Pith. sign in

REVIEW 4 major objections 3 minor 67 references

The multiplicative constant in asymptotics of higher-order analogues of the Tracy-Widom distribution

T0 review · 4 major / 3 minor · reviewed 2026-08-10 · deepseek-v4-flash

Pith's one-line read The paper proves an explicit multiplicative constant in the large-gap asymptotics of higher-order Tracy-Widom distributions, expressed through the Hamiltonian of a special Painlevé I hierarchy solution.

desk verdict Settles a real open constant problem with a clever contour-switching method, but the proof of two uniformity claims is not fully displayed. read the letter →

arxiv 2501.12679 v2 pith:4GHHX24O submitted 2025-01-22 math-ph math.MP

classification math-phmath.MP MSC 60B2034M5541A6033E17
keywords higher-orderTracy-WidomdistributionPainlevéIhierarchymultiplicativeconstantlargegapasymptoticsFredholmdeterminantRiemann-HilbertproblemXXXIVparametrixAirykernel
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 determines the previously unknown constant term in the large-gap asymptotics of higher-order Tracy-Widom distributions, which describe eigenvalue fluctuations at critical edge points of unitary random matrix models. The constant is expressed in closed form using an integral of the Hamiltonian associated with a special real pole-free solution of the even Painlevé I hierarchy, together with known constants from the classical Tracy-Widom distribution. A by-product of the proof is that the total integral of this Hamiltonian difference vanishes for every order, and the same uniform asymptotics reveal a transition from the higher-order distribution to the classical Tracy-Widom distribution in the appropriate scaling regime. The approach is presented as a template for evaluating similar multiplicative constants in other Fredholm determinants arising in mathematical physics.

What carries the argument

The central object is the $\mathrm{P}_I^{2k}$ kernel, a higher-order analogue of the Airy kernel built from solutions of the even Painlevé I hierarchy, and the associated Fredholm determinant $F(s;x)=\log\det(I-K_s^{(k)})$. The argument uses the two differential identities $\partial F/\partial x=-(Y_{-1})_{12}$ and $\partial F/\partial s=\lim_{\zeta\to s}(2\pi i)^{-1}(X(\zeta)^{-1}X'(\zeta))_{21}$, which connect the derivatives of the determinant to Riemann-Hilbert problems, and then applies nonlinear steepest descent analysis in three asymptotic regions: algebraic growth, transition, and exponential decay. The transition region uses a Painlevé XXXIV parametrix whose asymptotics, when integrated over the short transition interval, produce the classical Tracy-Widom constant $\chi(0)$, while the Hamiltonian $h$, defined through $dh/dx=q$ for the special pole-free solution $q$ of $\mathrm{P}_I^{2k}$, enters through the $x$-derivative identity and supplies the new constant $I_h(x)$.

What would settle it

Compute the left and right sides of the cancellation identity (2.19) numerically for large $|s|$ with $k=1$; if their difference is not $O(|s|^{-1/3})$, the extracted constant is wrong. Alternatively, evaluate $\log\det(I-K_s^{(1)})$ directly for $s=-40$ and fixed $x=0$, subtract the explicit $s$-dependent terms, and compare the remainder with the $k=1$ case of the constant formula in Corollary 1.2.

Watch

Extended reading notes

Core claim

For $k=1,2,\ldots$, the paper establishes that as $s\to -\infty$, with all auxiliary parameters $t_j$ set to zero, the logarithm of the higher-order Tracy-Widom determinant admits the expansion stated in Theorem 1.1, whose constant term is now explicit: it involves $-I_h(x)$, where $I_h(x)=\int_{+\infty}^{x}[h(\mu)-h_{\mathrm{Asy}}(\mu)]d\mu$, plus polynomial and logarithmic terms in $x$ and the classical Tracy-Widom constant $\chi(0)$. This closes the open constant problem left by earlier work that derived the asymptotic expansion only up to an undetermined $s$-independent constant $C(k)$. The proof also yields the vanishing total integral $\int_{-\infty}^{+\infty}[h(\mu)-h_{\mathrm{Asy}}(\mu)]d\mu=0$ for all $k$, and, by substituting a suitable $x$ depending on $s$, the higher-order distribution converges to the classical Tracy-Widom distribution in the large-gap regime.

Load-bearing premise

The whole proof hinges on a delicate cancellation: the error term in the transition-region asymptotics of $\partial F/\partial s$ grows with $|x|$, and the derivation depends on its integrated contribution over the short transition interval being negligible when substituted into the contour identity (1.41).

Editorial extensions

If this is right

  • The formerly open constant $C(k)$ in the known expansion (1.20) is now explicit for every $k$, completing the large-gap asymptotics of higher-order Tracy-Widom distributions.
  • The total integral of the Hamiltonian difference vanishes for all $k$, extending the known $\mathrm{P}_I^{2}$ result and giving a global identity for the special solutions of the even Painlevé I hierarchy.
  • The uniform asymptotics in $x$ yield an explicit transition from the higher-order Tracy-Widom distribution to the classical Tracy-Widom distribution in the large-gap regime, matching a kernel-level transition established earlier.
  • The paper states that the same strategy, using uniform partial-derivative asymptotics and a two-path contour integration, can be adapted to compute analogous multiplicative constants in other problems from mathematical physics.

Reading between the lines

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

  • Because the formula (1.29) is uniform in $x$ over an $s$-dependent window, one can likely read off subleading $x$-dependent corrections at the transition to the Airy kernel, beyond the leading transition recorded in Corollary 1.4.
  • The vanishing total integral of $h-h_{\mathrm{Asy}}$ suggests that $h-h_{\mathrm{Asy}}$ has equal tail contributions at $\pm\infty$, which may admit a tau-function interpretation and could be derived independently from isomonodromic properties of the Painlevé I hierarchy.
  • For $k=1$, the explicit constant in Corollary 1.2 can be tested numerically against direct evaluation of the Fredholm determinant for moderately large $|s|$; agreement would provide strong evidence for the general formula, and disagreement would point to the delicate transition-region error cancellation as the source of error.
  • The same two-derivative contour trick may work for determinants where only the asymptotics of derivatives are accessible, such as kernels built from the Painlevé II hierarchy, with the constant again expressed through a Hamiltonian integral.
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.

Referee Report

4 major / 3 minor

Summary. The paper determines the previously unknown multiplicative constant C(k) in the large gap asymptotics of the higher-order Tracy-Widom distributions det(I - K_s^{(k)}) for k = 1, 2, 3, \ldots. Theorem 1.1 gives an explicit expansion of F(s;x) = log det(I - K_s^{(k)}) as s \to -\infty, with a constant term involving the integral I_h(x) of the Hamiltonian h of the special real pole-free solution of the even Painlev\'e I hierarchy P_I^{2k}, the classical Airy constant \chi(0), and explicit logarithmic terms, with error O(|s|^{-\epsilon_0}) uniformly in x over the window [-c_1|s|^{2k+1}, \alpha_k|s|^{2k+1} - c_2|s|^{2k/3+\epsilon}]. The proof uses Riemann-Hilbert steepest descent within the framework of Claeys, Its and Krasovsky, together with a new two-contour strategy in the (s,x)-plane that exploits uniform asymptotics of both \partial F/\partial s and \partial F/\partial x and compares the contours ending at x_0 = \pm|s|^{2k+1}. The paper also proves (Corollary 1.3) that the total integral of h - h_{Asy} vanishes for all k, and (Corollary 1.4) that the expansion reduces to the classical Tracy-Widom asymptotics near the right edge of the x-window.

Significance. If the main theorem is correct, this is a substantial result: it resolves an open problem from Claeys, Its and Krasovsky (2010) by giving an explicit formula for the previously unknown constant for every k, with no fitted parameters, and the appearance of the Painlev\'e I Hamiltonian integral is a new structural feature. The paper is self-aware about the delicate step (Remark 2.5 explicitly acknowledges that the transition-region error in Lemma 2.3 is large and is controlled only after integration), and the cancellation in (2.19) is verified by the explicit expansions (2.20)-(2.22). The result passes internal consistency tests: for k = 1 it yields the form (1.20) with a concrete constant; Corollary 1.3 generalizes the known k = 1 total-integral identity of [21]; and Corollary 1.4 recovers the classical Tracy-Widom expansion (1.15) with the correct constant \chi(0). The method is plausibly transferable to Pearcey-type and other Fredholm determinants. The main reservation is that the constant extraction relies on uniformity of the transition-region asymptotics in the parameter \kappa_0, which is asserted rather than fully demonstrated; this is the subject of the major comments.

major comments (4)
  1. [§5.3, Eqs. (5.71), (5.78); used at Eq. (2.16)] Lemma 2.3's estimate (2.4) has an additive error O(|x|^{(k-1/3)/(2k+1)}) which is not small as |x| \to \infty; as Remark 2.5 acknowledges, the estimate is used only through its integral over (s1, s2), which in (2.16) yields O(|s|^{-1/6}) because length x error = O(|s|^{-k+1/6}) x O(|s|^{k-1/3}). For this product to be a genuine error bound, the O-term in (2.4) must be uniform in s across the whole transition interval, including the point where \alpha_k s^{2k+1} + x = 0, i.e. where \kappa_0 = b_0 = 0 in the scaled variables of Section 5. The derivation in Section 5.3 obtains (5.71) and (5.78) from (5.69), (5.70), (B.20) and (B.21), but the passage is compressed ('readily seen'): Lemma B.4 is stated only for x \to +\infty, whereas the argument of \Phi_0 and \tilde{\Phi}_0 in (5.67), (5.68), (5.75), (5.76) is X = \lambda^{(4k+3)/3}\kappa_0 (respectively X = \lambda^{(4k+3)/3} f_3(r; r_0)), which takes values down to 0 at the point \alpha_k s^{2k+1} + x = 0 inside the transition region; no displayed argument supplies a uniform bound for |X| = O(1) or X \to 0 with constants independent of \kappa_0. The manuscript should complete this uniformity argument (for example by a three-range estimate using analyticity of \Phi_0 in X together with (B.20)), or restate Lemma B.4 as a uniform-in-X statement. Without this, the extraction of -\chi(0) - \log(2k+1)/24 - \log \alpha_k/(24(2k+1)) in (2.16)-(2.18) is not fully justified. This is a load-bearing gap, not a stylistic one.
  2. [§5.1, Eq. (5.19) and Lemma 5.1] There appears to be an internal inconsistency in the size of the local parametrix radius \rho_2 in the case where \lambda^{(4k+3)/3}\kappa_0 is bounded. Formula (5.19) sets \rho_2 = 1/(\lambda^{(4k+3)/2}|\kappa_0|) + \lambda^{2k/3+1/2}; for |\kappa_0| \le C\lambda^{-(4k+3)/3} the first term is at least of order \lambda^{(4k+3)/6} and the second term is exactly \lambda^{(4k+3)/6}, so \rho_2 \to +\infty. In the proof of Lemma 5.1, however, the same case is said to give \rho_2 behaving like C(\lambda)\lambda^{-2k/3-1/2}, which tends to 0. These two statements are incompatible, and the estimates in (5.25)-(5.28) and in Lemma 5.2 depend on the actual size of \rho_2 (through (5.23) and the factors (\lambda\rho_2)^{\pm\sigma_3/4}). Since this matching feeds into Lemma 5.2 and hence into the bound (5.70) used for Lemma 2.3, the authors should resolve the discrepancy and rerun the estimates with the correct \rho_2.
  3. [Appendix B, Lemmas B.1, B.2, B.4] Lemmas B.1 and B.2 are essential inputs for Lemma 2.3, yet their proofs are explicitly sketches, and the version of the P34 asymptotics actually needed in Section 5 is not literally the one stated: the applications require (B.1) and (B.12) to hold uniformly over a two-parameter regime in which the size of x = \lambda^{(4k+3)/3}\kappa_0 ranges from 0 to +\infty and the ratio |\zeta|/|x| ranges near the threshold L, with uniform constants; the manuscript neither proves nor cites such strengthened statements. Lemma B.4 similarly covers only x \to +\infty. The authors should provide full proofs, or precise reference results that cover the required uniformity, and should verify the hypotheses of Lemmas B.1 and B.2 on the actual contours of Section 5 (for example on \partial U(r; \rho_2) with \rho_2 as in (5.19)).
  4. No additional major comments.
minor comments (3)
  1. [Abstract and Section 1] The name is misspelled as 'Clayes' in the abstract (twice) and in the Introduction; it should be 'Claeys', Its and Krasovsky.
  2. [§5.3, text before Eq. (5.71)] The notation 'large \lambda behavior of \Phi_0(-\lambda)' is imprecise: the argument of \Phi_0 in (5.67)-(5.68) is \lambda^{(4k+3)/3}\kappa_0, not \lambda. Since the uniformity in \kappa_0 is precisely the delicate point, the argument should be written out explicitly.
  3. [Remark 2.5] The remark correctly flags the transition-region error as the delicate point; it would be helpful to point the reader to the specific estimates (5.71) and (5.78) and to state that uniformity in \kappa_0 is claimed there, since that is what makes the integrated error small.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the multiplicative constant is derived from an independent Riemann–Hilbert analysis, with external inputs for the Airy constant and the Painlevé Hamiltonian.

full rationale

The paper's central claim, Theorem 1.1, is an explicit large-gap asymptotic constant for higher-order Tracy–Widom determinants. The derivation does not fit any parameter to the quantity being predicted. The constant is assembled from: (i) an independent Riemann–Hilbert asymptotic analysis of ∂F/∂s and ∂F/∂x in the algebraic, transition, and exponential regions (Lemmas 2.1–2.4); (ii) the known Airy-kernel constant χ(0), imported from external literature [1,27]; and (iii) the Hamiltonian h of the special P_I^{2k} solution, whose relation to the RH problem is established in [14,17] and whose large-x asymptotics are proved in Appendix C. The transition-region contribution to χ(0) comes from integrating the Painlevé II Hamiltonian against the asymptotic expansion of the Airy determinant, using the external identity (A.7) and (1.15). This is independent support, not self-citation. Self-citations such as [21] are used only to record the k=1 special case of the total-integral corollary, not as input for the general theorem. The reviewer's flagged concern — that the O(|x|^{(k-1/3)/(2k+1)}) error in Lemma 2.3 is controlled only after integration and that uniformity in the transition region is delicate — is a mathematical correctness or gap risk, not circularity: it concerns whether the stated asymptotic error bounds are valid uniformly, not whether the conclusion is assumed by construction. No equation in the paper defines the predicted constant in terms of itself, and no fitted quantity is relabeled as a prediction. The score is therefore 1, reflecting only a minor self-citation in the background that is not load-bearing.

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

The paper introduces no fitted constants and no new physical entities. Its free-parameter count is zero. The central claim rests on established theorems about Painlevé hierarchies, RH solvability, and the Airy determinant constant, all from the cited literature, plus the paper's own lengthy asymptotic analysis. The only internal mathematical tool introduced beyond existing ones is the uniform P34 parametrix variant, which is a construction, not a postulated entity.

assumptions (6)
  • standard math There exists a real pole-free solution q of the even Painlevé I hierarchy P_I^{2k} satisfying the boundary condition (1.27), and its Hamiltonian h has the asymptotic behavior (1.32).
    Invoked in Theorem 1.1 and Appendix C to define I_h(x) and to justify the convergence of the integral. Established in the cited works [14, 19, 49].
  • standard math The Riemann-Hilbert problems for Psi, Y and X are uniquely solvable and the differential identities (3.14) and (3.21) hold.
    This is the bridge between the Fredholm determinant and the RH analysis, taken from the general framework of Deift, Its and Zhou and from Claeys, Its and Krasovsky.
  • standard math The classical Airy-kernel determinant has the known large gap asymptotics with constant chi(0) from (1.15).
    Used in the proof of Theorem 1.1 through (2.16), where the transition integral is expressed in terms of the Airy determinant asymptotic constant. This is an external benchmark proved in [1, 27].
  • standard math The P34 parametrix RH problem is solvable and relates to the Painlevé II Hamiltonian through (A.5)-(A.7).
    Used as the local parametrix in the transition region; the connection to the Airy determinant is from the literature on P34 and is assumed in Appendix A.
  • standard math The g-function g_2 built in (5.38) satisfies the sign and decay properties from Claeys (2012), Proposition 2.5 and Corollary 2.6.
    Used in Section 5.2 to ensure exponential decay of the jump matrices away from the local parametrix disk.
  • standard math Standard Deift-Zhou nonlinear steepest descent and small-norm RH theory applies to the constructed parametrices.
    The paper repeatedly invokes the standard small-norm argument to control the final RH problems; this is background mathematical technology.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The multiplicative constant in asymptotics of higher-order analogues of the Tracy-Widom distribution." pith.science (2026). https://pith.science/paper/4GHHX24O

@misc{pith2026250112679,
  author       = {Pith},
  title        = {Pith review of: The multiplicative constant in asymptotics of higher-order analogues of the Tracy-Widom distribution},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4GHHX24O}},
  note         = {Machine review of arXiv:2501.12679}
}
abstract

In this paper, we are concerned with higher-order analogues of the Tracy-Widom distribution, which describe the eigenvalue distributions in unitary random matrix models near critical edge points. The associated kernels are constructed by functions related to the even members of the Painlev\'{e} I hierarchy $\mathrm{P_{I}^{2k}}, k\in\mathbb{N}^{+}$, and are regarded as higher-order analogues of the Airy kernel. We present a novel approach to establish the multiplicative constant in the large gap asymptotics of the distribution, resolving an open problem in the work of Clayes, Its and Krasovsky. An important new feature of the expression is the involvement of an integral of the Hamiltonian associated with a special, real, pole-free solution for $\mathrm{P_{I}^{2k}}$. In addition, we show that the total integral of the Hamiltonian vanishes for all $k$, and establish a transition from the higher-order Tracy-Widom distribution to the classical one in the asymptotic regime. Our approach can also be adapted to calculate similar critical constants in other problems arising from mathematical physics.

Figures

Figures reproduced from arXiv: 2501.12679 by the authors.

Figure 1
Figure 1. The three regions in asymptotic studies of [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. The contours of integration in the (x, s)-plane. We choose the red lines if x0 > 0 and the blue lines if x0 < 0. The dashed curve is the critical curve x = −αks 2k+1, where αk is given in (1.21). In the proof of Theorem 1.1, it suffices to choose either the red or the blue lines of integration depicted in [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. The jump contours Γj , j = 1, 2, 3, 4, of the RH problem for Ψ. Recalling the definition of s1 in (2.11), when x0 = |s| 2k+1 > 0, it is straightforward to see that α 1 2k+1 k s1 = −|s| − |s| −k+ 1 6 2k + 1 + k |s| −2k− 2 3 (2k + 1)2 − k(4k + 1) 3(2k + 1)3 |s| −3k− 3 2 + O  |s| −4k− 7 3  . (2.22) as s → −∞. Finally, by combining the above three formulas, we arrive at the desired approxi￾mation in (2.19). This finis… view at source ↗
Figures from the paper (14 more)
Figure 4
Figure 4. Figure 4: Regions I–V and jump contours of the RH problem for [PITH_FULL_IMAGE:figures/full_fig_p018_4.png]
Figure 5
Figure 5. Figure 5: The jump contour ΣT of the RH problem for T. Global parametrix The global parametrix T (∞) reads as follows. RH problem for T (∞) (a) T (∞) (η) is analytic in C \ (−∞, 0]. (b) T (∞) satisfies the jump condition T (∞) + (η) = T (∞) − (η)  0 1 −1 0 , η ∈ (−∞, 0). (4.25…
Figure 6
Figure 6. Figure 6: The jump contour ΣR of the RH problem for R in Section 4.2. RH problem for R (a) R(η) is analytic in C \ Σ R, where ΣR := ΣT 1 ∪ Σ T 3 ∪ ∂U(0; ρ1) \ U(0; ρ1) is depicted in [PITH_FULL_IMAGE:figures/full_fig_p025_6.png]
Figure 7
Figure 7. Figure 7: The jump contour ΣP of RH problem for P. According to Lemma 4.4, there exist two positive constants M2 and C2 such that for η ∈ Σ P 1 ∪ Σ P 3 , λ 4k+3 2 Re g1(η) ≤ −M2λ 4k+3 2 |η − r| 3 2 ≤ −C2λ k 2 , (5.4) uniformly for |η − r| ≥ λ −k−1 and λ large enough. As a conseq…
Figure 8
Figure 8. Figure 8: The jump contour ΣR of the RH problem for R in Section 5.1. RH problem for R (a) R(η) is analytic in C \ Σ R, where ΣR = [PITH_FULL_IMAGE:figures/full_fig_p033_8.png]
Figure 9
Figure 9. Figure 9: Regions I and II and the jump contours of the RH problem for [PITH_FULL_IMAGE:figures/full_fig_p035_9.png]
Figure 10
Figure 10. Figure 10: The jump contour of the RH problem for R in Section 5.2 To solve this RH problem, we introduce the function f3(η; r0) =  3 2 g2(η) 2 3 , (5.54) where g2(η) is defined in (5.38). Clearly, f3(η; r0) is analytic in U(r0; ˜ρ2). Indeed, we have f3(η; r0) =  3p2(r0) 2 2…
Figure 11
Figure 11. Figure 11: The jump contours of the RH problem for Q, where r0 is defined in (5.40). (c) As η → r, we have Q(η) = O(log (η − r)). (d) As η → ∞, we have Q(η) = I + O(η −1 )  (λη) − 1 4 σ3N, (6.2) where N is defined in (3.4). As mentioned in Section 5.2, we know that Re g2(η) < 0…
Figure 12
Figure 12. Figure 12: The jump contour of the RH problem for R in Section 6.1. When η ∈ Γ R \ ∂U(r0; ρ3), we have from (6.3) and (6.4) that vR(η) = I + O(λ k−δ 2 e −M4λ k 2 + 3 2 δ ), λ → +∞, (6.24) for some constant M4 > 0. For η ∈ ∂U(r0; ρ3), substituting the large z asymptotic expansion…
Figure 13
Figure 13. Figure 13: The jump contour Γ of the RH problem for Φ. [PITH_FULL_IMAGE:figures/full_fig_p045_13.png]
Figure 14
Figure 14. Figure 14: The jump contour Σ of the RH problem for ˜ Φ. ˜ (c) As ζ → ∞, we have Φ( ˜ ζ; x) = I + O(ζ −1 )  ζ − 1 4 σ3 I + iσ1 √ 2 e − 2 3 ζ 3 2 σ3 , (A.10) uniformly for any positive bounded x. (d) As ζ → x, we have Φ( ˜ ζ; x) = Φ˜(0)(ζ; x)  1 log (ζ−x) 2πi 0 1  , (A.11) and…
Figure 15
Figure 15. Figure 15: The jump contour of the RH problem for R in the proof of Lemma B.1. as |x| → +∞, where Rj (η) is analytic in C \ ∂U(0; ρ). In particular, the structures of Jj in (B.8) imply that for η ∈ C \ U(0; ρ), Ri , i = 1, 2, 3, takes the following form: R1(η) =  0 ⋆ 0 0 η , R…
Figure 16
Figure 16. Figure 16: The jump contour of the RH problem for R in the proof of Lemma B.2. Here, Φ(Ai) is the Airy parametrix characterized by the jump condition (6.14) and the asymptotic behavior (6.15), and C2(η; x) = 1 2πi Z 1 1 2 e − 4 3 (xξ) 3 2 ξ − η dξ, η ∈ C \ h 1 2 , 1 i . (B.17) T…
Figure 17
Figure 17. Figure 17: The jump contour of the RH problem for S in Appendix C C Asymptotics of the Hamiltonian h Let h(x) = h(x, t1, . . . , t2k−1) be the Hamiltonian associated with the special solution q of the Painlev´e I hierarchy P2k I . Following the asymptotics analysis carried out i…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 66 canonical work pages

  1. [21]

    Dai and W.-G

    D. Dai and W.-G. Long, Asymptotics and total integrals of the P 2 I tritronqu´ ee solution and its Hamiltonian, SIAM J. Math. Anal. 56 (2024), 5350–5371

  2. [1]

    J. Baik, R. Buckingham and J. DiFranco, Asymptotics of Tracy-Widom distributions and the total integral of a Painlev´ e II function, Comm. Math. Phys. 280 (2008), 463–497

  3. [2]

    Benjamin and I

    F. Benjamin and I. Krasovsky, Sine-kernel determinant on two large intervals, Comm. Pure Appl. Math. 77 (2024), 1958–2029

  4. [3]

    Bleher and A

    P. Bleher and A. Its, Semiclassical asymptotics of orthogonal polynomials, Riemann-Hilbert problem, and universality in the matrix model, Ann. Math. 150 (1999), 185–266

  5. [4]

    Bogatskiy, T

    A. Bogatskiy, T. Claeys and A. Its, Hankel determinant and orthogonal polynomials for a Gaussian weight with a discontinuity at the edge, Comm. Math. Phys. 347 (2016), 127–162

  6. [5]

    Bothner and A

    T. Bothner and A. Its, Asymptotics of a Fredholm determinant corresponding to the first bulk critical universality class in random matrix models, Comm. Math. Phys. 328 (2014), 155–202

  7. [6]

    M. J. Bowick and E. Br´ ezin, Universal scaling of the tail of the density of eigenvalues in random matrix models, Phys. Lett. B 268 (1991), 21–28. 55

  8. [7]

    Br´ ezin, E

    E. Br´ ezin, E. Marinar and G. Parisi, A nonperturbative ambiguity free solution of a string model, Phys. Lett. B 242 (1990), 35–38

Show all 67 references
  1. [8]

    Byun and S

    S.-S. Byun and S. Park, Large gap probabilities of complex and symplectic spherical en- sembles with point charges, preprint arXiv:2405.00386

  2. [9]

    Cafasso, T

    M. Cafasso, T. Claeys and M. Girotti, Fredholm determinant solutions of the Painlev´ e II hierarchy and gap probabilities of determinantal point processes, Int. Math. Res. Not. 2021 (2021), 2437–2478

  3. [10]

    Charlier, Large gap asymptotics on annuli in the random normal matrix model, Math

    C. Charlier, Large gap asymptotics on annuli in the random normal matrix model, Math. Ann. 388 (2024), 3529–3587

  4. [11]

    Charlier, J

    C. Charlier, J. Lenells and J. Mauersberger, Higher order large gap asymptotics at the hard edge for Muttalib-Borodin ensembles, Comm. Math. Phys. 384 (2021), 829–907

  5. [12]

    Charlier, J

    C. Charlier, J. Lenells and J. Mauersberger, The multiplicative constant for the Meijer- G kernel determinant, Nonlinearity 34 (2021), 2837–2877

  6. [13]

    Claeys, Universality in critical random matrix ensembles and pole-free solutions of Painlev´ e equations, Thesis of doctor degree in Katholieke Universiteit Leuven, 2006

    T. Claeys, Universality in critical random matrix ensembles and pole-free solutions of Painlev´ e equations, Thesis of doctor degree in Katholieke Universiteit Leuven, 2006

  7. [14]

    Claeys, Pole-free solutions of the first Painlev´ e hierarchy and non-generic critical be- havior for the KdV equation, Phys

    T. Claeys, Pole-free solutions of the first Painlev´ e hierarchy and non-generic critical be- havior for the KdV equation, Phys. D 241 (2012), 2226–2236

  8. [15]

    Claeys and T

    T. Claeys and T. Grava, Universality of the break-up profile for the KdV equation in the small dispersion limit using the Riemann–Hilbert approach, Comm. Math. Phys. 286 (2009), 979–1009

  9. [16]

    Claeys and T

    T. Claeys and T. Grava, The KdV hierarchy: universality and a Painlev´ e transcendent, Int. Math. Res. Not. 2012 (2012), 5063–5099

  10. [17]

    Claeys, A

    T. Claeys, A. Its and I. Krasovsky, Higher-order analogues of the Tracy-Widom distribution and the Painlev´ e II hierarchy, Comm. Pure Appl. Math. 63 (2010), 362–412

  11. [18]

    Claeys and M

    T. Claeys and M. Vanlessen, Universality of a double scaling limit near singular edge points in random matrix models, Comm. Math. Phys. 273 (2007), 499–532

  12. [19]

    Claeys and M

    T. Claeys and M. Vanlessen, The existence of a real pole-free solution of the fourth order analogue of the Painlev´ e I equation, Nonlinearity 20 (2007), 1163–1184

  13. [20]

    Corwin, P

    I. Corwin, P. Deift and A. Its, Harold Widom’s work in random matrix theory, Bull. Amer. Math. Soc. (N.S.) 59 (2022), 155–173

  14. [22]

    Dai, S.-X

    D. Dai, S.-X. Xu and L. Zhang, Gap probability at the hard edge for random matrix ensembles with pole singularities in the potential, SIAM J. Math. Anal. 50 (2018), 2233– 2279

  15. [23]

    Dai, S.-X

    D. Dai, S.-X. Xu and L. Zhang, Asymptotics of Fredholm determinant associated with the Pearcey kernel, Comm. Math. Phys. 382 (2021), 1769–1809

  16. [24]

    Deift, Orthogonal Polynomials and Random Matrices: A Riemann-Hilbert Approach, Courant Lecture Notes, vol

    P. Deift, Orthogonal Polynomials and Random Matrices: A Riemann-Hilbert Approach, Courant Lecture Notes, vol. 3, New York University, 1999. 56

  17. [25]

    Deift and D

    P. Deift and D. Gioev, Universality at the edge of the spectrum for unitary, orthogonal, and symplectic ensembles of random matrices, Comm. Pure Appl. Math. 60 (2007), 867–910

  18. [26]

    Deift, A

    P. Deift, A. Its and I. Krasovsky, Toeplitz matrices and Toeplitz determinants under the impetus of the Ising model: some history and some recent results, Comm. Pure Appl. Math. 66 (2013), 1360–1438

  19. [27]

    Deift, A

    P. Deift, A. Its and I. Krasovsky, Asymptotics of the Airy-kernel determinant, Comm. Math. Phys. 278 (2008), 643–678

  20. [28]

    Deift, A

    P. Deift, A. Its, I. Krasovsky and X. Zhou, The Widom-Dyson constant for the gap prob- ability in random matrix theory, J. Comput. Appl. Math. 202 (2007), 26–47

  21. [29]

    Deift, A

    P. Deift, A. Its and X. Zhou, A Riemann-Hilbert approach to asymptotic problems arising in the theory of random matrix models, and also in the theory of integrable statistical mechanics, Ann. Math. 146 (1997), 149–235

  22. [30]

    Deift, I

    P. Deift, I. Krasovsky and J. Vasilevska, Asymptotics for a determinant with a confluent hypergeometric kernel, Int. Math. Res. Notices 2011 (2011), 2117–2160

  23. [31]

    Deift, T

    P. Deift, T. Kriecherbauer and K.T-R McLaughlin, New results on the equilibrium measure for logarithmic potentials in the presence of an external field, J. Approx. Theory 95 (1998), 388–475

  24. [32]

    Deift, T

    P. Deift, T. Kriecherbauer, K. T.-R. McLaughlin, S. Venakides and X. Zhou, Uniform asymptotics for polynomials orthogonal with respect to varying exponential weights and applications to universality questions in random matrix theory, Comm. Pure Appl. Math. 52 (1999), 1335–1425

  25. [33]

    Deift and X

    P. Deift and X. Zhou, A steepest descent method for oscillatory Riemann-Hilbert problems, Ann. Math. 137 (1993), 295–368

  26. [34]

    Dubrovin, On Hamiltonian perturbations of hyperbolic systems of conservation laws, II: Universality of critical behavior, Comm

    B. Dubrovin, On Hamiltonian perturbations of hyperbolic systems of conservation laws, II: Universality of critical behavior, Comm. Math. Phys. 267 (2006), 117–139

  27. [35]

    Dubrovin, On Universality of Critical Behavior in Hamiltonian PDEs, Geometry, Topol- ogy, and Mathematical Physics, American Mathematical Society Translation Series 2, vol

    B. Dubrovin, On Universality of Critical Behavior in Hamiltonian PDEs, Geometry, Topol- ogy, and Mathematical Physics, American Mathematical Society Translation Series 2, vol. 224, American Mathematical Society, Providence, (2008), 59–109

  28. [36]

    Dubrovin, T

    B. Dubrovin, T. Grava and C. Klein, On universality of critical behavior in the focusing nonlinear Schr¨ odinger equation, elliptic umbilic catastrophe and the tritronqu´ ee solution to the Painlev´ e-I equation, J. Nonlinear Sci. 19 (2009), 57–94

  29. [37]

    Dyson, Fredholm determinants and inverse scattering problems, Comm

    F. Dyson, Fredholm determinants and inverse scattering problems, Comm. Math. Phys. 47 (1976), 171–183

  30. [38]

    Ehrhardt, Dyson’s constant in the asymptotics of the Fredholm determinant of the sine kernel, Comm

    T. Ehrhardt, Dyson’s constant in the asymptotics of the Fredholm determinant of the sine kernel, Comm. Math. Phys. 262 (2006), 317–341

  31. [39]

    Ehrhardt, The asymptotics of a Bessel-kernel determinant which arises in random matrix theory, Adv

    T. Ehrhardt, The asymptotics of a Bessel-kernel determinant which arises in random matrix theory, Adv. Math. 225 (2010), 3088–3133

  32. [40]

    P. J. Forrester and N. S. Witte, Application of the τ -function theory of Painlev´ e equations to random matrices: PIV, PII and the GUE, Commun. Math. Phys. 219 (2001), 357–398. 57

  33. [41]

    Gordoa and A

    R. Gordoa and A. Pickering, Nonisospectral scattering problems: a key to integrable hier- archies, J. Math. Phys. 40 (1999), 5749–5786

  34. [42]

    Grava, A

    T. Grava, A. Kapaev and C. Klein, On the tritronqu´ ee solutions of P 2 I , Constr. Approx. 41 (2015), 425–466

  35. [43]

    S. P. Hastings and J. B. McLeod, A boundary value problem associated with the second Painlev´ e transcendent and the Korteweg-de Vries equation, Arch. Rational Mech. Anal. 73 (1980), 31–51

  36. [44]

    A. R. Its, A. B. J. Kuijlaars and J. ¨Ostensson, Critical edge behavior in unitary random matrix ensembles and the thirty fourth Painlev´ e transcendent, Int. Math. Res. Not. 2008 (2008), rnn017, 67pp

  37. [45]

    A. R. Its, A. B. J. Kuijlaars and J. ¨Ostensson, Asymptotics for a special solution of the thirty fourth Painlev´ e equation, Nonlinearity 22 (2009), 1523–1558

  38. [46]

    A. Its, O. Lisovyy and A. Prokhorov, Monodromy dependence and connection formulae for isomonodromic tau functions, Duke Math. J. 167 (2018), 1347–1432

  39. [47]

    Its and A

    A. Its and A. Prokhorov, Connection problem for the tau-function of the sine-Gordon reduction of Painlev´ e-III equation via the Riemann-Hilbert approach, Int. Math. Res. Not. 2016 (2016), 6856–6883

  40. [48]

    A. R. Its and A. Prokhorov, On some Hamiltonian properties of the isomonodromic tau functions, Rev. Math. Phys. 30 (2018), 1840008, 38 pp

  41. [49]

    A. A. Kapaev, Weakly nonlinear solutions of equation P 2 I , J. Math. Sci. 73 (1995), 468–481

  42. [50]

    Krasovsky, Large gap asymptotics for random matrices, XVth International Congress on Mathematical Physics, New Trends in Mathematical Physics, Springer, 2009, 413–419

    I. Krasovsky, Large gap asymptotics for random matrices, XVth International Congress on Mathematical Physics, New Trends in Mathematical Physics, Springer, 2009, 413–419

  43. [51]

    Krasovsky, Gap probability in the spectrum of random matrices and asymptotics of polynomials orthogonal on an arc of the unit circle, Int

    I. Krasovsky, Gap probability in the spectrum of random matrices and asymptotics of polynomials orthogonal on an arc of the unit circle, Int. Math. Res. Not. 2004 (2004), 1249–1272

  44. [52]

    Krasovsky and T

    I. Krasovsky and T. H. Maroudas, Airy-kernel determinant on two large intervals, Adv. Math. 440 (2024), Paper No. 109505, 79 pp

  45. [53]

    N. A. Kudryashov, The first and second Painlev´ e equations of higher order and some relations between them, Phys. Lett. A 224 (1997), 353–360

  46. [54]

    A. B. J. Kuijlaars and K. T.-R. McLaughlin, Generic behavior of the density of states in random matrix theory and equilibrium problems in the presence of real analytic external fields, Comm. Pure Appl. Math. 53 (2000), 736–785

  47. [55]

    A. B. J. Kuijlaars, K. T-R. McLaughlin, W. Van Assche and M. Vanlessen, The Riemann- Hilbert approach to strong asymptotics for orthogonal polynomials on [ −1, 1], Adv. Math. 188 (2004), 337–398

  48. [56]

    A. B. J. Kuijlaars and M. Vanlessen, Universality for eigenvalue correlations from the modified Jacobi unitary ensemble, Int. Math. Res. Not. 2002 (2002), 1575–1600

  49. [57]

    M. L. Mehta, Random Matrices, 2nd edn, Academic Press, Boston, 1991. 58

  50. [58]

    Mugan and F

    U. Mugan and F. Jrad, Painlev´ e test and the first Painlev´ e hierarchy, J. Phys. A: Math. Gen. 32 (1999), 7933–7952

  51. [59]

    Pastur and M

    L. Pastur and M. Shcherbina, Universality of the local eigenvalue statistics for a class of unitary invariant random matrix ensembles, J. Stat. Phys. 86 (1997), 109–147

  52. [60]

    E. B. Saff and V. Totik, Logarithmic Potentials with External Fields, Springer-Verlag, New York, 1997

  53. [61]

    Shimomura, A certain expression for the first Painlev´ e hierarchy, Proc

    R. Shimomura, A certain expression for the first Painlev´ e hierarchy, Proc. Japan Acad. Ser. A 80 (2004), 105–109

  54. [62]

    C. A. Tracy and H. Widom, Level-spacing distributions and the Airy kernel, Comm. Math. Phys. 159 (1994), 151–174

  55. [63]

    Vanlessen, Strong asymptotics of Laguerre-type orthogonal polynomials and applica- tions in random matrix theory, Constr

    M. Vanlessen, Strong asymptotics of Laguerre-type orthogonal polynomials and applica- tions in random matrix theory, Constr. Approx. 25 (2007), 125–175

  56. [64]

    Widom, The strong Szeg˝ o limit theorem for circular arcs, Indiana Univ

    H. Widom, The strong Szeg˝ o limit theorem for circular arcs, Indiana Univ. Math. J. 21 (1971), 277–283

  57. [65]

    Xu and D

    S.-X. Xu and D. Dai, Tracy-Widom distributions in critical unitary random matrix ensem- bles and the coupled Painlev´ e II system, Comm. Math. Phys. 365 (2019), 515–567

  58. [66]

    Xu and Y

    S.-X. Xu and Y. Q. Zhao, Painlev´ e XXXIV asymptotics of orthogonal polynomials for the Gaussian weight with a jump at the edge, Stud. Appl. Math. 127 (2011), 67–105

  59. [67]

    Yao and L

    L.-M. Yao and L. Zhang, Asymptotics of the hard edge Pearcey determinant, SIAM J. Math. Anal. 56 (2024), 137–172. 59

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.