration ( $U$ ) as a sensitive geometric diagnostic that distinguishes between two types of RG evolution:

Depletion: In fixed-dimensional flows, the loss of degrees of freedom corresponds to an increasing complexity acceleration (anti-correlation).
Reorganisation: In flows across dimensions, the reorganisation of degrees of freedom into lower-dimensional sectors corresponds to a co-monotonic behavior.

The authors argue that this reversal indicates that $U$ captures the structural reorganisation of the Krylov chain (potentially involving block-Lanczos structures or symmetry-resolved sectors) rather than just the total count of degrees of freedom. The work suggests that the acceleration of spread complexity contains information not captured by the complexity or the central function separately, offering a new bridge between RG irreversibility, spectral recursion, and quantum information spreading.

Limitations and Caveats
The authors note that for Dp-branes, the supergravity description has a finite range of validity due to divergent dilaton and curvature invariants; calculations beyond these bounds should be viewed as qualitative extrapolations. Furthermore, the proper-momentum equality is strictly established for specific AdS $_3$ /CFT $_2$ states, and its application to general top-down, non-conformal backgrounds is a physically motivated extension. The paper also acknowledges that $U$ is not itself a Lanczos coefficient but constrains an aggregate probability current, and a complete field-theory derivation would require reconstructing the full Lanczos sequence.

The Paper's Story: Falling Particles and Changing Dimensions

What They Actually Did and Found

How Sure Are They?

The Takeaway

More like this