REVIEW 2 major objections 4 minor 1 cited by
Supersymmetry and trace formulas III. Frenkel trace formula
T0 review · 2 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read The paper derives an exact trace formula for the heat kernel on compact semisimple Lie groups with left and right translations, generalizing the Frenkel trace formula via supersymmetric localization.
desk verdict A plausible generalization of the Frenkel trace formula for all semisimple compact Lie groups, but the derivation rests entirely on an unproved localization principle imported from the authors' earlier work. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The machinery is the supersymmetric localization principle: one adds a $Q$-exact deformation, written in Eq. (2.12), whose fixed-point set is the constant-velocity locus $g(\tau)=e^{\gamma\tau/\beta}g_0$ with $\gamma\in\Gamma$. Around this locus the path integral factorizes into a classical action, a one-loop Pfaffian/determinant ratio, and a finite fermionic zero-mode integral, and the Harish-Chandra orbital integral converts the remaining $G/T$ integral into the Weyl-group sum. Two lattice lemmas (that $e^{\langle\alpha,\gamma\rangle}=1$ and $e^{\langle\rho,w\gamma\rangle}=e^{\langle\rho,\gamma\rangle}$ for $\gamma\in\Gamma$) move the $\gamma$-dependence out of the Weyl denominator and produce the clean factor $e^{\langle\rho,\gamma\rangle}$ in the final formula.
What would settle it
Compute the two sides of Eq. (2.24) for $G=SO(3)$ at fixed $\beta$ and regular $h_l,h_r$: the spectral side is the sum $\sum_\lambda d_\lambda \chi_\lambda(e^{h_l})\chi_\lambda(e^{-h_r})e^{-\beta C_2(\lambda)/2}$ over irreducible representations, and the lattice side sums over $W\times\Gamma$; any nonzero difference would kill the identity.
Extended reading notes
Core claim
The central claim is Eq. (2.24): for a compact semisimple Lie group $G$ with characteristic lattice $\Gamma$, for regular $h_l,h_r\in\mathfrak{t}$, the trace of the heat kernel with left and right translation insertions equals the Weyl-group/lattice sum on the right-hand side. The left side is the operator trace in $L^2(G)$; the right side has denominator $s(h_l)s(-h_r)$, a factor $e^{\frac12\beta\langle\rho,\rho\rangle}$, and the new factor $e^{\langle\rho,\gamma\rangle}$ inside the lattice sum. This generalizes Frenkel's formula, which was stated for simply connected simple compact groups; in that case $\Gamma=2\pi iQ^\vee$ and the extra factor is identically $1$, so the formula reduces to the original. The paper presents this identity as an exact localization result, with both derivations reducing the path integral to a one-loop evaluation around constant-velocity loops.
Load-bearing premise
The proof assumes that the supersymmetric localization principle from the authors' earlier work applies exactly to the twisted actions (2.6) and (3.8), reducing the path integral to the constant-velocity locus; the paper offers no independent proof of that localization step.
Editorial extensions
If this is right
- The identity (2.24) holds for every compact semisimple Lie group, not just simply connected ones, adding the factor $e^{\langle\rho,\gamma\rangle}$ to the lattice sum.
- The left-right trace can be evaluated exactly without expanding in $\beta$; each term in the lattice sum has the structure of a one-loop determinant around a constant-velocity loop.
- The two derivations give the same formula from different localization loci, connecting the Eskin-type derivation on $G$ with the Selberg-type derivation on the gauged $G\times G$ model.
- When $G$ is simply connected, the new formula reduces to the original Frenkel trace formula because the extra factor equals one.
Reading between the lines
- The same twisted-action technique could be tested on non-compact or infinite-dimensional targets, where the characteristic lattice would be replaced by the appropriate lattice of periods; the exactness of the localization would have to be checked anew in each case.
- The appearance of $\Gamma/2\Gamma$ in the gauged-model derivation hints at a double-cover structure of the maximal torus that the paper does not exploit; a representation-theoretic interpretation of this decomposition could be a next step.
- For a non-simply-connected group such as $SO(3)$, the extra factor $e^{\langle\rho,\gamma\rangle}$ could be detected numerically by comparing the lattice-sum side with the spectral side at finite $\beta$.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents two path-integral derivations of a generalized Frenkel trace formula for a compact semisimple Lie group G. The main result, Eq. (2.24), expresses Tr_{L^2(G)}[L_{e^{h_l}}R_{e^{-h_r}}e^{-(1/2)\beta\Delta_G}] as a sum over (w,\gamma) in W\times\Gamma, with a Weyl denominator and a phase e^{\langle\rho,\gamma\rangle}. This generalizes Frenkel's formula, originally stated for simply connected simple compact groups, to arbitrary semisimple compact groups. Section 2 uses a supersymmetric non-linear sigma model on G with a right-twisted action; Section 3 uses a gauged sigma model on G\times G. Both derivations rely on the authors' earlier supersymmetric localization principle and on the Harish-Chandra orbital integral formula.
Significance. If the localization assumption is granted, the paper offers a conceptually unified physical derivation of the Eskin, Selberg, and Frenkel trace formulas, and a natural extension to non-simply connected groups through the characteristic lattice phase e^{\langle\rho,\gamma\rangle}. The two derivations are structurally independent and explicitly bridge the authors' earlier works [4,5]. The final formula is explicit and reduces exactly to Frenkel's formula in the simply connected case, which is a valuable consistency check. The main weakness is that the core localization principle is not proved here, and the right-twisted setting is not a trivial corollary of the earlier cases; the significance of the paper is therefore conditional on closing that gap and on correcting the Harish-Chandra formula issues described below.
major comments (2)
- [Eq. (2.20), Section 2] The Harish-Chandra formula stated in Eq. (2.20) is not the analytic continuation of Eq. (1.9). For a rank-one example, the left-hand side of (2.20) behaves as e^{xy}/(xy) for large regular arguments, while the right-hand side with the factors \pi(X)\pi(\lambda) grows like xy e^{xy}. With Eq. (2.20) as written, the factors \pi(h_l+\gamma)\pi(h_r) in Eq. (2.19) do not cancel, and the constant (2\pi\beta)^{r/2} in Eq. (2.21) does not follow. The correct formula should be the analytic continuation of (1.9), namely a product of 2\pi/(\langle\alpha,X\rangle\langle\alpha,\lambda\rangle) times the Weyl sum, up to a sign convention. The authors should correct (2.20) and rederive the prefactors in (2.21) and in Section 3 accordingly.
- [Section 2, Eqs. (2.6), (2.7), (2.12); Section 3, Eqs. (3.8)-(3.15)] The derivation of the main theorem is conditional on the supersymmetric localization principle of [4,5], but the paper does not prove that this principle applies to the right-twisted actions used here. The action (2.6) contains the h_r-dependent fermion coupling (\psi,(\partial_\tau+\mathrm{ad}_{h_r/\beta})\psi) and the shifted bosonic current J_{l,r}=J+\mathrm{Ad}_{g^{-1}}h_l/\beta-h_r/\beta, while the supersymmetry (2.7) and deformation (2.12) are written in terms of J_l only. The assertion that the supersymmetry is 'exactly the same' as in [4] does not by itself establish invariance of the full twisted action or the Q-exactness of V in the presence of the h_r term. A direct computation of \delta S_{h_l,h_r} and of the bosonic part of \delta V, or an explicit reduction to the proof in [4], is needed. The same gap appears in the gauged sigma model derivation, where the right-hand side of (3.11) is localized using the deformations (3.15).
minor comments (4)
- [Eq. (2.11)] The argument of the hyperbolic sine is written as '\langle h_r,r\rangle'; it should be '\langle\alpha,h_r\rangle'.
- [Paragraph before Eq. (2.19)] The text refers to 'the localized path integral I in (2.4)', but I is defined in Eq. (2.9); the cross-reference should be corrected.
- [Section 3, Eq. (3.10)] The definition of \chi_p(A) introduces a factor 2^{r/2} relative to the coefficient c_r in Section 2; a sentence explaining this normalization, beyond the parenthetical remark on the fermion measure, would improve readability.
- [Throughout] The paper relies on the localization principle from [4,5]; since [5] is listed as 'to appear', the authors should ensure that the cited results are available to the reader or summarize the needed statements in an appendix.
Circularity Check
No significant circularity: the generalized Frenkel trace formula is derived by explicit localization computations and is not used as an input.
full rationale
The paper's derivation chain starts from the path-integral representation of the supertrace (2.9), computes the localizing deformation (2.12), the localization locus (2.14), the one-loop determinants (2.16), the fermionic constant-mode integral (2.18), and the Harish-Chandra orbital integral (2.20); the final formula (2.24) is then assembled by algebraic simplification using Lemma 1. The gauged sigma model derivation in Section 3 is likewise self-contained after gauge fixing: the localization data (3.15)-(3.18), the gamma-sum rearrangement (3.23), and the Gaussian integral (3.24) are all evaluated in the paper. The target trace formula is never used as an input, there are no fitted parameters, and no equation is assumed that is equivalent to the result being derived. The only imported ingredient is the supersymmetric localization principle from the authors' prior papers [4,5], but that is a general method previously applied to the Eskin and Selberg trace formulas, not the Frenkel formula itself, and so it constitutes real external support rather than a circular premise. Concerns about whether the localization argument is fully justified for the right-twisted action are proof-gap or correctness concerns, not circularity. The reduction to Frenkel's simply-connected formula is a consistency check and does not reveal a circular step.
Assumptions & free parameters
assumptions (5)
- domain assumption Supersymmetric localization principle: the path integral with fermionic zero mode insertions and deformation s delta V localizes exactly to the locus dot J = 0.
- standard math Harish-Chandra orbital integral formula (Eqs. (1.9), (2.20)): integral over G/T of e^{(X, g^{-1} lambda g)} equals (2 pi)^{dim(G/T)/2} / (pi(X) pi(lambda)) sum_W eps(w) e^{(w lambda, X)}.
- standard math Freudenthal-de Vries 'strange formula': R = 6 <rho,rho> for the scalar curvature of the Cartan-Killing metric.
- domain assumption Path integral representation of the operator trace as an integrated propagator (Eq. (1.7)).
- domain assumption Twisting by global symmetry insertions is equivalent to coupling to constant background gauge fields h_l/beta and h_r/beta (Eqs. (2.6), (3.7)).
Cite this review
Pith. "Pith review of Supersymmetry and trace formulas III. Frenkel trace formula." pith.science (2026). https://pith.science/paper/5JWBJFBA
@misc{pith2026250210210,
author = {Pith},
title = {Pith review of: Supersymmetry and trace formulas III. Frenkel trace formula},
year = {2026},
howpublished = {\url{https://pith.science/paper/5JWBJFBA}},
note = {Machine review of arXiv:2502.10210}
}
abstract
By applying the new supersymmetric localization principle introduced in \cite{Choi:2021yuz,Choi:2023pjn}, we present two complementary approaches for the path integral derivation of the `non-chiral' trace formula for a semisimple compact Lie group $G$, which generalizes the so-called Frenkel trace formula. Corresponding physical systems for each picture are the quantum mechanical sigma model on $G$ and the gauged sigma model on $G\times G$, and the approaches closely follow the spirit of the Eskin trace formula \cite{Choi:2021yuz} and the Selberg trace formula \cite{Choi:2023pjn} respectively. These methods provide a natural conceptual bridge between two seemingly independent derivations in \cite{Choi:2021yuz} and \cite{Choi:2023pjn}.
Forward citations
Cited by 1 Pith paper
-
Localization of strings on group manifolds
Supersymmetric localization reproduces the WZW partition function as a sum over abelian classical solutions, verified for SU(2) and extended to SL(2,R) and H_3^+.
Reference graph
Works this paper leans on
-
[4]
Supersymmetry and trace formulas I. Compact Lie groups
Changha Choi and Leon A. Takhtajan, Supersymmetry and trace formulas. Part I. Compact Lie groups , JHEP 06 (2024), 026; arXiv:2112.07942
work page Pith review arXiv 2024
-
[1]
N. Berline, E. Getzler, and M. Vergne, Heat kernels and Dirac operators , Springer Science & Business Media, 2003
work page 2003
-
[2]
Jean-Michel Bismut, The hypoelliptic Laplacian on a compact Lie group , J. Funct. Anal. 255 (2008), no. 9, 2190–2232
work page 2008
-
[3]
Roberto Camporesi, Harmonic analysis and propagators on homogeneous spaces , Phys. Rep. 196 (1990), no. 1-2, 1–134
work page 1990
-
[5]
, Supersymmetry and trace formulas. Part II. Selberg trace fo rmula, to appear in Adv. Theor. Math. Phys. (2025), arXiv:2306.13636
work page Pith review arXiv 2025
-
[6]
J. S. Dowker, When is the ‘sum over classical paths’ exact? , J. Phys. A 3 (1970), no. 5, 451
work page 1970
-
[7]
, Quantum mechanics on group space and Huygens’ principle , Annals Phys. 62 (1971), 361–382
work page 1971
-
[8]
L. D. Èskin, Heat equation on Lie groups , In Memoriam: N. G. Čebotarev (Russian), Izdat. Kazan. Univ., Kazan, 1964, pp. 113–132. SUPERSYMMETRY AND TRACE FORMULAS III. FRENKEL TRACE FORMUL A 13
work page 1964
Show all 14 references
-
[9]
I. B. Frenkel, Orbital theory for affine Lie algebras , Invent. Math. 77 (1984), 301–352
1984
-
[10]
Harish-Chandra, Differential operators on a semisimple Lie algebra , Amer. J. Math. (1957), 87–120
1957
-
[11]
M. S. Marinov and M. V. Terentyev, Dynamics on the group manifold and path integral , Fortsch. Physik 27 (1979), no. 11-12, 511–545
1979
-
[12]
R. F. Picken, The propagator for quantum mechanics on a group manifold from an infinite dimensional analog of the Duistermaat-Heckman integratio n formula , J. Phys. A 22 (1989), 2285
1989
-
[13]
Schulman, A path integral for spin , Phys
L. Schulman, A path integral for spin , Phys. Rev. 176 (1968), 1558–1569
1968
-
[14]
Selberg, Harmonic analysis and discontinuous groups in weakly symme tric Riemannian spaces with applications to Dirichlet series , J
A. Selberg, Harmonic analysis and discontinuous groups in weakly symme tric Riemannian spaces with applications to Dirichlet series , J. Indian Math. Soc. (N.S.) 20 (1956), no. 1-3, 47–87. Perimeter Institute for Theoretical Physics, W aterloo, Ontario, N2L 2Y5, Canada Departm...
1956
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.