Constrained Pad\'e Ensembles for Thermal $\mathcal{N}{=}4$ SYM with the Exact $\mathcal O(\lambda^{5/2})$ Coefficient

Qianqian Du; Ubaid Tantary

arxiv: 2604.16109 · v1 · submitted 2026-04-17 · ✦ hep-th

Constrained Pad\'e Ensembles for Thermal mathcal{N}{=}4 SYM with the Exact mathcal O(λ^(5/2)) Coefficient

Ubaid Tantary , Qianqian Du This is my paper

Pith reviewed 2026-05-10 07:26 UTC · model grok-4.3

classification ✦ hep-th

keywords N=4 SYMthermal SYMPadé approximantsweak-coupling expansionstrong-coupling expansionthermodynamicsconstrained ensemblessupersymmetric gauge theory

0 comments

The pith

Exact O(λ^{5/2}) coefficient collapses the constrained LSTP ensemble for thermal N=4 SYM to a single curve

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

The paper updates the constrained log-subtracted two-point Padé ensemble used to interpolate the thermodynamics of thermal N=4 supersymmetric Yang-Mills theory in four dimensions. It incorporates the newly exact weak-coupling coefficient at order λ^{5/2} and shifts the weak-side matching points into the regime where this term is numerically important. The admissible set of approximants then reduces from nine nominal survivors on three distinct curves to one unique curve, the crossover region shrinks to a single value, and the pointwise uncertainty band vanishes within numerical resolution. The resulting LSTP survivor still differs from the central curve of the Hermite-Padé ensemble, so the new coefficient removes LSTP scan uncertainty but leaves the method dependence intact.

Core claim

By upgrading the weak-coupling truncation to the exact O(λ^{5/2}) coefficient while retaining the same interpolation ansatz and shifting the weak-side matching points, the admissible set of constrained LSTP approximants collapses to a single distinct curve, the crossover range becomes unique, and the pointwise band width drops to zero within numerical resolution. The Hermite-Padé central curve does not coincide with this unique LSTP survivor, removing the LSTP scan uncertainty but preserving the difference between the two routes.

What carries the argument

The constrained log-subtracted two-point Padé (LSTP) ensemble, which builds interpolating functions from weak- and strong-coupling series subject to matching constraints on both sides

If this is right

The LSTP route now supplies a unique thermodynamic prediction without residual scan uncertainty.
The discrepancy between LSTP and Hermite-Padé central curves survives, showing that the exact coefficient does not eliminate all method dependence.
The next calculational step required is the unknown O(λ^{-3}) strong-coupling coefficient to impose further constraints.
Pointwise uncertainty in the ensemble is eliminated to within the numerical resolution of the calculation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Higher-order terms in either expansion can be used systematically to discriminate among different resummation schemes in gauge-theory thermodynamics.
The remaining LSTP-HP difference suggests that the two constructions are sensitive to different features of the underlying analytic structure.
The same constrained-ensemble technique could be applied to other thermodynamic observables such as entropy density or speed of sound to check internal consistency.

Load-bearing premise

The original interpolation ansatz remains valid after the weak-side matching points are moved into the regime where the new O(λ^{5/2}) term is numerically significant, and the chosen constraints are sufficient to force uniqueness without introducing artifacts.

What would settle it

A direct high-precision evaluation of the energy density or pressure at an intermediate 't Hooft coupling near the former crossover region that lies outside the single LSTP survivor curve would falsify the collapse of the admissible set.

Figures

Figures reproduced from arXiv: 2604.16109 by Qianqian Du, Ubaid Tantary.

**Figure 2.** Figure 2: FIG. 2. Individual LSTP survivor curves before and after imposing the exact [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: FIG. 3. Curvature diagnostic [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

read the original abstract

We revisit the constrained log-subtracted two-point Pad\'e (LSTP) ensemble for thermal $\mathcal{N}=4$ supersymmetric Yang--Mills (SYM) thermodynamics in four spacetime dimensions after upgrading the weak-coupling truncation from $\mathcal{O}(\lambda^2)$ to the exact $\mathcal{O}(\lambda^{5/2})$ coefficient. We keep the interpolation ansatz unchanged and shift the weak-side matching points to the regime where the new term is numerically significant. The admissible set collapses from $9$ nominal survivors ($3$ distinct curves) to a single distinct curve, the crossover range shrinks to a unique value, and the pointwise band width drops to zero within numerical resolution. The Hermite-Pad\'e (HP) central curve does not coincide with the unique LSTP survivor, so the exact weak-coupling coefficient removes the LSTP scan uncertainty but not the difference between the two routes. The next step is to compute the unknown $\mathcal{O}(\lambda^{-3})$ strong-coupling coefficient.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Adding the exact O(λ^{5/2}) term and shifting the weak-side points collapses the LSTP ensemble to one curve, but the result still differs from the Hermite-Padé central curve.

read the letter

The main thing to know is that including the exact O(λ^{5/2}) weak-coupling coefficient and moving the matching points into the regime where that term matters reduces the constrained LSTP admissible set from nine survivors across three curves down to a single distinct curve, with the pointwise band width dropping to zero and the crossover range becoming unique. The paper keeps the original interpolation ansatz fixed and simply upgrades the input truncation plus the point locations. This produces a concrete, unique interpolation for the thermodynamics of thermal N=4 SYM that was not available before. It also records that this unique LSTP survivor does not coincide with the central curve obtained from the Hermite-Padé route, so the two methods still disagree even after the improvement. That is the useful output: a tighter practical curve between weak and strong coupling for this benchmark theory. The calculation is straightforward and the numerical outcome is reported clearly. The paper does not claim the result is independent of the method; it flags the remaining difference with HP. The soft spot is that the collapse is tied to the specific choice of matching points and the fixed ansatz. Shifting the points is a reasonable response to the new coefficient, but it assumes the original functional form remains adequate once those points sit where the λ^{5/2} term is numerically important. The paper does not test how the single survivor changes under small variations in the constraints or points, so the robustness of the uniqueness is not fully mapped. The circularity burden noted in the stress test is real but modest: the constraints are author-chosen, yet the paper is transparent about keeping the ansatz unchanged and about the non-coincidence with HP. This work is for people in the AdS/CFT and finite-temperature field-theory communities who need a reduced-uncertainty interpolation for N=4 SYM thermodynamics as a benchmark. A reader who wants an explicit, single-curve result from the LSTP route will get value from it. I would send it to peer review. The update is concrete, the central claim is numerically checkable, and the remaining method dependence is stated plainly rather than hidden.

Referee Report

3 major / 2 minor

Summary. The manuscript revisits the constrained log-subtracted two-point Padé (LSTP) ensemble for thermal N=4 SYM thermodynamics after upgrading the weak-coupling series to include the exact O(λ^{5/2}) coefficient. Keeping the interpolation ansatz fixed and shifting the weak-side matching points into the regime where the new term matters, the authors report that the admissible set collapses from 9 nominal survivors (3 distinct curves) to a single curve, the crossover range becomes unique, and the pointwise band width drops to zero within numerical resolution. The unique LSTP survivor does not coincide with the Hermite-Padé central curve. The next step proposed is computation of the unknown O(λ^{-3}) strong-coupling coefficient.

Significance. If the collapse is robust, the work shows that an exact higher-order weak-coupling coefficient can eliminate scan uncertainty in constrained Padé ensembles for N=4 SYM, yielding a sharply constrained interpolation between weak and strong coupling. This provides a concrete illustration of how perturbative input reduces theoretical uncertainty in gauge-theory thermodynamics and highlights persistent method dependence between LSTP and HP routes. The approach is reproducible in principle once matching-point values and numerical details are supplied.

major comments (3)

[Abstract and results section] Abstract and results section: the central claim of collapse to a single curve with zero band width is load-bearing, yet the manuscript provides neither the explicit numerical values of the shifted weak-side matching points nor tables or supplementary data showing the 9-to-1 reduction and verification that the O(λ^{5/2}) term was inserted correctly. Without these, it is impossible to exclude post-hoc fitting artifacts or constraint-induced uniqueness.
[Methods/ansatz section] Methods/ansatz section: the uniqueness result presupposes that the original LSTP interpolation ansatz remains a faithful representation once the weak-side matching points are moved into the O(λ^{5/2})-significant regime. No explicit consistency check or justification is supplied that the fixed ansatz does not become internally inconsistent with the higher-order data, raising the possibility that the observed collapse is an artifact of the constraint implementation rather than a genuine reduction in uncertainty.
[Comparison paragraph] Comparison paragraph: the reported non-coincidence between the unique LSTP survivor and the Hermite-Padé central curve is noted but not quantified (e.g., via pointwise difference or integrated discrepancy). This difference is important for assessing route dependence; the manuscript does not test whether additional terms would resolve or amplify it.

minor comments (2)

[Notation] Notation: the definition of the log-subtracted two-point Padé approximants should be restated explicitly when the matching points are shifted, to avoid ambiguity in the constraint equations.
[References] References: ensure the original LSTP ensemble paper is cited when describing the admissible-set construction, for reader context.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the thorough review and valuable suggestions. We agree that additional explicit details are needed to support the central claims and have revised the manuscript accordingly. Below we provide point-by-point responses to the major comments.

read point-by-point responses

Referee: [Abstract and results section] Abstract and results section: the central claim of collapse to a single curve with zero band width is load-bearing, yet the manuscript provides neither the explicit numerical values of the shifted weak-side matching points nor tables or supplementary data showing the 9-to-1 reduction and verification that the O(λ^{5/2}) term was inserted correctly. Without these, it is impossible to exclude post-hoc fitting artifacts or constraint-induced uniqueness.

Authors: We agree that the explicit numerical values and verification data are essential for reproducibility and to rule out artifacts. In the revised manuscript we have added a dedicated table in the results section listing the shifted weak-side matching points (now placed at values where the λ^{5/2} term is numerically relevant), the full set of weak-coupling coefficients before and after the upgrade, and a step-by-step demonstration of the reduction from nine nominal survivors to one. We also include a short verification subsection confirming correct insertion of the exact O(λ^{5/2}) coefficient by direct comparison of the truncated series. These additions directly address the concern. revision: yes
Referee: [Methods/ansatz section] Methods/ansatz section: the uniqueness result presupposes that the original LSTP interpolation ansatz remains a faithful representation once the weak-side matching points are moved into the O(λ^{5/2})-significant regime. No explicit consistency check or justification is supplied that the fixed ansatz does not become internally inconsistent with the higher-order data, raising the possibility that the observed collapse is an artifact of the constraint implementation rather than a genuine reduction in uncertainty.

Authors: The referee correctly identifies the need for an explicit consistency check. We have added a new paragraph and accompanying numerical test in the methods section of the revised version. The test verifies that the fixed LSTP ansatz continues to satisfy the matching conditions at the new points without introducing internal inconsistencies or spurious poles. Because the ansatz was constructed to accommodate the perturbative series through O(λ^{5/2}), the observed collapse is a direct consequence of the tighter constraints imposed by the exact coefficient rather than an artifact of the implementation. The added residuals and consistency metrics support this conclusion. revision: yes
Referee: [Comparison paragraph] Comparison paragraph: the reported non-coincidence between the unique LSTP survivor and the Hermite-Padé central curve is noted but not quantified (e.g., via pointwise difference or integrated discrepancy). This difference is important for assessing route dependence; the manuscript does not test whether additional terms would resolve or amplify it.

Authors: We agree that quantitative measures are necessary. The revised manuscript now includes both the maximum pointwise difference and the integrated L2 discrepancy between the unique LSTP survivor and the Hermite-Padé central curve over the crossover interval. Regarding the effect of additional terms, the O(λ^{-3}) strong-coupling coefficient is still unknown, so a direct test is not possible at present. We have expanded the comparison paragraph to discuss this limitation and to emphasize that the persistent difference illustrates method dependence between the LSTP and HP routes even after the weak-coupling upgrade. revision: partial

Circularity Check

0 steps flagged

No significant circularity; numerical outcome of updated constrained ensemble

full rationale

The paper updates the weak-coupling truncation with the exact O(λ^{5/2}) coefficient, shifts the weak-side matching points into the relevant regime, and reports that the pre-defined LSTP constraints then admit only a single curve. This is a direct numerical consequence of applying the existing interpolation ansatz and constraint set to the new input data, not a reduction of any claimed prediction or first-principles result to the inputs by construction. The ansatz is explicitly held fixed, the non-coincidence with the HP central curve is stated, and no load-bearing self-citation, uniqueness theorem, or ansatz smuggling appears in the derivation chain. The result is self-contained as an exploration of the constrained Padé method with upgraded data.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that the constrained LSTP ansatz remains a faithful interpolant when the new coefficient is included, together with the accuracy of the external O(λ^{5/2}) coefficient; no free parameters are explicitly listed in the abstract but the choice of matching points functions as an implicit tunable.

free parameters (1)

weak-side matching points
Shifted to the regime where the new term is numerically significant; their precise locations are not stated in the abstract but control which curves survive the constraints.

axioms (1)

domain assumption The log-subtracted two-point Padé ansatz with the stated constraints provides a valid family of interpolants for the thermodynamic functions of thermal N=4 SYM.
Invoked throughout the construction of the ensemble; no independent justification is given in the abstract.

pith-pipeline@v0.9.0 · 5490 in / 1610 out tokens · 60212 ms · 2026-05-10T07:26:34.061003+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

14 extracted references · 2 canonical work pages · 1 internal anchor

[1]

(4) The O(λ5/2) calculation does not contribute at order λ2 logλ , so A2 log is carried forward unchanged

log(1 + √ 2) − 1 64(6 + √ 2)π2 − 35 + 212 log 2 128 √ 2 ≃0.7537170300. (4) The O(λ5/2) calculation does not contribute at order λ2 logλ , so A2 log is carried forward unchanged. Through- out this paper we run two parallel LSTP scans with identical construction: abeforescan that truncates the weak series at O(λ2) (matching Ref. [ 7], i.e., with no A5/2) an...
[2]

0.75≤f(λ)≤1 on the evaluation grid, 2.d f /d(logλ)≤0 on the evaluation grid,
[3]

9” and the “5

no poles on the positive real λ axis after mapping back from thezplane. The 9 unknown coefficients in Eq. (10) are fixed by impos- ing 9 matching conditions on R(z(λ)): we require R(z(λi)) to equal the truncated target values at five weak-side points λi and at four strong-side points λj (the specific values of λi, λj are given below). The “9” and the “5” ...
[4]

to a single distinct LSTP curve, with pointwise band width falling to zero within numeri- cal resolution

the old 9 nominal LSTP survivors (3 distinct curves) collapse to 3 nominal survivors that are numerically identical, i.e. to a single distinct LSTP curve, with pointwise band width falling to zero within numeri- cal resolution
[5]

the old crossover range λc ∈ [2.9520, 6.7321] col- lapses to the unique value λc = 4 .7863, with f(λc) = 0.8371. This value lies inside the old range and the upward shift relative to the HP value (3 .52) and the M¨ uller value (3.14) is consistent with the positive sign of the exact A5/2 coefficient, which raises the weak-coupling series in the intermedia...
[6]

The comparison is asymmetric: only the LSTP route is updated with A5/2

the HP central curve remains distinct atλc = 3.5223. The comparison is asymmetric: only the LSTP route is updated with A5/2. The exact weak-side coef- ficient therefore removes within-route uncertainty but not structural route dependence
[7]

The main result is not simply that the band narrows: the exact O(λ5/2) coefficient selects a unique LSTP inter- polant

the strong-coupling residual estimator ˆS3(20) changes sign between the two routes (+2 .14 for LSTP, −21.86 for HP), giving a finite-coupling diag- nostic whose interpretation as the trueS 3 requires the next strong-coupling coefficient. The main result is not simply that the band narrows: the exact O(λ5/2) coefficient selects a unique LSTP inter- polant....
[8]

Q. Du, M. Strickland, and U. Tantary, JHEP08, 064 (2021) [Erratum: JHEP02, 053 (2022)]

2021
[9]

${\cal N}=4$ supersymmetric Yang-Mills thermodynamics to order $\lambda^{5/2}$

M. E. Carrington, G. Kunstatter, and U. Tantary, “ N= 4 supersymmetric Yang-Mills thermodynamics to order λ5/2,” arXiv:2604.05919 [hep-th] (2026)

work page internal anchor Pith review Pith/arXiv arXiv 2026
[10]

J. O. Andersen, Q. Du, M. Strickland, and U. Tantary, Phys. Rev. D105, 015006 (2022)

2022
[11]

J. M. Maldacena, Adv. Theor. Math. Phys.2, 231 (1998)

1998
[12]

S. S. Gubser, I. R. Klebanov, and A. A. Tseytlin, Nucl. Phys. B534, 202 (1998)

1998
[13]

M¨ uller, Phys

B. M¨ uller, Phys. Rev. D112, 054007 (2025)

2025
[14]

Tantary, Phys

U. Tantary, Phys. Rev. D (2026), doi:10.1103/ks69-drw2 (in press)

work page doi:10.1103/ks69-drw2 2026

[1] [1]

(4) The O(λ5/2) calculation does not contribute at order λ2 logλ , so A2 log is carried forward unchanged

log(1 + √ 2) − 1 64(6 + √ 2)π2 − 35 + 212 log 2 128 √ 2 ≃0.7537170300. (4) The O(λ5/2) calculation does not contribute at order λ2 logλ , so A2 log is carried forward unchanged. Through- out this paper we run two parallel LSTP scans with identical construction: abeforescan that truncates the weak series at O(λ2) (matching Ref. [ 7], i.e., with no A5/2) an...

[2] [2]

0.75≤f(λ)≤1 on the evaluation grid, 2.d f /d(logλ)≤0 on the evaluation grid,

[3] [3]

9” and the “5

no poles on the positive real λ axis after mapping back from thezplane. The 9 unknown coefficients in Eq. (10) are fixed by impos- ing 9 matching conditions on R(z(λ)): we require R(z(λi)) to equal the truncated target values at five weak-side points λi and at four strong-side points λj (the specific values of λi, λj are given below). The “9” and the “5” ...

[4] [4]

to a single distinct LSTP curve, with pointwise band width falling to zero within numeri- cal resolution

the old 9 nominal LSTP survivors (3 distinct curves) collapse to 3 nominal survivors that are numerically identical, i.e. to a single distinct LSTP curve, with pointwise band width falling to zero within numeri- cal resolution

[5] [5]

the old crossover range λc ∈ [2.9520, 6.7321] col- lapses to the unique value λc = 4 .7863, with f(λc) = 0.8371. This value lies inside the old range and the upward shift relative to the HP value (3 .52) and the M¨ uller value (3.14) is consistent with the positive sign of the exact A5/2 coefficient, which raises the weak-coupling series in the intermedia...

[6] [6]

The comparison is asymmetric: only the LSTP route is updated with A5/2

the HP central curve remains distinct atλc = 3.5223. The comparison is asymmetric: only the LSTP route is updated with A5/2. The exact weak-side coef- ficient therefore removes within-route uncertainty but not structural route dependence

[7] [7]

The main result is not simply that the band narrows: the exact O(λ5/2) coefficient selects a unique LSTP inter- polant

the strong-coupling residual estimator ˆS3(20) changes sign between the two routes (+2 .14 for LSTP, −21.86 for HP), giving a finite-coupling diag- nostic whose interpretation as the trueS 3 requires the next strong-coupling coefficient. The main result is not simply that the band narrows: the exact O(λ5/2) coefficient selects a unique LSTP inter- polant....

[8] [8]

Q. Du, M. Strickland, and U. Tantary, JHEP08, 064 (2021) [Erratum: JHEP02, 053 (2022)]

2021

[9] [9]

${\cal N}=4$ supersymmetric Yang-Mills thermodynamics to order $\lambda^{5/2}$

M. E. Carrington, G. Kunstatter, and U. Tantary, “ N= 4 supersymmetric Yang-Mills thermodynamics to order λ5/2,” arXiv:2604.05919 [hep-th] (2026)

work page internal anchor Pith review Pith/arXiv arXiv 2026

[10] [10]

J. O. Andersen, Q. Du, M. Strickland, and U. Tantary, Phys. Rev. D105, 015006 (2022)

2022

[11] [11]

J. M. Maldacena, Adv. Theor. Math. Phys.2, 231 (1998)

1998

[12] [12]

S. S. Gubser, I. R. Klebanov, and A. A. Tseytlin, Nucl. Phys. B534, 202 (1998)

1998

[13] [13]

M¨ uller, Phys

B. M¨ uller, Phys. Rev. D112, 054007 (2025)

2025

[14] [14]

Tantary, Phys

U. Tantary, Phys. Rev. D (2026), doi:10.1103/ks69-drw2 (in press)

work page doi:10.1103/ks69-drw2 2026