REVIEW 1 major objections 4 minor
$L^p$-Estimates for maximal averages along mixed homogeneous hypersurfaces in $\mathbb{R}^{3}$
T0 review · 1 major / 4 minor · reviewed 2026-08-27 · deepseek-v4-flash
Pith's one-line read The paper determines the exact critical Lebesgue exponent for maximal averages along mixed homogeneous surfaces in $\mathbb{R}^3$ from the multiplicities of the real roots of the Hessian determinant.
desk verdict Serious advance on sharp maximal estimates for mixed homogeneous hypersurfaces, but the sufficiency proof for Type B_T relies on an unproved curved-cone FIO estimate that must be supplied before acceptance. 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 argument is carried by a classification of the real roots of the Hessian determinant $H_\Phi$ into Types A, $B_T$, $B_{NT}$, and C, according to which of $\Phi_1,\Phi_2$ vanish along the root orbit $\gamma_\tau=\{(x_1,\tau x_1^{\kappa_2/\kappa_1}):x_1>0\}$, and by the factorization $H_\Phi(x)=(x_2-\tau x_1^{\kappa_2/\kappa_1})^{N_\tau}V_\tau(x)$ near each root. Sufficiency is proved by reducing to narrow neighborhoods of these roots, applying stationary phase to obtain Airy-type integrals, and then bounding the resulting Fourier multipliers: Type $B_T$ roots require the theory of FIO-cone multipliers on curved cones, while Type C roots reduce to a rescaled flat-cone multiplier; necessity is obtained from local lower bounds based on a quantitative transversality lemma.
What would settle it
A concrete check is whether the curved-cone square-function estimate used for Theorem 7.1 holds for the specific curved cone arising in Section 7.3; if it fails, the claimed Type $B_T$ bound for $p_\Phi$ collapses, and a counterexample to Theorem 1.3 would follow.
Extended reading notes
Core claim
The central claim is Theorem 1.3: if $\Phi$ is real-analytic on $\mathbb{R}^2\setminus\{0\}$, $\kappa$-homogeneous of degree $1$, with $\kappa_1,\kappa_2>0$, $|\kappa|\le 1$, $\kappa_1\ne\kappa_2$, $H_\Phi$ not identically zero, and Assumptions 1.2 satisfied, then the maximal operator $\mathcal{M}$ is $L^p$-bounded for every $p>p_\Phi$ and unbounded for every $p<p_\Phi$, where $p_\Phi=\max\{3/2,2(M_{B_T}+1)/(M_{B_T}+3),2M_C/(M_C+1)\}$. Here $M_{B_T}$ and $M_C$ are maxima, over roots of Type $B_T$ and Type $C$, of $M(\zeta)=N(\zeta)+2$, with $N(\zeta)$ the multiplicity of the root $\zeta$ of the Hessian determinant; if $p_c=3/2$, the operator is unbounded also at the endpoint $p=3/2$.
Load-bearing premise
The whole Type $B_T$ sufficiency argument rests on an unproved $L^4$ estimate for curved-cone FIO multipliers, Theorem 7.1, which the paper says follows by adapting the methods of [6] and the curved-cone square-function estimate from [11]; if that extension fails, the claimed $B_T$ bound collapses.
Editorial extensions
If this is right
- For every mixed homogeneous polynomial $\Phi$ in the stated class, the critical exponent can be read off directly from the Hessian-root multiplicities as $p_c=\max\{3/2,2(M_{B_T}+1)/(M_{B_T}+3),2M_C/(M_C+1)\}$.
- Surfaces through the origin can have $p_c$ as low as $3/2$, so non-transversality does not force $p_c\ge 2$; earlier general sufficient ranges above $2$ are not sharp.
- Only roots of Type $B_T$ or Type $C$ can raise $p_c$ above $3/2$; Type A and Type $B_{NT}$ roots impose only the baseline threshold.
- Known sharp bounds for convex mixed homogeneous surfaces, such as $p_c=2a_2/(a_2+1)$ for $\Phi(x)=x_1^2+x_2^{a_2}$ with even $a_2>2$, are recovered as special cases.
- When $p_c=3/2$, the operator fails to be bounded at the endpoint $p=3/2$, so the bounded range is open at the critical value.
Reading between the lines
- A natural extension would be to test whether the same root-type formula persists when $|\kappa|>1$ or when $H_\Phi$ vanishes identically; the authors explicitly say these cases need different tools, but if the formula survived it would complete the classification.
- The same FIO-cone multiplier route could plausibly locate $p_c$ for non-transversal maximal averages in higher dimensions, with the baseline exponent $n/(n-1)$ replacing $3/2$ and root-type contributions taking analogous multiplicity-driven forms.
- If the curved-cone square-function estimate used for Theorem 7.1 is verified, the endpoint statement at $3/2$ would rule out any endpoint $L^{3/2}$ bound, mirroring the spherical maximal operator.
- A testable variation is to replace real-analytic by merely smooth mixed-homogeneous functions; the factorization and finite-multiplicity arguments rely on analyticity, so a smooth counterexample would show that the threshold is not purely combinatorial.
Signed reviews
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies maximal averaging operators in R^3 along hypersurfaces that are graphs of mixed (κ-)homogeneous functions Φ analytic away from the origin, with κ1,κ2>0, κ1≠κ2 and |κ|≤1. The main result, Theorem 1.3, gives a sharp critical exponent p_Φ = max{3/2, 2(M_BT+1)/(M_BT+3), 2M_C/(M_C+1)}, where M_BT and M_C are defined through multiplicities of real roots of the Hessian determinant HΦ. The proof splits roots of HΦ into Types A, B_T, B_NT and C; necessary conditions are proved in Section 3 via a quantitative transversality argument, and sufficiency is reduced to uniform maximal estimates for localized Fourier multipliers in Theorem 4.2. The Type B cases are handled by reducing to FIO-cone multiplier estimates in Section 7, and the Type C case is handled in Section 8.
Significance. If the proof is completed, the result is significant: it determines the sharp L^p-range for a broad class of non-transversal surfaces in R^3, going beyond previous work that assumed transversality, and it connects the problem with the recent FIO-cone multiplier theory. The critical exponent is expressed directly in terms of root multiplicities rather than fitted constants, and the paper includes several illustrative examples, including polynomial and non-polynomial cases. The necessary-condition part is self-contained and appears sound. However, a central estimate for Type B roots, Theorem 7.1, is stated without proof and is deferred to the companion/in preparation work [6] and to [11], so the sufficiency direction for Type B roots is not established in the present manuscript.
major comments (1)
- [§7.2, Theorem 7.1; §7.3, (7.51)–(7.61)] Theorem 7.1 is the sole source of the L^4 FIO-cone multiplier estimate for curved cones, and Corollary 7.2 together with (7.51)–(7.61) uses it to prove Theorem 4.2(b) for both Type B_T and Type B_NT roots. The theorem is not proved in the manuscript: the paragraph beginning "As for the proof of this theorem" states only that an inspection of [6] and the curved-cone analogue of (7.39), cited to [11, Remark 1.7], yields the result. Neither the analogue of (7.39) for the cone (7.30) nor the adaptation of the induction-on-scales argument to the phase class F_{1,γ/2,γ} with the small mixed derivative condition (7.37) is stated or verified. Since the sharp exponent 2(M_BT+1)/(M_BT+3) collapses if Theorem 7.1 fails, the sufficiency direction of Theorem 1.3 is not established for roots of Type B in the present manuscript.
minor comments (4)
- [§1.3, Example 1.11] The example discusses the roots γ(±1,−1), but the text writes "both of multiplicity N(±1,1)=m−2"; the notation should be N(±1,−1).
- [§7.2, display (7.38)] The operator norm in (7.38) should be typeset as ∥T_{m_{λ,R}}∥_{L^4→L^4}; as printed it reads "∥T mλ,R∥L4→L4" with a missing subscript.
- [References] Reference [31] has a typo: "Hamonic Analysis" should be "Harmonic Analysis".
- [§7.3, Lemma 7.3 and amplitude verification] The verification that φ0+φγ lies in F_{1,γ/2,γ} and satisfies the SMD condition, and the assertion that the amplitude a_{λ,R} is admissible, are compressed into phrases such as "easy to see", "similarly", and "by Leibniz' rule"; one or two displayed estimates for the mixed derivative and for the amplitude would make this part substantially easier to check.
Circularity Check
No circular derivation: p_c is defined by root multiplicities and proved via independent necessary/sufficient estimates; the main caveat is the external, self-cited curved-cone FIO multiplier estimate (Theorem 7.1), which is a gap rather than a circularity.
full rationale
The critical exponent p_Phi in Theorem 1.3 is computed directly from multiplicities N(zeta) = ord H_Phi(zeta) and M(zeta) = N(zeta)+2; these quantities are properties of the input function Phi, not constants fitted to the maximal operator. The necessity part (Theorem 3.1) is proved in the paper from Proposition 3.2 and Corollary 3.4 / Proposition 3.3, which are self-contained lower-bound arguments based on transversality and convex bodies. The sufficiency part reduces to Theorem 4.2 and then to Fourier multiplier estimates. For Type B_T and B_NT it invokes Theorem 7.1, an L^4 estimate for FIO-cone multipliers on curved cones. This theorem is stated without proof; the text says it follows from an inspection of [6] and the curved-cone analogue of the Guth-Wang-Zhang estimate from [11]. [6] is by overlapping authors, so this is a load-bearing self-citation in the sufficiency argument, and the extension from the light cone to curved cones is not demonstrated in the manuscript. However, this is a correctness/completeness gap, not an instance of the paper's central claim being equivalent to its input by construction: no equation is defined in terms of the target L^p exponent, no fitted parameter is renamed as a prediction, and no result is derived from itself. The endpoint and all lower bounds are independent of Theorem 7.1. Thus the circularity score remains low (2), reflecting the self-citation dependency, not a circular derivation.
Assumptions & free parameters
assumptions (4)
- domain assumption Theorem 7.1: L^4 estimate for FIO-cone multipliers on curved cones with phase in F_{κ1,κ2,γ} and SMD.
- standard math Guth-Wang-Zhang square-function estimate (7.39) and its extension to curved cones.
- standard math Sharp L^p maximal estimates for hypersurfaces with non-vanishing Gaussian curvature ([9], [13]).
- standard math Airy-type integral asymptotics of [17, Lemma 2.2].
Cite this review
Pith. "Pith review of $L^p$-Estimates for maximal averages along mixed homogeneous hypersurfaces in $\mathbb{R}^{3}$." pith.science (2026). https://pith.science/paper/7WWUBUI3
@misc{pith2026260822012,
author = {Pith},
title = {Pith review of: $L^p$-Estimates for maximal averages along mixed homogeneous hypersurfaces in $\mathbbR^3$},
year = {2026},
howpublished = {\url{https://pith.science/paper/7WWUBUI3}},
note = {Machine review of arXiv:2608.22012}
}
abstract
In this paper, we study $L^p$-estimates for maximal averaging operators $\mathcal M$ along hypersurfaces $S$ in $\mathbb{R}^{3}$ which are the graph of a mixed homogeneous function $\Phi$ which is analytic away from the origin. The closure of such a surface will pass through the origin, so that the usual transversality condition that had been imposed in many previous works on maximal averages along hypersurfaces will not hold even when $\Phi$ is analytic at the origin. As our main result, under mild assumptions which are satisfied for instance for every mixed homogeneous polynomial $\Phi,$ we determine the critical Lebesgue exponent $p_c$ for which $\mathcal M$ is $L^p$-bounded for every $p>p_c,$ but unbounded for $p<p_c,$ in terms of multiplicities of the real roots of the Hessian determinant of $\Phi.$ It turns out that the study of the contributions by neighborhoods of a certain type of roots is closely related to recent work by Dendrinos, Ikromov and the first and third author on sharp estimates for a maximal averaging operator along a transversal hypersurface of an ``exceptional'' class, whose $L^p$-boundedness had been an open problem for a long time and which has recently been established by means of their new theory of FIO-cone multipliers.
Reviewed August 27, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.