Yang-Lee edge singularity at O(d 4) from the functional renormalization group: revised analysis

arXiv:2611.14206 · hep-th · Submitted 2026-11-10 · 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: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Yang-Lee edge singularity at O(d 4) from the functional renormalization group".

Mira: Here's the gist of the paper. The paper determines the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group from page 1; this is important because it provides a non-perturbative approach to studying critical phenomena in complex magnetic fields from page 3.

Kai: Determining Critical Exponents The authors determine the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group retaining all operators at second and fourth order in the derivative expansion from page 1; at fourth order, they obtain specific values for η in d = 5, 4 and 3, which agree with the six-loop epsilon expansion from page 1.

Mira: In three dimensions, these results give σ = 0.076 and νc = 0.359 from page 1; they locate the zero of the edge exponent at dσ = 2.65(1) from page 2, which is consistent with the six-loop estimate from page 3.

Kai: Field Dependence and Convergence The convergence of the field expansion for the complex Yang–Lee fixed point depends crucially on its reference point from page 2; expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1.3 from page 5; this representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order from page 5.

Title and authors: Kai: So we're looking at this paper now called "Yang-Lee edge singularity at O(d four) from the functional renormalization group: revised analysis" <ref:2611.14206#pg1,4) from the functional renormalization group: revised analysis>. It sounds like they're refining some of the old work on how critical phenomena behave near these complex magnetic fields.

Mira: Yeah, it really is about taking that original approach and pushing it to fourth order in the derivative expansion. They're focusing on determining how those critical exponents behave as continuous functions of dimension d, which is a key piece for understanding these systems more broadly.

Lev: From an error correction standpoint, being able to get these exponents as continuous functions of dimension helps us map out where the system might actually be stable or unstable in different physical regimes.

Kai: Exactly. The original work was important, but this revision is trying to give us a much more complete picture by including all the operators up to fourth order in that derivative expansion.

The paper's summary: Mira: What they're doing here is using the functional renormalization group, or FRG, to find these critical exponents as continuous functions of dimension d. They are retaining every operator at second and fourth order in the derivative expansion, which is a big step for getting a more complete picture than what was done before.

Kai: That means they're not just looking at some simple approximations; they're keeping all the terms up to that fourth order, which is crucial when you're trying to get accurate numbers. They also get specific values for the exponent eta in dimensions d equals five, four, and three.

Lev: Getting those specific values is important because we need concrete targets to see how well our theories match up with what we can actually measure on real hardware.

Mira: Right, and they found that these fourth-order results for eta match the six-loop epsilon expansion results in dimensions five, four, and three. In three dimensions specifically, they get sigma equals zero point zero seven six and nu c equals zero point three five nine.

Kai: And they also locate the zero of the edge exponent at d sigma equals two point six five one with one uncertainty. That finding is consistent with the estimate from the six-loop calculation as well.

The paper's improvements: Lev: So what are these improvements, really? It sounds like they fixed some issues that were there before, especially concerning how we approach the complex Yang-Lee fixed point.

Mira: They addressed a big problem with the field expansion for this complex Yang-Lee fixed point by changing the reference point. They found that expansions about zero fail with their old regulator, but expanding around a running imaginary field value at which the mass vanishes converges all the way down to dimension d equals one point three.

Kai: That convergence down to one point three is significant because it allows them to preserve an exact relation between the thermal eigenvalue and the field dimension, which they call theta h plus theta t equals d, at every truncation order.

Lev: That preservation of that identity across all orders means they have a much more robust way to calculate things than just stopping at a certain point in the expansion.

Mira: And on top of that, they derived the flows for U, Z and W one two three by matching momentum dependence at a constant background field. They did this by doing angular averages over loop momentum analytically for any dimension d.

Conclusion: Kai: So to wrap up, this paper shows how these continuous functions of dimension work when you use the functional renormalization group at second and fourth order in the derivative expansion. The main thing is that they get specific values for eta in d equals five, four, and three that agree with the six-loop epsilon expansion.

Mira: And they also found the zero of the edge exponent at d sigma equals two point six five one, which checks out against those estimates too. Plus they have this new way to handle the field dependence by expanding around a different point that converges down to dimension one point three, preserving that key hyperscaling identity.

Lev: For me, what's really interesting is how they test the correction-to-scaling exponent omega equal negative theta three is positive for the leading correction to scaling, and those values match the six-loop constrained Padé estimates. That means we can test that truncation independently of eta.

Kai: It seems like this paper gives us a much better set of tools for studying these complex magnetic fields because it's more complete in its derivative expansion and has a convergent way to handle the field dependence.

Mira: It really does, especially since they are pushing the calculations to fourth order, and they have to be careful about truncation errors, which is something we talked about earlier. We should keep an eye on how this method performs when we try it on even more complex systems.

Lev: I think that's the path forward. Getting these results into a framework that can handle the complexity of real hardware constraints is what matters next, since these theoretical tools are getting better and better at describing these critical points.

Konrad Kockler, Jan M. Pawlowski, Vladimir V. Skokov

Institut für Theoretische Physik, University of Heidelberg · ExtreMe Matter Institute EMMI, GSI Helmholtzzentrum für Schwerionenforschung · Department of Physics, North Carolina State University

hep-th

Submitted: 2026-11-10

Updated: 2026-11-10

Comments: 16 pages, 8 figures, 2 tables

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 64/100

The gist: The paper determines the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group from page 1…

Key concepts

Functional Renormalization Group (FRG)
A non-perturbative technique used to study critical phenomena by systematically summing up contributions from different length scales. Here, it is applied to find the critical exponents of the Yang-Lee edge singularity in complex magnetic fields.
Yang-Lee Edge Singularity
This describes a specific type of phase transition occurring at the boundary of a region in parameter space, particularly relevant for systems with complex magnetic fields. The paper studies how its critical behavior changes as the dimension of the system varies.
Field Expansion Convergence
The method used to analyze the complex Yang-Lee fixed point is sensitive to its reference point. Expansions around zero field fail, but expanding around a specific running imaginary field value allows for convergence down to low dimensions, preserving important hyperscaling relations.

Terminology

Summary

The paper determines the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group from page 1; this is important because it provides a non-perturbative approach to studying critical phenomena in complex magnetic fields from page 3. p. 0

Where the paper says it

Pawlowski,1, 2 and Vladimir V. Skokov3, 2 1 Institut f¨ur Theoretische Physik, Universit¨at Heidelberg, Philosophenweg 16, 69120 Heidelberg, Germany 2ExtreMe Matter Institute EMMI, GSI Helmholtzzentrum f¨ur Schwerionenforschung, Planckstrasse 1, 64291 Darmstadt, Germany 3Department of Physics, North Carolina State University, Raleigh, NC 27695, USA (Dated: November 11, 2026) We determine the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group, retaining all operators at second and fourth order in the derivative expansion.

Determining Critical Exponents

The authors determine the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group retaining all operators at second and fourth order in the derivative expansion from page 1; at fourth order, they obtain specific values for η in d = 5, 4 and 3, which agree with the six-loop epsilon expansion from page 1. In three dimensions, these results give σ = 0. p. 0

Where the paper says it

575(9) in d = 5, 4 and 3, respectively, in agreement with the six-loop ϵ expansion. In three dimensions, these results give σ = 0.

p. 0
Where the paper says it

575(9) in d = 5, 4 and 3, respectively, in agreement with the six-loop ϵ expansion. In three dimensions, these results give σ = 0.

076 and νc = 0. p. 0
Where the paper says it

In three dimensions, these results give σ = 0. 076 and νc = 0.

p. 0
Where the paper says it

In three dimensions, these results give σ = 0. 076 and νc = 0.

359 from page 1; they locate the zero of the edge exponent at dσ = 2. p. 0
Where the paper says it

076 and νc = 0. We locate the zero of the edge exponent at dσ = 2.

p. 0
Where the paper says it

076 and νc = 0. We locate the zero of the edge exponent at dσ = 2.

65(1) from page 2, which is consistent with the six-loop estimate from page 3. p. 0
Where the paper says it

We locate the zero of the edge exponent at dσ = 2. 65(1), also consistent with the six-loop estimate.

p. 0
Where the paper says it

We locate the zero of the edge exponent at dσ = 2. 65(1), also consistent with the six-loop estimate.

Field Dependence and Convergence

The convergence of the field expansion for the complex Yang–Lee fixed point depends crucially on its reference point from page 2; expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1.3 from page 5; this representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order from page 5. p. 0

Where the paper says it

Expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1. This representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order.

p. 0
Where the paper says it

Expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1. This representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order.

p. 0
Where the paper says it

Expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1. This representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order.

Flow Equations Derivation

The authors derive the flows of U, Z and W1,2,3 by matching the momentum dependence of the vertex functions at a constant background field from page 4; they perform angular averages over the loop momentum analytically for arbitrary d from page 4. The flows are extracted by matching coefficients of the resulting momentum structures, and consistency checks verify that the field-derivative relations hold for all y from page 4. p. 3

Where the paper says it

addition, the flows of Z ′ , W′ 1 and the other field derivatives extracted from higher-point vertices must agree with the field derivatives of the flows obtained from lowerpoint vertices. We verify that the residuals of all redundant equations vanish identically and that the field-derivative relations hold for all y.

Results and Benchmarks

The fourth-order results for η are given as η = −0.153(7) in d = 5, −0. p. 0

Where the paper says it

At fourth order, we obtain η = −0. 153(7), −0.

352(9) in d = 4 and −0. p. 0
Where the paper says it

153(7), −0. 352(9) and −0.

575(9) in d = 3 from page 6; these agree with the six-loop epsilon-expansion results in d = 5, 4 and 3 from page 7. p. 0
Where the paper says it

575(9) in d = 5, 4 and 3, respectively, in agreement with the six-loop ϵ expansion. In three dimensions, these results give σ = 0.

The correction-to-scaling exponent ω is found to be ω = −θ3 > 0 for the leading correction to scaling from page 8; these values agree with the six-loop constrained-Pad´e estimates from page 8. p. 7
Where the paper says it

It agrees with the six-loop constrained-Pad´e estimates [19] and, in d = 3, with the fuzzy-sphere result [27]; as for η, the step from O(∂ 2 ) to O(∂ 4 ) moves it toward these values. In d = 2, they agree with the exact values of η, σ and νc, and the correction-to-scaling exponent ω = 1.

Uncertainty and Limitations

The main remaining limitation is the uncertainty associated with the derivative expansion itself from page 9; they estimate this uncertainty from the difference between second and fourth order from page 9. The paper also notes that omitting two of the three four-derivative operators changes the fourth-order correction by up to 50% from page 7; this is because the sign of the discrepancy depends on dimension from page 7. p. 7

Where the paper says it

Two features distinguish our work from earlier FRG studies: First, complete derivative expansions: We retain all operators at O(∂ 2 ) and O(∂ 4 ). Omitting two of the three four-derivative operators changes the fourth-order correction by up to 50%, and the sign of the discrepancy depends on dimension.

The field expansion converges more slowly as d decreases, and for d = 2, they find η = −0. p. 7
Where the paper says it

5034† −1 (exact) values becomes relevant for α = 3 and our calculations become unreliable. The field expansion converges more slowly as d decreases.

795(20) in agreement with the exact value −4/5 from page 8. p. 7
Where the paper says it

795(20) ± 0. 007 at O(∂ 4 ), in agreement with the exact value −4/5 [5].

Conclusion

The authors have determined the critical exponents of the Yang–Lee edge singularity using the functional renormalization group at second and fourth order in the derivative expansion from page 9; two features distinguish their work: complete derivative expansions and a convergent representation of the field dependence from page 8. Their fourth-order values of η agree with the six-loop epsilon-expansion results in d = 5, 4 and 3 from page 7; the correction-to-scaling exponent ω is not fixed by η and therefore tests the truncation independently from page 8. p. 7

Where the paper says it

Our fourth-order values of η agree, within the quoted uncertainties, with the six-loop ϵ-expansion results in d = 5, 4 and 3 [19]. The correction-to-scaling exponent, ω = 0.

Summary

The Yang–Lee edge singularity at O(d 4) from the functional renormalization group determines critical exponents as continuous functions of dimension, providing a non-perturbative approach to studying critical phenomena in complex magnetic fields from page 3.

Pruning redundant reasoning tokens

The authors determine the critical exponents of the Yang–Lee edge singularity as continuous functions of dimension using the functional renormalization group retaining all operators at second and fourth order in the derivative expansion from page 1; at fourth order, they obtain specific values for η in d = 5, 4 and 3, which agree with the six-loop epsilon expansion from page 7. In three dimensions, these results give σ = 0.076 and νc = 0.359 from page 1; they locate the zero of the edge exponent at dσ = 2.65(1) from page 2, which is consistent with the six-loop estimate from page 3.

How the training signal is built

The convergence of the field expansion for the complex Yang–Lee fixed point depends crucially on its reference point from page 5; expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1.3 from page 5; this representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order from page 5.

Where the gains come from

The authors derive flows of U, Z and W1,2,3 by matching momentum dependence of vertex functions at a constant background field from page 4; they perform angular averages over loop momentum analytically for arbitrary d from page 4. The flows are extracted by matching coefficients of resulting momentum structures, and consistency checks verify that the field-derivative relations hold for all y from page 4. p. 3

Where the paper says it

addition, the flows of Z ′ , W′ 1 and the other field derivatives extracted from higher-point vertices must agree with the field derivatives of the flows obtained from lowerpoint vertices. We verify that the residuals of all redundant equations vanish identically and that the field-derivative relations hold for all y.

p. 3
Where the paper says it

addition, the flows of Z ′ , W′ 1 and the other field derivatives extracted from higher-point vertices must agree with the field derivatives of the flows obtained from lowerpoint vertices. We verify that the residuals of all redundant equations vanish identically and that the field-derivative relations hold for all y.

The convergence of the field expansion for the complex Yang–Lee fixed point depends crucially on its reference point from page 5; expansions about ϕ = 0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d = 1.3 from page 5; this representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θh + θt = d, at every field-truncation order from page 5.

The authors derive flows of U, Z and W1,2,3 by matching momentum dependence of vertex functions at a constant background field from page 4; they perform angular averages over loop momentum analytically for arbitrary d from page 4. The flows are extracted by matching coefficients of resulting momentum structures, and consistency checks verify that the field-derivative relations hold for all y from page 4.

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 FRG analysis of Yang-Lee edge singularity by retaining all operators at second and fourth order in derivative expansion allows for a determination of critical exponents η, σ, and νc as continuous functions of dimension d.

  2. The improved system can determine the anomalous dimension η = -0.153(7) in d=5, -0.352(9) in d=4, and-0.575(9) in d=3 at fourth order, which is consistent with the six-loop epsilon expansion.

  3. The system can locate the zero of the edge exponent at dimension dσ = 2.65(1), which is consistent with the six-loop estimate, and it can determine η = -0.795(21) and ω = 1.90(6) in d=2 compared to exact values.

  4. The improved system can handle complex Yang-Lee fixed points by utilizing expansions about the running imaginary field value at which the mass vanishes, achieving convergence down to d = 1.3, thereby preserving the exact relation between the thermal eigenvalue and the field dimension, θh + θt = d.

  5. The improved system can detect hyperscaling violations reported in earlier works by distinguishing between truncation errors and expansion-point motion when linearizing the flow, as suggested by Section IX: The omission leaves the fixed-point solution unchanged, because ∂tχ = 0 at the fixed point. It does change the linearized flow: the stability matrix no longer has ∆ϕ as an exact eigenvalue.

Abstract

We determine the critical exponents of the Yang-Lee edge singularity as continuous functions of dimension using the functional renormalization group, retaining all operators at second and fourth order in the derivative expansion. At fourth order, we obtain η=-0.153(7), -0.352(9) and-0.575(9) in d=5, 4 and 3, respectively, in agreement with the six-loop ε expansion. In three dimensions, these results give σ=0.076 and ν c=0.359. We locate the zero of the edge exponent at d σ=2.65(1), also consistent with the six-loop estimate. Continuing the fixed point through d σ, we obtain η=-0.795(21) and ω=1.90(6) in d=2, compared with the exact values-4/5 and 2. For the complex Yang-Lee fixed point, convergence of the field expansion depends crucially on its reference point. Expansions about ϕ=0 fail to converge with the smooth regulator used here, whereas expansions about the running imaginary field value at which the mass vanishes converge down to d=1.3. This representation preserves the exact relation between the thermal eigenvalue and the field dimension, and hence the hyperscaling identity θ h+θ t=d, at every field-truncation order. We trace the large violations reported in earlier works to the omission of the expansion-point motion when linearizing the flow, rather than to the truncation itself.

Sources

Related papers