Local subtractions and renormalization-group monotonicity: exact defect and sphere flows

arXiv:2608.02447 · hep-th · Submitted 2026-08-03 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: High Energy Physics: Theory. The day in high-energy theory: gravity, strings, fields and the questions left open.

Kai: We're looking at the paper: "Local subtractions and renormalization-group monotonicity".

Mira: Here's the gist of the paper. Local subtractions and renormalization-group monotonicity: exact defect and sphere flows is a study on when subtracting local counterterms in a quantum field theory preserves monotonicity along the renormalization-group flow, which matters for understanding RG behavior in various physical systems The authors investigate conditions under which the sphere free energy decreases monotonically after subtracting local terms fixed by a renormalization prescription.

Kai: Defining Admissible Subtractions and Obstructions The paper defines admissible subtractions as polynomials in the scale derivative D that remove local terms proportional to R3 and R while preserving fixed-point values, with the normalization reproducing the fixed-point constant.

Mira: It establishes an infrared obstruction: if a subtracted free energy approaches an ordered IR endpoint from below under decay assumptions, every admissible subtraction is nonmonotone.

Kai: This obstruction arises because the bilateral Laplace transform of the scale derivative has a real zero, which is preserved by every finite polynomial in D.

Title and authors: Kai: So we're looking at this paper called "Local subtractions and renormalization-group monotonicity: exact defect and sphere flows." It basically asks a question about when you can take local terms out of a quantum field theory without breaking the flow along the renormalization group path.

Mira: Right, so it’s about those counterterms that fix certain zeros in the subtraction polynomial, but they don't automatically guarantee that the free energy keeps decreasing monotonically as you change your scale >

Lev: From an engineering standpoint, it means we need to be really precise about what we subtract; if we pick the wrong local terms, our scaling behavior goes haywire on real hardware >

Kai: Exactly. The authors investigate when subtracting those local terms preserves monotonicity along the renormalization group flow. They look at specific conditions under which this happens in various physical systems >

Mira: It's interesting because they find that just having positive spectral data isn't enough to guarantee that the free energy stays monotone >

Lev: So what does this mean for modeling? Does it mean we have to be more restrictive about what local terms we subtract before we even start designing an experiment?

Kai: It means you need these specific spectral conditions, like complete monotonicity of the heat trace for some systems, because that’s what actually guarantees the flow stays monotone >

Mira: They've got this classification system for different types of flows—like Robin boundary flows and massive Dirac fields—to show exactly where those sufficient conditions apply >

Lev: So it’s not just a general theory; it’s showing us specific physical systems where we can be confident about the renormalization group behavior under these subtractions >

The paper's summary: Kai: Okay, so the core of "Local subtractions and renormalization-group monotonicity: exact defect and sphere flows" is that they are investigating when subtracting local counterterms doesn't mess up the flow along the renormalization group path. They show that positivity of spectral data alone isn't enough to guarantee that monotonicity >

Mira: That’s a key point; if you just look at the numbers, it doesn't tell you the whole story; you need more information about how those spectral functions behave under derivatives >

Lev: What does this change for someone who is trying to model a physical system? Does it mean we have to be more restrictive about what we subtract before we even start designing an experiment?

Mira: And they've got this classification system for different types of flows—like Robin boundary flows and massive Dirac fields—to show exactly where those sufficient conditions apply >

Kai: Exactly, they've given us concrete rules for when to subtract local terms without losing that monotonicity, and they've mapped out exactly which geometries work >

Mira: It’s a very helpful classification tool because it separates the cases where subtraction works from those where it fails due to the infrared obstruction >

Lev: So we're moving from just knowing what happens to being able to predict *if* a specific physical setup will maintain stability under scale changes >

The paper's improvements: Lev: Beyond just stating what they found, what are the actual improvements suggested by this work that we should take away? What’s the practical application there?

Kai: They suggest a way to improve our understanding of nonmonotonicity at large N by incorporating conditions one through three for the critical O N model >

Mira: That means we can conditionally predict when every single admissible subtraction is nonmonotone if those specific spectral inequalities are not met >

Lev: So it’s like adding a predictive layer to error correction; if we meet those conditions, we know the flow will be good enough for our hardware >

Kai: Also, they suggest that the endpoint ordering holds for every alpha in AdS space, but running monotonicity is restricted to specific ranges >

Mira: So while the endpoint behavior is universal across AdS space, the actual running monotonicity depends on those specific constraints we just discussed >

Lev: This helps us narrow down the parameter space for our theoretical models; we know exactly where we need to focus our attention when dealing with large N limits >

Conclusion: Kai: So to wrap up on this paper, "Local subtractions and renormalization-group monotonicity: exact defect and sphere flows," it’s about when local subtraction works in RG flow.

Mira: The main thing is that positivity of spectral data isn't enough; you need specific spectral conditions for the flow to be monotone >

Lev: For me, this means we have to be extremely cautious about how we handle those local terms when scaling up to large N, because the theory seems to push us towards needing more than just positivity of spectral data >

Lev: And I think this paper highlights that when we look at these local subtractions and renormalization-group monotonicity, we see that dynamics supply spectral inequalities in broader classes of flows are still an open problem >

Kai: Yeah, the final word is that a monotone admissible subtraction exists if and only if some specific conditions hold for the flow to be good >

Mira: And I think this paper sets a clear standard for what's possible in these types of flows, giving us a solid baseline to measure against moving forward >

Lev: For me, this means we have to keep asking questions about those spectral inequalities because it seems like dynamics supply them in broader classes of flows is still an open problem >

Francesco Scardino

Department of Physics, School of Mathematics and Natural Sciences, Bergische Universität Wuppertal

hep-th

Submitted: 2026-08-03

Updated: 2026-11-03

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

Importance score: 72/100

The gist: Local subtractions and renormalization-group monotonicity: exact defect and sphere flows is a study on when subtracting local counterterms in a quantum field theory preserves monotonicity along the…

Key concepts

Admissible Subtractions
These are specific polynomial subtractions of local terms proportional to R3 and R that maintain fixed-point values. They must satisfy a normalization condition related to the fixed-point constant. If these subtractions lead to a free energy approaching an ordered infrared endpoint from below, they are nonmonotone.
Infrared Obstruction
This obstruction occurs when the bilateral Laplace transform of the scale derivative has a real zero that is preserved by every admissible subtraction. This mathematical feature ensures that if the subtracted free energy decays toward an ordered IR endpoint from below under certain decay assumptions, monotonicity cannot be guaranteed for any admissible subtraction.
Stress-Tensor Formula
This formula relates the leading infrared sphere coefficient of a flow to a difference of non-negative stress-tensor spectral moments. For massive critical O(N) models, this moment difference vanishes at leading order and becomes negative at order N₀, providing a key tool to determine the sign and behavior of the flow in the infrared phase.

Terminology

Summary

Local subtractions and renormalization-group monotonicity: exact defect and sphere flows is a study on when subtracting local counterterms in a quantum field theory preserves monotonicity along the renormalization-group flow, which matters for understanding RG behavior in various physical systems The authors investigate conditions under which the sphere free energy decreases monotonically after subtracting local terms fixed by a renormalization prescription. p. 0

Defining Admissible Subtractions and Obstructions

The paper defines admissible subtractions as polynomials in the scale derivative D that remove local terms proportional to R3 and R while preserving fixed-point values, with the normalization reproducing the fixed-point constant. It establishes an infrared obstruction: if a subtracted free energy approaches an ordered IR endpoint from below under decay assumptions, every admissible subtraction is nonmonotone. p. 1 This obstruction arises because the bilateral Laplace transform of the scale derivative has a real zero, which is preserved by every finite polynomial in D. p. 1

Sufficient Conditions for Monotonicity

The authors derive sufficient spectral conditions for monotonicity based on the sign of derivatives of spectral functions. For Gaussian and leading large-N double-trace determinants on S p, complete monotonicity of the rescaled integrated heat trace is sufficient for monotonicity, which holds for Robin flows on every sphere dimension p ≥ 2. p. 0 A related differential identity proves monotonicity for massive Dirac fields on every odd sphere of dimension d ≥ 3. p. 1 For double-trace families driven by O2 on spherical surface defects and leading order on S3, a monotone subtraction exists exactly when the dimension of O is at least 1/2 and (3 − p37/13)/2, respectively. p. 0

Classification of Surface-Defect Flows

The paper classifies generalized-free scalar operators on spherical surface defects, finding that a monotone admissible subtraction exists if and only if 1/2 ≤ ∆b < 1, realized exactly by a four-dimensional monodromy defect. For a scalar operator O of dimension 1/2 < ∆ < 3/2 in a CFT with large-N factorization deformed by O2, monotonicity exists on S3 if and only if ∆ ≥ ∆∗ = (3 − p37/13)/2. p. 1

Stress-Tensor Formula and Infrared Behavior

The leading infrared sphere coefficient of a flow from a three-dimensional conformal field theory to a gapped phase is proportional to a difference of nonnegative stress-tensor spectral moments. For the massive critical O(N) model, this moment difference vanishes at leading order and is negative at order N0. p. 1 This result is verified by an independent calculation using flat-space stress-tensor diagrams. p. 2

Control of the Minimum Near Threshold

The paper analyzes the onset of monotonicity near thresholds, showing that the minimum of the running function Sb moves to a scale Xε proportional to e log(15/8)/(2δ) as δ = ε − 1 ↓ 0. This implies that for sufficiently small δ > 0, there is exactly one stationary point on the positive coupling axis, which is the global minimum and obeys CIR − Cε(ymin) ∼ ν2a3⁄2∗3√3b∗δ3/2.

Comparison with Holographic Flows

The results are compared with holographic F-functions of Ref. [4], showing that the dividing line for monotonicity in the double-trace family is fixed by the sign of the IR coefficient, through a factor 37 − 13ε2 in Eq. p. 7 (38). The comparison also suggests that nonmonotonicity at large finite N is conditional on spectral inequalities, which are not proven for the continuum critical O(N) model. p. 2

Physical Source and Local Terms

The allowed source counterterms and the finite-N scaling of the source do not change c−1, as local contributions proportional to mjRj are cancelled by terms in Eq. (63). The result is that for sufficiently large finite N, BN < 0, which invokes the criterion of Sec. p. 0 II C and excludes every finite polynomial subtraction. p. 0

Spectral Zeta Function and Endpoints

The analysis of the spectral zeta function shows that the pole at w = -1/ε produces a term proportional to A(ε)y in DCε, which fixes the sign of the IR coefficient. The endpoint difference CUV − CIR is shown to be positive, consistent with standard double-trace endpoint formulas. p. 6

Infrared Expansion and Minimum Location

The infrared expansion near the threshold is given by a combination of terms A(ε)y, B(ε)y3, and L(ε)y2ε, which determines the position and depth of the minimum near the threshold. The location of this minimum is found to be ymin ∼ r3b∗a∗δ−1/2. p. 7

Exact Flat-Space Stress-Tensor Calculation

The exact flat-space result for the moment difference is derived as a position-space sum rule, yielding m(M2 − 3M0)(1) = I0 − 4π2/75π3 <0, which agrees with the general formula. This calculation provides an independent verification of the stress-tensor relation in the same prescription. p. 9

Conclusion on Monotonicity

Local subtraction can produce an RG monotone only when the spectrum supplies sign information beyond positivity, and this holds for Robin boundary flows in every dimension and massive Dirac fields on every odd sphere. The loss of monotonicity is driven by the infrared coefficient having the wrong sign, where no admissible subtraction works. p. 6 The authors conclude that a monotone admissible subtraction exists if and only if 1/2 ≤ ∆b < 1 for surface defects, and ∆ ≥ ∆∗ for double-trace flows on S3. p. 0 The paper also notes that the endpoint ordering holds for every α in AdS space, but running monotonicity is restricted to specific ranges. p. 7 The final problem remains whether dynamics supplies the required spectral inequalities in broader classes of flows. p. 8 No other information is provided for this conclusion on page 31. p. 0

--- Page 31 ---

The paper investigates when subtracting local counterterms in a quantum field theory preserves monotonicity along the renormalization-group flow, which matters for understanding RG behavior in various physical systems The authors investigate conditions under which subtracting local counterterms in a quantum field theory preserves monotonicity along the renormalization-group flow, which matters for understanding RG behavior in various physical systems. p. 0

The paper investigates when subtracting local counterterms in a quantum field theory preserves monotonicity along the renormalization-group flow, which matters for understanding RG behavior in various physical systems The authors investigate conditions under which subtracting local counterterms in a quantum field theory preserves monotonicity along the renormalization-group flow, which matters for understanding RG behavior in various physical systems.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved ability to distinguish between sufficient and necessary conditions for RG monotonicity in different field theories by explicitly mapping spectral data to required sign information beyond mere positivity. The AI can now determine when positivity of the spectral data alone does not determine monotonicity versus when it is sufficient, referencing specific theorems like Bernstein's theorem.

  2. Enhanced capability to classify flows on various geometric backgrounds and defects by applying exact classifications derived in Sections IV and V for surface-defect and three-sphere double-trace flows, linking them to specific dimension thresholds like Delta star.

  3. Improved precision in calculating the infrared coefficient of the subtracted sphere free energy by using the stress-tensor formula derived from flat-space data, allowing accurate determination of whether a flow approaches its IR value from below or above based on sign differences in spectral moments.

  4. Increased robustness in analyzing nonmonotonicity at large N by incorporating conditions (i) through (iii) for the critical O(N) model, allowing the system to conditionally predict when every admissible subtraction is nonmonotone.

  5. Stronger capacity to analyze the onset of the minimum near thresholds by using discrete summation identities and rational coefficient inequalities to prove strict monotonicity for specific parameter ranges like 0 < epsilon <= 13/10.

Sources

Related papers