⚛️ high-energy theory

$\mathcal{N}=2$ RG flows, Non-Invertible Symmetries and Matrix Factorisations

This paper demonstrates that topological defect lines in $\mathcal{N}=2$ minimal models, which are preserved under the least relevant perturbation to first order, fail to define symmetries in the massless IR theory due to a second-order obstruction linked to a supersymmetry anomaly, whereas they remain consistent symmetries in the associated massive integrable flow.

Original authors: Federico Ambrosino, Matthias R. Gaberdiel, Yu Nakayama

Published 2026-08-05

📖 4 min read🧠 Deep dive

CC BY 4.0

Original authors: Federico Ambrosino, Matthias R. Gaberdiel, Yu Nakayama

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 river flowing from a turbulent, high-energy source (the "UV") down to a calm, low-energy lake (the "IR"). Physicists call this journey a "Renormalization Group flow." For decades, they've believed that if you have a special rule or symmetry in the source river—like a specific way the water swirls—that rule must survive the journey and still exist in the lake. It's like assuming that if a river has a unique pattern of eddies at the top of a waterfall, those same eddies must still be swirling in the pool at the bottom.

In recent years, scientists discovered that these "rules" aren't just simple on/off switches; they can be complex, non-invertible "topological defect lines." Think of these lines as invisible, magical ribbons woven into the fabric of space-time. If you stretch or twist the ribbon, the physics doesn't change. The big question was: if you push the river with a gentle nudge (a "perturbation") that doesn't seem to tangle these ribbons, do the ribbons stay intact all the way to the bottom? Most physicists assumed the answer was a resounding "yes." This paper, however, tells a different story: sometimes, even if the nudge looks harmless at first glance, the ribbons actually snap.

The authors of this paper, Federico Ambrosino, Matthias R. Gaberdiel, and Yu Nakayama, decided to test this assumption using a very specific, mathematically clean playground: the $N=2$ minimal models. These are like simplified, two-dimensional universes where the rules of quantum physics are strict and solvable. They focused on a family of these magical ribbons (called "non-invertible defect lines") and asked what happens when they push the system with the weakest possible nudge that usually leads to a smooth, massless transition.

Their investigation revealed a surprising twist. When they tried to carry these ribbons through the flow, everything looked fine at the very first step. The ribbons seemed to survive the initial push. However, as they looked closer at the second step of the journey, they found a hidden snag. The ribbons couldn't be adjusted to fit the new shape of the river without breaking a fundamental rule of the universe: supersymmetry. It's as if the ribbon tried to stretch, but the fabric of space-time refused to let it, creating a "glitch" or an "anomaly" right on the ribbon itself.

The paper proves that for the standard, massless flow (where the river ends in a calm, massless lake), these ribbons cannot survive. The assumption that "if the nudge doesn't touch the ribbon, the ribbon survives" is false. The obstruction appears specifically at the second order of the deformation, meaning it's a subtle effect that only shows up when you look beyond the immediate, first reaction. The authors show this using two different methods: one based on the language of Conformal Field Theory (the study of the river's shape) and another using "Matrix Factorisations" (a way of breaking down the river's potential energy into algebraic blocks). Both methods agree: the ribbons break.

But here is the playful part of the story: the authors didn't just say "it's impossible." They asked, "What if we change the river itself?" They found that if you tweak the nudge slightly—adding a specific, higher-order correction to the push—you can save the ribbons. This new path leads to a completely different destination: a "massive" theory where the river doesn't end in a calm lake but in a landscape of $k+1$ distinct, heavy vacua (like a series of deep, isolated pools). Along this specific, "integrable" path, all the ribbons survive perfectly, keeping their complex relationships intact.

In short, the paper demonstrates that preserving these exotic symmetries is not automatic. It requires a very specific, finely tuned journey. If you take the "easy" route (the standard massless flow), the symmetries break down due to a hidden anomaly. If you take the "hard" route (the massive Chebyshev deformation), the symmetries are preserved, but the universe you end up in is fundamentally different. The authors have shown that the universe is more picky about its symmetries than we thought; you can't just assume they survive a flow, even if the flow looks friendly at first.

Technical Summary: $N=2$ RG Flows, Non-Invertible Symmetries, and Matrix Factorisations

Problem Statement
The paper investigates the fate of topological defect lines (specifically non-invertible symmetries) under renormalisation group (RG) flows in two-dimensional $N=2$ supersymmetric theories. The central puzzle arises from the standard expectation that if a topological defect line commutes with a perturbing operator in the ultraviolet (UV) theory, it should survive as a symmetry of the infrared (IR) fixed point.

The authors focus on the flow from the $k$ -th $N=2$ minimal model to the $(k-2)$ -nd minimal model, induced by the least relevant chiral perturbation of the superpotential $W = X^{k+2}$ . The perturbation is $W = X^{k+2} + \lambda X^k$ . The UV theory possesses a family of non-invertible topological defect lines forming the fusion category $so(3)_k$ (the integer-spin subring of the $su(2)_k$ fusion ring). Since the perturbing field commutes with all lines in $so(3)_k$ , one would expect this entire category to survive the flow. However, the IR theory is the $(k-2)$ -nd minimal model, which possesses a strictly smaller fusion category. The authors demonstrate that no subcategory of the IR theory is isomorphic to $so(3)_k$ (except for trivial cases), creating a contradiction between the commutation of the defect with the perturbation and the structure of the IR fixed point.

Methodology
The paper employs two complementary frameworks to analyze this obstruction:

Conformal Field Theory (CFT) Arguments: The authors use the Verlinde formula and the structure of the modular $S$ -matrix to analyze the fusion rings of the $k$ -th and $(k-2)$ -nd minimal models. They explicitly calculate quantum dimensions to show that the generators of the UV defect category cannot map to a consistent set of generators in the IR theory while preserving fusion rules.
Matrix Factorisations (Landau-Ginzburg Description): The $N=2$ minimal models are described by Landau-Ginzburg (LG) models with superpotentials. Topological B-type defects are represented by matrix factorisations of the superpotential difference $W(X) - W(Y)$ . The authors analyze the deformation of these matrix factorisations order-by-order in the coupling $\lambda$ . They investigate whether the matrices $E$ and $J$ (satisfying $EJ = W \cdot \mathbb{1}$ ) can be deformed to satisfy the deformed superpotential $W(\lambda) = X^{k+2} + \lambda X^k$ while maintaining the defect's topological nature.

Key Contributions and Results

The Second-Order Obstruction: The analysis reveals that while the defect lines can be deformed to first order in $\lambda$ (consistent with the commutation condition), they encounter an obstruction at second order. Specifically, for the strictly linear deformation $W = X^{k+2} + \lambda X^k$ , the condition for the existence of a second-order deformation term fails. The obstruction manifests as a non-vanishing class $[E_1 J_1]$ in the quotient ring $C[X, Y]/(E_0, J_0)$ , which cannot be removed by any choice of first-order counterterms.
Interpretation as an Anomaly: The authors interpret this obstruction as a supersymmetry (BRST) anomaly localised on the defect line. While the perturbing field commutes with the line (ensuring first-order survival), the "collision" of two insertions of the perturbing field on the line generates a term that violates the supersymmetry algebra ( $Q^2 = W$ ) unless a specific bulk counterterm is introduced. The linear deformation does not provide this counterterm.
Resolution via Massive Flows: The paper demonstrates that the obstruction can be resolved if the superpotential is allowed to receive higher-order corrections in $\lambda$ . By systematically cancelling the obstruction order-by-order, the superpotential is uniquely fixed to the Chebyshev (Dickson) deformation:
$W(X, \lambda) = \xi^d + \frac{a^d}{\xi^d}, \quad \text{where } X = \xi + \frac{a}{\xi}, \ a = -\frac{\lambda}{d}.$
This deformation corresponds to an integrable, massive RG flow. Unlike the massless flow to the $(k-2)$ -nd minimal model, this flow ends in a theory with $k+1$ massive vacua. Along this massive trajectory, the entire $so(3)_k$ fusion category is preserved to all orders, and the defects define symmetries of the massive IR theory.
Massless Flow Interpretation: For the strictly linear (massless) flow, the defect lines cannot be preserved as supersymmetric (B-type) defects. The authors suggest that the obstruction may be understood as the defect lines attempting to act on the additional massive vacua that appear in the IR (at $X \sim \pm i\sqrt{\lambda}$ ) alongside the massless sector. In the massless limit, the defect fails to be a pure CFT operator.

Significance
The paper challenges the assumption that a symmetry commuting with a perturbation is automatically preserved in the IR fixed point. It highlights two critical subtleties in RG flows involving non-invertible symmetries:

Defect Dressing: Carrying a defect through a flow requires dressing it with defect-localised counterterms. This dressing can fail due to localised anomalies (specifically supersymmetry anomalies in this context).
Massive Vacua: The IR theory may contain massive sectors that are not captured by the massless CFT analysis. The preservation of a defect may depend on its action on these massive vacua.

The authors conclude that while the symmetry-preserving deformation exists, it forces the theory onto a massive, integrable trajectory (the Chebyshev flow) rather than the expected massless flow to a smaller minimal model. This result provides a concrete counterexample to standard symmetry-based RG flow conjectures and elucidates the precise mechanism (second-order anomalies) by which such expectations can fail.

Technical Summary: N=2N=2N=2 RG Flows, Non-Invertible Symmetries, and Matrix Factorisations

More like this

Technical Summary: $N=2$ RG Flows, Non-Invertible Symmetries, and Matrix Factorisations