Gradient Flows and the Curvature of Theory Space

Andreas Stergiou; William H. Pannell

REVIEW 3 major objections 4 minor 3 cited by

Gradient Flows and the Curvature of Theory Space

T0 review · 3 major / 4 minor · reviewed 2026-08-08 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper computes the Ricci scalar of the multiscalar gradient-flow metric in d=4−ε, finds the leading curvature is fixed and nonzero, and identifies the flow potential and metric with F-tilde and the Zamolodchikov metric.

desk verdict Solid new Ricci scalar computation for the gradient-flow metric, but the 'theory space is curved' claim needs an unstated N-independence postulate and the F-tilde/Zamolodchikov identifications are conditional. read the letter →

arxiv 2502.06940 v2 pith:WJIO4EDC submitted 2025-02-10 hep-th cond-mat.stat-mechhep-ph

William H. Pannell , Andreas Stergiou This is my paper · ORCID

classification hep-thcond-mat.stat-mechhep-ph

keywords gradientflowrenormalisationgroupRicciscalartheoryspacemultiscalarfieldepsilonexpansionF-tildetheoremZamolodchikovmetric

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper studies the space of quartic couplings of $N$ massless scalar fields in $d=4-\varepsilon$ dimensions, equipped with the metric that makes renormalisation-group flow a gradient flow. It computes the Ricci scalar of that metric through the first two orders in the coupling and finds that gradient flow fixes the leading curvature to $R_0=\frac{5}{3456}N^7+O(N^6,\varepsilon N^7)$, so no theory-independent choice of the remaining freedom in the flow solution can make the space flat. The paper then shows that the flow potential can be chosen to agree at fixed points with the sphere free-energy quantity $\widetilde F$, and that the metric can be chosen to agree with the Zamolodchikov metric of nearly marginal quartic operators. If these identifications hold, the conjectured weak $\widetilde F$-theorem becomes a perturbative gradient-flow statement interpolating between the $d=4$ $a$-theorem and the $d=3$ $F$-theorem, and the space of theories is naturally curved rather than flat.

What carries the argument

The machinery is the gradient-flow equation $\partial_I A = G_{IJ}\beta^J$, where $I=(ijkl)$ is a generalised index and $G_{IJ}$ is a positive-definite symmetric tensor treated as a Riemannian metric on the space of couplings. A and G are expanded as perturbative series in the quartic tensor $\lambda_{ijkl}$, with coefficients fixed by solving the gradient-flow equations order by order; the Ricci scalar is then evaluated from $R=G^{IJ}(\Gamma^K_{IK}\Gamma^L_{JL}-\Gamma^K_{LI}\Gamma^L_{KJ}+\partial_K\Gamma^K_{IJ}-\partial_I\Gamma^K_{JK})$ using diagrammatic tensor contractions. Scheme changes act as diffeomorphisms on the couplings, and scheme invariance is used both to verify the result and to bootstrap the dependence of $R$ on the unfixed coefficient $g_1$. The physical identifications use the sphere free-energy quantity $\widetilde F=\sin(\pi d/2)\log Z_{S^d}$ and the Zamolodchikov norm of quartic operators on the sphere, whose one-loop mixing matrix is obtained from the renormalisation of $\phi^4$.

What would settle it

Compute $\widetilde F$ at a generic fixed point such as the hypercubic one, to $O(\varepsilon^5)$ in the MS scheme, and compare it with $\pi/288$ times $A$ evaluated with the coefficient choices of the paper; any mismatch at that point would refute the claimed equality at all fixed points. Alternatively, an $N$-independent assignment of the remaining metric coefficients that makes $R_0=0$ at order $N^7$ while preserving gradient flow would refute the curvature claim.

Watch

Extended reading notes

Core claim

The central claim is that the gradient-flow metric on multiscalar theory space is not flat and that its curvature is determined by physical data. After imposing gradient flow, the leading Ricci scalar is $R_0=\frac{5}{3456}N^7+O(N^6,\varepsilon N^7)$, independent of all unfixed coefficients, so there is no theory-independent choice of the free solution with $R_0=0$; at next order $R_1$ is proportional to $\lambda_{iijj}$, and since inequivalent fixed points differ in this invariant the curvature distinguishes them. The paper further claims that fixing $a_2'=-\frac{13}{24}+\frac{1}{12}g_1$ and $a_5^3=-\frac{1}{9}(3\pi^2-89)-g_1$, together with an overall rescaling by $\pi/288$, makes the potential $A$ equal to $\widetilde F$ at every fixed point, and that choosing $g_1=-6$ makes the metric coincide, up to a fixed rescaling, with the Zamolodchikov metric of quartic operators on the sphere. This recasts the conjectured $\widetilde F$-theorem as a perturbative theorem about gradient flow in $d=4-\varepsilon$.

Load-bearing premise

The load-bearing step is the claim that matching the potential to the sphere free-energy function at the O(N) and cubic fixed points fixes it at every fixed point; the paper infers this from the shared tensor structure of the vacuum diagrams, but no explicit check at a generic fixed point is given.

Editorial extensions

If this is right

The coupling space of N-scalar theories in d=4−ε carries an intrinsic, scheme-independent curvature; no field redefinition or choice of gradient-flow solution can flatten it.
The next-order Ricci scalar is proportional to $\lambda_{iijj}$, so it takes different values at inequivalent fixed points; the explicit O(N) versus hypercubic difference does not change sign at N=4, so it does not order those two flows.
With the specified coefficient choices, A becomes a perturbative strong-monotonic gradient potential whose fixed-point values equal $\widetilde F$, interpolating between the d=4 a-theorem and the d=3 F-theorem.
The same choices make the metric, up to a fixed rescaling, the Zamolodchikov metric of nearly marginal quartic operators, giving a d=4−ε analogue of the two-dimensional c-theorem setting.
Under these identifications the large-N leading Ricci scalar is positive, including its O(ε) correction, which the paper interprets as compactness of theory space at large N.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

A natural test, not performed in the paper, is to compute $\widetilde F$ at a generic fixed point such as the hypercubic one to $O(\varepsilon^5)$ and compare it with $\pi/288$ times A; the paper checks O(N) and cubic, but the generic case rests on a tensor-structure argument rather than an explicit computation.
The scheme-invariance bootstrap used here for R looks transferable: any metric with the same diagrammatic expansion will have its Ricci scalar partly forced by diffeomorphism covariance, so the method could constrain curvature computations in other sectors of theory space without brute-force tensor algebra.
If gradient flow holds beyond perturbation theory, the $N^7$ curvature term would be an exact invariant of the space of scalar theories, giving a geometric label for theory space that is invisible to beta-function computations alone.
For defect and boundary theories, where interaction tensors have fewer indices, the same Ricci-scalar computation should be substantially simpler, so the geometric programme could be tested there first.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Let me give you my read on Pannell–Stergiou. The genuine novelty is the explicit Ricci scalar computation for the gradient-flow metric in multiscalar phi^4 in d = 4 - epsilon. The leading term R0 = 5/3456 N^7 + ... with the scheme-independence check and the bootstrap from g1 to fix the g2^m coefficients is solid work. That part deserves credit and is the paper's real contribution.

The trouble starts with the interpretation. The paper claims the space of theories is curved because no 'theory-independent' choice of the unfixed coefficients can set R0 = 0. But 'theory-independent' is doing a lot of work. The gradient-flow equations do not require the arbitrary coefficients to be N-independent. For each fixed N they are just constants, and you can choose them as functions of N. My stress-test note gives an explicit example: set g1 = 0, a5^3 = 0, and choose g2^1(N) to cancel B; then R0 = 0 for every N >= 1, as an admissible solution of the gradient-flow problem. So the conclusion that gradient flow 'naturally induces' curvature relies on an unstated postulate, namely that the same coefficients must work for all N. That postulate is physically reasonable, but it is not gradient flow alone, and the paper doesn't flag it as an assumption.

The identifications of A with F-tilde and G with the Zamolodchikov metric are also conditional. The coefficients are fixed by hand to match the O(N) and cubic fixed points, and the metric matching needs an operator rescaling. The claim that this automatically works at every fixed point is argued by saying F-tilde differs only in the tensor structure lambda_ijkl multiplying the same vacuum diagrams. That is suggestive, not a proof; there could be additional scheme-dependent contributions at generic fixed points. So the 'F-tilde theorem extends to gradient flow' should be read as a conjecture backed by two examples, not as a theorem.

The compactness remark is an overreach, as the reader noted: a positive leading large-N Ricci scalar doesn't tell you the manifold is compact.

Net: the paper is worth engaging with. The Ricci scalar computation is a tangible new result, and the identifications are interesting proposals. I'd send it to a serious referee, with a request that the authors state the N-independence postulate and either prove or soften the all-fixed-point claim for F-tilde. It should not be desk-rejected.

Referee Report

3 major / 4 minor

Summary. The paper studies the geometry of the space of couplings for multiscalar φ^4 theories in d=4−ε, using the metric G_IJ and potential A of the gradient-flow equation ∂_I A = G_IJ β_J. The authors compute the leading and next-to-leading terms in the Ricci scalar, perform a scheme-independence check of the leading term, and then attempt to give physical meaning to A and G_IJ by matching them at fixed points to the interpolating quantity \tilde{F} of Giombi–Klebanov and to the Zamolodchikov metric of quartic operators. They conclude that the space of multiscalar theories is curved, with R_0 = (5/3456) N^7 + O(N^6, ε N^7) at large N, and that the \tilde{F}-theorem can be extended perturbatively to a gradient-flow theorem.

Significance. If the claims hold, the paper is a valuable step toward a geometric interpretation of gradient flow and a perturbative bridge between gradient flow and monotonicity theorems. The explicit computation of the Ricci scalar, including the diagrammatic method in Appendix C and the scheme-independence check of R_0, is a substantial technical contribution and appears reproducible. The matching of A with \tilde{F} at the O(N) and cubic fixed points is suggestive, as is the identification of the metric with a rescaled Zamolodchikov metric. However, the central curvature claim is conditional on an N-independence requirement on free coefficients that is not derived from the gradient-flow equations, and the matching claims rest on unproven assumptions about higher-loop tensor structures. These issues materially affect the strength of the conclusions.

major comments (3)

[§3, Eqs. (3.3)–(3.4)] The claim that no choice of free coefficients can make R_0 vanish is only true under the stated N-independence postulate. The counterexample above is a valid solution of the gradient-flow equations for each fixed N and shows that the O(N^7) leading term is not fixed by gradient flow alone.
[§4, Eqs. (4.6)–(4.11)] The matching is partly by construction: the free coefficients are chosen to reproduce known values of \tilde{F}. The independent content is that a single choice works at the two tested fixed points; the universal claim is an extrapolation.
[§5, Eqs. (5.27)–(5.32)] This is a load-bearing issue for the claim that the metric coincides with the Zamolodchikov metric, because the identification is enforced by the operator rescaling.

minor comments (4)

[§5, Eq. (5.30)] The notation "22∆α" is ambiguous; it should be typeset as 2^{2Δα} for clarity.
[§5, after Eq. (5.9)] The sentence "working in the MS scheme rather than the MS scheme" appears to contain a typo; presumably the intended contrast is between the MS-bar scheme and the MS scheme.
[§6] The statement that theory space is "compact at large N" is not supported by a perturbative computation of the Ricci scalar; positivity of the leading coefficient does not establish compactness of the full space.
[§3.2] The scheme-independence check is presented in detail for R_0, but for R_1 the authors rely on a bootstrap argument. It would be helpful to state explicitly to what extent the R_1 result has been independently verified.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the gradient-flow curvature computation is self-contained, and the matching of A and G to F-tilde and the Zamolodchikov metric is presented as an explicit choice of unfixed freedom, not as a prediction.

full rationale

The paper's central derivation is the perturbative solution of the gradient-flow equation (2.7) and the explicit computation of the Ricci scalar in (3.3)-(3.4). The displayed leading O(N^7) term in (3.4) is fixed once the unfixed coefficients are required to be N-independent, a condition the paper states explicitly in the same paragraph; the conclusion that no theory-independent choice can set R0=0 is therefore a direct consequence of the displayed formula, not of a fitted input. The identifications in Sections 4 and 5 are explicitly framed as choices: Eq. (4.6) fixes a2' and a5^3 in order to match the independently computed eF at the O(N) fixed point, and the cubic fixed point is then checked as a non-trivial consistency test; Eq. (5.29) fixes g1=-6 and Eq. (5.31) rescales the operators to match the Zamolodchikov two-point function. Because these are transparent 'there exists a choice' constructions rather than predictions derived from the matched data, they do not constitute circularity. The use of the authors' prior work [20,29] is motivational or data-supplying and is not needed to force the main result; the Ricci scalar computation is carried out in this paper from the gradient-flow solution through three and four loops. Hence no step reduces by construction to its own input.

Assumptions & free parameters 4 free parameters · 6 assumptions · 0 invented entities

The central curvature result relies on the gradient-flow solution and the completeness of the tensor basis; the F-tilde and Zamolodchikov identifications require fitting several coefficients to known data. No new particles, forces, or dimensions are introduced.

free parameters (4)

a2_prime = -13/24 + 1/12 g1
Fixed in (4.6) so that A matches F-tilde at the O(N) fixed point through O(ε^4).
a5_3 = -(3π^2 - 89)/9 - g1
Fixed in (4.6) to match F-tilde at O(ε^5).
g1 = -6
Fixed in (5.29) to match the metric to the Zamolodchikov two-point function; enters A and the Ricci scalar.
overall rescaling = π/288
Applied to A and G so that the potential matches the normalization of F-tilde; a normalization choice that affects all later expressions.

assumptions (6)

domain assumption A perturbative solution to ∂_I A = G_IJ β^J exists with positive-definite, symmetric G_IJ through the orders used.
Assumed in section 2 and used to compute R0 and R1 in section 3; the solution is taken from prior work [17-20].
domain assumption Scheme changes act as diffeomorphisms on the space of couplings.
Used in section 3.1, eqs (3.5)-(3.6), to establish scheme independence of the Ricci scalar; standard result from [26].
domain assumption The values of F-tilde at fixed points computed in [30] are correct.
Used in section 4 to fix a2' and a5^3 and to claim matching at all fixed points.
domain assumption The tensor structures appearing in the expansions of A and G form a complete basis at each order in λ.
Required for the claim in section 4 that matching at O(N) and cubic fixed points fixes A=F-tilde at all fixed points; not proved.
domain assumption Perturbative expansion in λ and ε is valid and the metrics considered are positive definite.
Throughout the paper; standard in perturbative RG.
ad hoc to paper The rescaled operators [O'] defined in (5.31)-(5.32) are the correct physical operators for identifying the metric with the Zamolodchikov metric.
The rescaling is introduced to force the matching in (5.30); it is a choice rather than derived.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Gradient Flows and the Curvature of Theory Space." pith.science (2026). https://pith.science/paper/WJIO4EDC

@misc{pith2026250206940,
  author       = {Pith},
  title        = {Pith review of: Gradient Flows and the Curvature of Theory Space},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WJIO4EDC}},
  note         = {Machine review of arXiv:2502.06940}
}

abstract

The metric and potential associated with the gradient property of renormalisation group flow in multiscalar models in $d=4-\varepsilon$ dimensions are studied. The metric is identified with the Zamolodchikov metric of nearly marginal operators on the sphere. An explicit form for the associated Ricci scalar in $d=4-\varepsilon$ is derived, which shows that the space of multiscalar field theories is curved. The potential is identified with a quantity $\widetilde{F}$ that was previously proposed as a weakly monotonic function interpolating between the $a$-theorem in four dimensions and the $F$-theorem in three dimensions. This implies that the $\widetilde{F}$-theorem can be extended perturbatively to a theorem about gradient flow in $d=4-\varepsilon$.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

Gradient RG Flow in Scalar-Fermion QFTs
hep-th 2025-11 conditional novelty 7.0 of 10

RG flow in scalar-fermion theories is gradient through four loops only when the 'beta shift' is included, via over a thousand scheme-independent constraints that hold wherever current data allow a check.
$\phi^6$ at $6$ (and some $8$) loops in $3d$
hep-th 2026-05 unverdicted novelty 6.0 of 10

Six-loop beta-function graphs for general ϕ⁶ theory in 3d are recalculated (differing from Hager, agreeing with recent work), with large-N eight-loop results, O(ε³) exponents, and gradient-flow linear relations.
Gradient properties of $\varphi^3$ in $d=6-\varepsilon$
hep-th 2025-07 conditional novelty 6.0 of 10

The d = 6 gradient (A-function) structure of the phi^3 RG flow extends to d = 6 − epsilon, requiring one new three-loop constraint, Eq. (22), that the MS coefficients satisfy.

Reference graph

Works this paper leans on

40 extracted references · 16 canonical work pages · cited by 3 Pith papers

[1]

Renormalization group and critical phenomena. 1. Renormalization group and the Kadanoff scaling picture

K. G. Wilson,“Renormalization group and critical phenomena. 1. Renormalization group and the Kadanoff scaling picture”, Phys. Rev. B4, 3174 (1971)

work page 1971
[2]

Irreversibility of the Flux of the Renormalization Group in a 2D Field Theory

A. B. Zamolodchikov,“Irreversibility of the Flux of the Renormalization Group in a 2D Field Theory”, JETP Lett.43, 730 (1986)

work page 1986
[3]

Seeing a c-theorem with holography

R. C. Myers & A. Sinha,“Seeing a c-theorem with holography”, Phys. Rev. D 82, 046006 (2010), arXiv:1006.1263 [hep-th]

arXiv 2010
[4]

Towards the F-Theorem: N=2 Field Theories on the Three-Sphere

D. L. Jafferis, I. R. Klebanov, S. S. Pufu & B. R. Safdi,“Towards the F-Theorem: N=2 Field Theories on the Three-Sphere”, JHEP 1106, 102 (2011), arXiv:1103.1181 [hep-th]

arXiv 2011
[5]

F-Theorem without Supersymmetry

I. R. Klebanov, S. S. Pufu & B. R. Safdi,“F-Theorem without Supersymmetry”, JHEP 1110, 038 (2011), arXiv:1105.4598 [hep-th]

arXiv 2011
[6]

On the RG running of the entanglement entropy of a circle

H. Casini & M. Huerta, “On the RG running of the entanglement entropy of a circle”, Phys. Rev. D85, 125016 (2012), arXiv:1202.5650 [hep-th]

arXiv 2012
[7]

Is There a c Theorem in Four-Dimensions?

J. L. Cardy,“Is There a c Theorem in Four-Dimensions?”, Phys. Lett. B215, 749 (1988)

work page 1988
[8]

Analogs for thec Theorem for Four-dimensional Renormalizable Field Theories

I. Jack & H. Osborn,“Analogs for thec Theorem for Four-dimensional Renormalizable Field Theories”, Nucl. Phys. B343, 647 (1990)

work page 1990

Show all 40 references

[9]

Weyl consistency conditions and a local renormalization group equation for general renormalizable field theories

H. Osborn, “Weyl consistency conditions and a local renormalization group equation for general renormalizable field theories”, Nucl. Phys. B363, 486 (1991)

1991
[10]

Constraints on RG Flow for Four Dimensional Quantum Field Theories

I. Jack & H. Osborn,“Constraints on RG Flow for Four Dimensional Quantum Field Theories”, Nucl. Phys. B883, 425 (2014), arXiv:1312.0428 [hep-th]

2014 arXiv
[11]

On Renormalization Group Flows in Four Dimensions

Z. Komargodski & A. Schwimmer,“On Renormalization Group Flows in Four Dimensions”, JHEP 1112, 099 (2011), arXiv:1107.3987 [hep-th]. 32

2011 arXiv
[12]

Averaged null energy and the renormalization group

T. Hartman & G. Mathys,“Averaged null energy and the renormalization group”, JHEP 2312, 139 (2023), arXiv:2309.14409 [hep-th]

2023 arXiv
[13]

The g-theorem and quantum information theory

H. Casini, I. Salazar Landea & G. Torroba,“The g-theorem and quantum information theory”, JHEP 1610, 140 (2016), arXiv:1607.00390 [hep-th]

2016 arXiv
[14]

Renormalization Group Flows on Line Defects

G. Cuomo, Z. Komargodski & A. Raviv-Moshe,“Renormalization Group Flows on Line Defects”, Phys. Rev. Lett.128, 021603 (2022), arXiv:2108.01117 [hep-th]

2022 arXiv
[15]

Constraint on Defect and Boundary Renormalization Group Flows

K. Jensen & A. O’Bannon,“Constraint on Defect and Boundary Renormalization Group Flows”, Phys. Rev. Lett.116, 091601 (2016), arXiv:1509.02160 [hep-th]

2016 arXiv
[16]

Gradient flows in three dimensions

I. Jack, D. R. T. Jones & C. Poole,“Gradient flows in three dimensions”, JHEP 1509, 061 (2015), arXiv:1505.05400 [hep-th]

2015 arXiv
[17]

Gradient Flow and the Renormalization Group

D. J. Wallace & R. K. P. Zia,“Gradient Flow and the Renormalization Group”, Phys. Lett. A 48, 325 (1974)

1974
[18]

Gradient Properties of the Renormalization Group Equations in Multicomponent Systems

D. J. Wallace & R. K. P. Zia,“Gradient Properties of the Renormalization Group Equations in Multicomponent Systems”, Annals Phys.92, 142 (1975)

1975
[19]

Scheme invariants in ϕ4 theory in four dimensions

I. Jack & C. Poole,“Scheme invariants in ϕ4 theory in four dimensions”, Phys. Rev. D98, 065011 (2018), arXiv:1806.08598 [hep-th]

2018 arXiv
[20]

Gradient properties of perturbative multiscalar RG flows to six loops

W. H. Pannell & A. Stergiou,“Gradient properties of perturbative multiscalar RG flows to six loops”, Phys. Lett. B853, 138701 (2024), arXiv:2402.17817 [hep-th]

2024 arXiv
[21]

The a-function in six dimensions

J. A. Gracey, I. Jack & C. Poole,“The a-function in six dimensions”, JHEP 1601, 174 (2016), arXiv:1507.02174 [hep-th]

2016 arXiv
[22]

Constraints on 3- and 4-loop β-functions in a general four- dimensional Quantum Field Theory

C. Poole & A. E. Thomsen, “Constraints on 3- and 4-loop β-functions in a general four- dimensional Quantum Field Theory”, JHEP 1909, 055 (2019), arXiv:1906.04625 [hep-th]

2019 arXiv
[23]

Stochastic Quantization on Two-dimensional Theory Space and Morse Theory

S. R. Das, G. Mandal & S. R. Wadia,“Stochastic Quantization on Two-dimensional Theory Space and Morse Theory”, Mod. Phys. Lett. A4, 745 (1989)

1989
[24]

Critical phenomena and renormalization group theory

A. Pelissetto & E. Vicari,“Critical phenomena and renormalization group theory”, Phys. Rept. 368, 549 (2002), cond-mat/0012164

2002 arXiv
[25]

Interpolating between a and F

S. Giombi & I. R. Klebanov, “Interpolating between a and F”, JHEP 1503, 117 (2015), arXiv:1409.1937 [hep-th]

2015 arXiv
[26]

Scheme Dependence and Multiple Couplings

I. Jack & H. Osborn, “Scheme Dependence and Multiple Couplings”, arXiv:1606.02571 [hep-th]

arXiv
[27]

xAct: Efficient Tensor Computer Algebra for Mathematica

J. Martín-García, “xAct: Efficient Tensor Computer Algebra for Mathematica” , http://www.xact.es. 33
[28]

Minimally subtracted six loop renormalization of O(n)-symmetric ϕ4 theory and critical exponents

M. V. Kompaniets & E. Panzer, “Minimally subtracted six loop renormalization of O(n)-symmetric ϕ4 theory and critical exponents” , Phys. Rev. D 96, 036016 (2017), arXiv:1705.06483 [hep-th]

2017 arXiv
[29]

Heavy handed quest for fixed points in multiple coupling scalar theories in theϵ expansion

H. Osborn & A. Stergiou, “Heavy handed quest for fixed points in multiple coupling scalar theories in theϵ expansion”, JHEP 2104, 128 (2021), arXiv:2010.15915 [hep-th]

2021 arXiv
[30]

Generalized F-Theorem and the ϵ Expansion

L. Fei, S. Giombi, I. R. Klebanov & G. Tarnopolsky, “Generalized F-Theorem and the ϵ Expansion”, JHEP 1512, 155 (2015), arXiv:1507.01960 [hep-th]

2015 arXiv
[31]

Critical properties ofϕ4 theories

H. Kleinert & V. Schulte-Frohlinde,“Critical properties ofϕ4 theories”, World Scientific (2001)

2001
[32]

Dimensional Regularization of Massless Theories in Spherical Space-Time

I. T. Drummond,“Dimensional Regularization of Massless Theories in Spherical Space-Time”, Nucl. Phys. B94, 115 (1975)

1975
[33]

Bounds on multiscalar CFTs in theϵ expansion

M. Hogervorst & C. Toldo,“Bounds on multiscalar CFTs in theϵ expansion”, JHEP2104, 068 (2021), arXiv:2010.16222 [hep-th]

2021 arXiv
[34]

General gauge-Yukawa-quarticβ-functions at 4-3-2-loop order

J. Davies, F. Herren & A. E. Thomsen,“General gauge-Yukawa-quarticβ-functions at 4-3-2-loop order”, JHEP 2201, 051 (2022), arXiv:2110.05496 [hep-ph]

2022 arXiv
[35]

Defects in scalar field theories, RG flows and dimensional disentangling

D. Rodriguez-Gomez & J. G. Russo,“Defects in scalar field theories, RG flows and dimensional disentangling”, JHEP 2211, 167 (2022), arXiv:2209.00663 [hep-th]

2022 arXiv
[36]

RG flows and stability in defect field theories

I. Carreño Bolla, D. Rodriguez-Gomez & J. G. Russo,“RG flows and stability in defect field theories”, JHEP 2305, 105 (2023), arXiv:2303.01935 [hep-th]

2023 arXiv
[37]

Defects, rigid holography, and C- theorems

I. Carreño Bolla, D. Rodriguez-Gomez & J. G. Russo, “Defects, rigid holography, and C- theorems”, Phys. Rev. D108, L041701 (2023), arXiv:2306.11796 [hep-th]

2023 arXiv
[38]

Line defect RG flows in theε expansion

W. H. Pannell & A. Stergiou,“Line defect RG flows in theε expansion”, JHEP 2306, 186 (2023), arXiv:2302.14069 [hep-th]

2023 arXiv
[39]

A note on defect stability in d = 4 − ε

W. H. Pannell, “A note on defect stability in d = 4 − ε”, JHEP 2412, 187 (2024), arXiv:2408.15315 [hep-th]

2024 arXiv
[40]

Multiscalar critical models with localised cubic interactions

S. Harribey, W. H. Pannell & A. Stergiou,“Multiscalar critical models with localised cubic interactions”, JHEP 2412, 092 (2024), arXiv:2407.20326 [hep-th]. 34

2024 arXiv

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.

[1] [1]

Renormalization group and critical phenomena. 1. Renormalization group and the Kadanoff scaling picture

K. G. Wilson,“Renormalization group and critical phenomena. 1. Renormalization group and the Kadanoff scaling picture”, Phys. Rev. B4, 3174 (1971)

work page 1971

[2] [2]

Irreversibility of the Flux of the Renormalization Group in a 2D Field Theory

A. B. Zamolodchikov,“Irreversibility of the Flux of the Renormalization Group in a 2D Field Theory”, JETP Lett.43, 730 (1986)

work page 1986

[3] [3]

Seeing a c-theorem with holography

R. C. Myers & A. Sinha,“Seeing a c-theorem with holography”, Phys. Rev. D 82, 046006 (2010), arXiv:1006.1263 [hep-th]

arXiv 2010

[4] [4]

Towards the F-Theorem: N=2 Field Theories on the Three-Sphere

D. L. Jafferis, I. R. Klebanov, S. S. Pufu & B. R. Safdi,“Towards the F-Theorem: N=2 Field Theories on the Three-Sphere”, JHEP 1106, 102 (2011), arXiv:1103.1181 [hep-th]

arXiv 2011

[5] [5]

F-Theorem without Supersymmetry

I. R. Klebanov, S. S. Pufu & B. R. Safdi,“F-Theorem without Supersymmetry”, JHEP 1110, 038 (2011), arXiv:1105.4598 [hep-th]

arXiv 2011

[6] [6]

On the RG running of the entanglement entropy of a circle

H. Casini & M. Huerta, “On the RG running of the entanglement entropy of a circle”, Phys. Rev. D85, 125016 (2012), arXiv:1202.5650 [hep-th]

arXiv 2012

[7] [7]

Is There a c Theorem in Four-Dimensions?

J. L. Cardy,“Is There a c Theorem in Four-Dimensions?”, Phys. Lett. B215, 749 (1988)

work page 1988

[8] [8]

Analogs for thec Theorem for Four-dimensional Renormalizable Field Theories

I. Jack & H. Osborn,“Analogs for thec Theorem for Four-dimensional Renormalizable Field Theories”, Nucl. Phys. B343, 647 (1990)

work page 1990