Pith. sign in

REVIEW 3 major objections 5 minor 30 references

Long-Time Dynamics and Modified Scattering for the Coulomb Vlasov--Hartree System

T0 review · 3 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper claims that the Coulomb Vlasov–Hartree system, a classical–quantum model of a Bose–Fermi mixture, has global smooth solutions for small localized data and that its long-time behavior is modified scattering in which the…

desk verdict First modified scattering for the coupled Coulomb Vlasov–Hartree system, with a genuine but easily patched regularity gap in the main theorem. read the letter →

arxiv 2608.04402 v1 pith:4M64EEMJ submitted 2026-08-05 math.AP

Wenrui Huang , Mengyi Xie This is my paper · ORCID
classification math.AP MSC 35Q8335Q5535B40
keywords Vlasov–HartreesystemCoulombinteractionmodifiedscatteringlong-timeasymptoticsBose–Fermimixtureglobalwell-posednessGalileanvectorfieldslogarithmicphasecorrection
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 studies the Vlasov–Hartree system in three dimensions with Coulomb interaction, a mean-field model of a Bose–Fermi mixture in which the bosonic density generates a force on the fermions and the fermionic density generates a potential for the bosons. For sufficiently small, regular, and localized initial data it proves global well-posedness and, more importantly, identifies the exact long-time asymptotics. The central claim is modified scattering with the coupling visible at leading order: the fermionic distribution converges along characteristics shifted by $\log(t) F_\infty(v)$, where $F_\infty$ is built from the limiting bosonic profile, while the bosonic wave acquires the phase $e^{-i \log(t)\phi_\infty(x/t)}$, with $\phi_\infty$ built from the limiting fermionic density. A sympathetic reading is that the two components never decouple: each species' memory of the other survives in the asymptotic data at order one.

What carries the argument

The proof is carried out in the common self-similar variable $y=x/t$. For the Coulomb potential it uses an exact shell decomposition $1/|z|=C_0\int_0^\infty \chi(z/R)\,dR/R^2$, which lets the physical field $E(t,x)$ be compared with an effective field $E^0(t,x/t)$ whose evolution can be controlled through the transport equation for the free-transport profile $\gamma(t,x,v)=f(t,x+tv,v)$. On the Schrödinger side the central objects are the Galilean vector field $G=ix+2t\nabla$ and the renormalized profile $h(t,y)=t^{3/2}e^{i|y|^2t/4}u(t,ty)e^{i\theta(t,y)}$, where $\theta$ removes the growing Coulomb phase. The key structural fact is that both components propagate in the same variable: convergence of $h$ to $h_\infty$ gives the limiting force $F_\infty=-\nabla(V*|h_\infty|^2)$ that enters the Vlasov characteristics, and convergence of $\gamma$ gives the density $\rho_\infty$ that defines the phase $\phi_\infty=V*\rho_\infty$ in the Hartree law.

What would settle it

Solve (1.1) numerically in three dimensions for small Gaussian data with $V(x)=\lambda|x|^{-1}$ and check whether $\tilde\gamma(t,x,v)=\gamma(t,x-\log(t)F_\infty(v),v)$ converges in $L^\infty_{x,v}$ with the stated $t^{-\delta_0}$ rate and whether the rescaled Schrödinger profile, after removing the phase $e^{-i\log(t)\phi_\infty(x/t)}$, converges in $L^\infty_x$; failure of either for arbitrarily small data would refute Theorem 1.1. Alternatively, finding data satisfying (1.7) but not $C^1$ for which the system has no strong solution would refute the theorem's regularity statement.

Watch

Extended reading notes

Core claim

With $V(x)=\lambda|x|^{-1}$, the paper proves (Theorem 1.1) that for initial data satisfying the smallness condition (1.7) there is a unique global strong solution, and that as $t\to\infty$ the free-transport profile obeys $f(t, x+tv-\log(t)F_\infty(v), v) = f_\infty(x,v) + O_{L^\infty_{x,v}}(t^{-\delta_0})$, while the Hartree wave obeys $u(t,x) = t^{-3/2} e^{-i|x|^2/(4t)} e^{-i\log(t)\phi_\infty(x/t)} h_\infty(x/t) + O_{L^\infty_x}(t^{-3/2-\delta_0})$. Here $\rho_\infty(v)=\int f_\infty^2(x,v)\,dx$, $\phi_\infty = V*\rho_\infty$, and $F_\infty = -\nabla(V*|h_\infty|^2)$, so the asymptotic fermionic state sets the bosonic phase and the asymptotic bosonic state bends the fermionic trajectories. The same coupled argument, with integrable time weights, yields ordinary scattering for inverse-power interactions $V(x)=\lambda|x|^{-\alpha}$, $1<\alpha<3/2$, with rates $t^{1-\alpha}$.

Load-bearing premise

The whole long-time picture rests on the small-data bootstrap closing with the logarithmic growth rates stated in Proposition 3.10, on the validity of the unpublished companion's physical-space Hartree estimates [25] used as a black box, and on smallness in (1.7) implying the $C^1$ regularity that the local well-posedness argument requires.

Editorial extensions

If this is right

  • The asymptotic state of the system is genuinely coupled: $f_\infty$ and $h_\infty$ cannot be read off from the free evolution of the initial data, because each carries a leading-order imprint of the other.
  • The fermionic sector satisfies a modified scattering law along the corrected characteristics $x+tv-\log(t)F_\infty(v)$, so the long-time force on the fermions is entirely described by the limiting bosonic profile.
  • The bosonic sector obeys $u(t,x)\approx t^{-3/2}e^{-i|x|^2/(4t)}e^{-i\log(t)\phi_\infty(x/t)}h_\infty(x/t)$, meaning the fermionic density acts as a logarithmic phase clock on the condensate.
  • For $1<\alpha<3/2$, the same bootstrap yields ordinary (unmodified) scattering with the explicit rates (1.11)–(1.12), confirming that $\alpha=3/2$ is the borderline for long-range behavior in this coupled setting.
  • Conservation laws imply the constructed asymptotic profiles preserve the initial fermionic and bosonic masses, so the scattering map, once defined, will be mass-preserving.

Reading between the lines

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

  • A natural next step, not taken in the paper, is to construct the modified wave operator: Theorem 1.1 gives the asymptotic pair $(f_\infty,h_\infty)$ that a direct scattering map would need to hit, and the effective fields $F_\infty,\phi_\infty$ are exactly the data such an operator would invert.
  • The common self-similar variable $y=x/t$ suggests the method may carry over to other kinetic–dispersive systems with borderline long-range interactions, such as Vlasov–Nordström or Vlasov–Schrödinger with gravitational coupling, where a similar two-sided log correction should appear.
  • The restriction $\alpha<3/2$ is technical (an endpoint Sobolev embedding); a sharper embedding or a different norm might push the short-range result up to $\alpha=2$, and the failure at $\alpha=3/2$ would then be a genuine threshold rather than an artifact.
  • One testable consequence of the coupling is that the asymptotic phase $\phi_\infty$ should equal the Coulomb potential of the limiting fermionic spatial density, equation (4.24); measuring the phase difference in a numerical two-species simulation would verify the self-consistent closure of the asymptotic data.
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

3 major / 5 minor

Summary. The paper studies the three-dimensional Vlasov--Hartree system with Coulomb interaction, in which the Vlasov force is generated by the bosonic density |u|^2 and the Hartree potential by the fermionic density \int f^2 dv. The main theorem (Theorem 1.1) asserts that sufficiently small, localized, and regular data produce a unique global strong solution and that both components exhibit modified scattering: f(t, x+tv-\log t F_\infty(v), v) -> f_\infty in L^\infty, while u(t,x) is asymptotic to t^{-3/2} e^{-i|x|^2/4t} e^{-i\log t \phi_\infty(x/t)} h_\infty(x/t). Theorem 1.2 states a short-range analogue with ordinary scattering for interactions |x|^{-\alpha}, 1<\alpha<3/2. The proof couples a bootstrap for the Vlasov free-transport profile \gamma with Galilean vector-field estimates for the Hartree component and derives the asymptotic profiles by constructing limits rather than assuming them.

Significance. If the main theorem is correct, this is the first sharp long-time asymptotic description for this coupled kinetic-dispersive model, and it shows that the two components interact at leading order through logarithmic phase and characteristic corrections. The proof is largely self-contained and contains parameter-free constructions: the asymptotic profiles are limits of the evolution, not fitted objects, and the Lorentz-space estimates used are mostly standard. The extension to short-range interactions is a useful additional result. However, the regularity gap in Theorem 1.1, the sign/phase inconsistency in Section 4.2, and the incomplete energy estimates in Section 3.2 must be fixed before the claims are fully supported.

major comments (3)
  1. [Section 1, Theorem 1.1; Section 3.1, Proposition 3.1] The hypotheses of Theorem 1.1 do not imply the regularity required by the local well-posedness theory. Assumption (1.7) controls \nabla_{x,v} f0 only in L^2 \cap L^\infty and with a weight in L^\infty; this does not imply that f0 is C^1, whereas Proposition 3.1 requires f0 \in C^1 and yields f \in C^1 on a short interval. Since the proof neither approximates f0 by C^1 data nor supplies a separate local existence argument under (1.7), the global C^1 statement in Theorem 1.1 is not justified. This is a genuine gap in the theorem as stated, although it is localized: adding f0 \in C^1 to (1.7), or weakening the conclusion for f to W^{1,\infty} \cap W^{1,2} in (x,v), would close it.
  2. [Section 4.2, Eqs. (4.3)-(4.7) and the paragraph preceding (1.10)] The proof of the Hartree asymptotic is inconsistent with the definition of h_\infty and with the sign of the phase in (1.10). Since h(t,y)=g(t,y)e^{i\theta(t,y)} and \theta(t,y)=l(y)+\log(t)\phi_\infty(y)+O(t^{-1/40}), the correct relation is g(t,y)=h(t,y)e^{-i\theta(t,y)}, so g converges to h_\infty(y)e^{-i(l(y)+\log(t)\phi_\infty(y))}. The displayed identity in the text instead writes g(t,y)-h_\infty(y)e^{+i\lambda(\log(t)\phi_\infty(y)+l(y))} and concludes convergence to the plus-phase profile. Consequently, as written, the proof yields a different phase than the e^{-i\log(t)\phi_\infty} in (1.10), and the l(y) phase is also missing from (1.10). This is fixable by redefining the final h_\infty to absorb e^{-il(y)} and by correcting the signs throughout Section 4.2, but the current text does not support the stated law.
  3. [Section 3.2, Lemmas 3.5 and 3.7] The energy estimates for Gu and G^2u omit the potential term proportional to \phi. From (i\partial_t-\Delta)Gu=\phi Gu+2t(\nabla\phi)u, the standard identity gives \partial_t\|Gu\|_{L^2} \le \|\phi\|_{L^\infty}\|Gu\|_{L^2}+2t\|\nabla\phi\|_{L^\infty}\|u\|_{L^2}. The lemma states only the second term. The missing term is handled by Gr\"onwall using Corollary 3.4 (\|\phi\|_{L^\infty} \lesssim \varepsilon^2\langle t\rangle^{-1}), which inserts an additional factor of t^{O(\varepsilon^2)} or \log^{O(\varepsilon^2)}; for \varepsilon small this is compatible with the bootstrap, but the estimate as printed is not justified. The same omission occurs in the G^2 equation in Lemma 3.7. Please provide the corrected inequality and verify that the log powers absorb the extra factor.
minor comments (5)
  1. [Eq. (4.16)] The definition of F_\infty(y) has a typo: it integrates with dy and uses h_\infty(y) inside a convolution in x. The intended formula is F_\infty(y)=\lambda\int (x-y)/|x-y|^3 |h_\infty(x)|^2 dx, as used in Corollary 4.4.
  2. [Section 1.2 and references] The paper should state explicitly which estimates from the unpublished companion [25] are actually used as inputs. The proof appears self-contained, but Section 1.2 describes [25] as the driving force; the reader needs a precise dependency list to audit the argument.
  3. [Title, Lemma 2.1, and Lemma 3.7] There are several typographical errors: 'SCA TTERING' in the title, 'gives' in Lemma 2.1(2), and 'femionic' in Lemma 3.7.
  4. [Section 3.3, Lemma 3.9] The chain-rule identity '\nabla_x\gamma(t,x-tv,v)=t^{-1}\nabla_v\gamma(t,x-tv,v)-t^{-1}\nabla_v[\gamma(t,x-tv,v)]' is confusing; please rewrite it with the intended variable change and clarify how integration by parts in v is used.
  5. [Section 4.2 and Proposition 4.3] In the final Hartree step, the rate t^{-1/80} is stated without specifying the choice of m \in (3/2,2) and \kappa in Proposition 4.3; please spell out the choice so the reader can verify the exponent.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the asymptotic profiles are constructed as limits rather than assumed, and the only self-referential element is a non-load-bearing citation to the authors' own forthcoming work [25]; the regularity gap under (1.7) is a correctness issue, not circularity.

full rationale

Inspection of the derivation chain shows that the asymptotic objects in Theorem 1.1 are not inputs but constructed limits. In Section 4.1, h(t)=g(t)e^{i\theta(t)} is introduced and Proposition 4.1 proves convergence to h_infty using only the bootstrap bounds on G^2u from Proposition 3.10; Corollary 4.4 then defines F_infty by (4.16) and proves t^2F(t,tv) -> F_infty. No parameter is fitted and no asymptotic profile is inserted by hand; the logarithmic corrections are derived from the time-integrated potential. Similarly, Lemma 4.5 constructs phi_infty as the limit of the effective potential t*phi^0(t,a), and Lemma 4.7 verifies phi_infty = V*rho_infty after convergence, so the coupling identities in (1.8) are proven consequences, not assumptions. The Section 3 bootstrap closes by strict improvement of the norms; assumptions (3.9) are auxiliary and are later consequences, so no circularity is present. The only self-referential element is the citation of the authors' own forthcoming work [25] in Section 1.2, described as 'the driving force of our paper.' However, no displayed estimate from [25] is invoked in Sections 3-4; the needed physical-space Hartree bounds are derived directly, so this citation is not load-bearing. Separately, the theorem as stated has a regularity gap: assumption (1.7) does not imply f0 in C^1, which Proposition 3.1 requires, and no approximation argument is supplied. That gap affects correctness of the stated regularity, not circularity of the derivation.

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

No numbers are fitted to data and no new physical objects are introduced. The proof rests on standard analytic estimates plus explicit smallness and regularity assumptions. The only non-public input is the unpublished companion paper [25], cited as a source of the physical-space method but largely reproduced in the present proof.

assumptions (6)
  • standard math Strichartz estimates, including endpoint Strichartz, for the three-dimensional Schrodinger propagator (Keel-Tao).
    Used in Section 2.3 and in the local well-posedness iteration to control u in mixed space-time norms.
  • standard math Lorentz-space Holder, convolution, Riesz potential, and Riesz transform estimates (O'Neil, Stein, Grafakos), such as (2.3), (2.6)-(2.8).
    These estimates are the main analytic tools for bounding Coulomb potentials and fields in Sections 2 and 3.
  • standard math Liouville theorem: divergence-free Hamiltonian vector fields preserve phase-space measure, so transport along characteristics preserves L^p norms (Lemma 2.3).
    Used throughout the Vlasov component analysis, including bootstrap bounds for weighted moments and derivatives of gamma.
  • standard math The Coulomb kernel shell representation with a compactly supported chi and the identity C0/|z| = integral_0^infty chi(z/R) dR/R^2.
    Used in Section 3.2 to define the effective fields phi_R, E_R, phi_r^0, and E_r^0 and to derive sharp decay.
  • domain assumption Initial data smallness and localization (1.7) with epsilon_0 below a universal threshold.
    The theorem is conditional on this small-data regime; no explicit admissible value of epsilon_0 is given.
  • domain assumption C^1 regularity of f0 for the local well-posedness theory in Proposition 3.1.
    Proposition 3.1 explicitly assumes f0 in C^1, while Theorem 1.1 states C^1 solutions under assumption (1.7) only. This mismatch is a gap in the statement.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Long-Time Dynamics and Modified Scattering for the Coulomb Vlasov--Hartree System." pith.science (2026). https://pith.science/paper/4M64EEMJ

@misc{pith2026260804402,
  author       = {Pith},
  title        = {Pith review of: Long-Time Dynamics and Modified Scattering for the Coulomb Vlasov--Hartree System},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/4M64EEMJ}},
  note         = {Machine review of arXiv:2608.04402}
}
read the original abstract

We study the three-dimensional Vlasov--Hartree system with Coulomb interaction, a classical--quantum model for Bose--Fermi mixtures. For sufficiently small, regular, and localized initial data, we prove global well-posedness and determine the precise long-time dynamics. The coupling remains visible at leading order: the Vlasov distribution scatters along logarithmically corrected free characteristics determined by the asymptotic bosonic profile, while the Hartree wave acquires a logarithmic phase correction generated by the asymptotic fermionic density. Our proof combines techniques from kinetic theory with vector-field methods for dispersive equations and also extends to shorter-range inverse-power interactions.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 28 canonical work pages

  1. [25]

    Pausader and M

    B. Pausader and M. Xie,A Fourier-Free New Method for the 3D Hartree Equation: Full Space and half-space with Robin Condition,in preparation

    work page

  2. [1]

    Bardos and P

    C. Bardos and P. Degond,Global existence for the Vlasov–Poisson equation in 3 space variables with small initial data, Ann. Inst. H. Poincar´ e Anal. Non Lin´ eaire2(1985), no. 2, 101–118

    work page 1985

  3. [2]

    Bigorgne and R

    L. Bigorgne and R. Velozo Ruiz,Late-time asymptotics of small data solutions for the Vlasov–Poisson system,Nonlinearity39(2026), no. 4, Paper No. 045007

    work page 2026

  4. [3]

    C´ ardenas, J

    E. C´ ardenas, J. K. Miller and N. Pavlovi´ c,On the effective dynamics of Bose–Fermi mixtures,Ars Inven. Anal. 2025, Paper No. 5, 54 pp

    work page 2025

  5. [4]

    Global Existence and Time Decay for the Vlasov-Hartree System

    K. Cavanagh,Global existence and time decay for the Vlasov–Hartree system,arXiv:2607.04444

    work page Pith review arXiv

  6. [5]

    Choi and S

    S.-H. Choi and S. Kwon,Modified scattering for the Vlasov–Poisson system, Nonlinearity29(2016), 2755– 2774. MODIFIED SCATTERING FOR THE COULOMB VLASOV–HARTREE SYSTEM 29

    work page 2016

  7. [6]

    Flynn, Z

    P. Flynn, Z. Ouyang, B. Pausader and K. Widmayer,Scattering map for the Vlasov–Poisson system,Peking Math. J.6(2023), no. 2, 365–392

    work page 2023

  8. [7]

    Ginibre and T

    J. Ginibre and T. Ozawa,Long range scattering for non-linear Schr¨ odinger and Hartree equations in space dimensionn≥2,Comm. Math. Phys.151(1993), no. 3, 619–645

    work page 1993

Show all 30 references
  1. [8]

    R. T. Glassey,The Cauchy problem in kinetic theory, SIAM, Philadelphia, PA, 1996; MR1379589

    1996

  2. [9]

    Grafakos,Classical Fourier analysis, third edition, Graduate Texts in Mathematics, 249, Springer, New York, 2014; MR3243734

    L. Grafakos,Classical Fourier analysis, third edition, Graduate Texts in Mathematics, 249, Springer, New York, 2014; MR3243734

    2014

  3. [10]

    Hayashi and P

    N. Hayashi and P. I. Naumkin,Asymptotics for large time of solutions to the nonlinear Schr¨ odinger and Hartree equations,Amer. J. Math.120(1998), no. 2, 369–389

    1998

  4. [11]

    Hayashi, P

    N. Hayashi, P. I. Naumkin and T. Ozawa,Scattering theory for the Hartree equation,SIAM J. Math. Anal. 29(1998), no. 5, 1256–1267

    1998

  5. [12]

    Hong and S

    Y. Hong and S. Pankavich,Modified scattering for the Vlasov–Riesz system with long-range interactions, arXiv:2604.04256

    arXiv

  6. [13]

    Huang and H

    W. Huang and H. Kwon,Scattering of the Vlasov–Riesz system in the three dimensions,Ann. Henri Poincar´ e (2026)

    2026

  7. [14]

    Huang and H

    W. Huang and H. Kwon,Scattering map for the Vlasov–Poisson system with a repulsive harmonic potential, arXiv:2602.21344

  8. [15]

    Huang, B

    W. Huang, B. Pausader and M. Suzuki,The Vlasov–Poisson system with a perfectly conducting wall: Convex domains,Comm. Math. Phys., to appear arXiv:2412.13434

    arXiv

  9. [16]

    Iacobelli, S

    M. Iacobelli, S. Rossi, and K. Widmayer,On the stability of vacuum in the screened Vlasov–Poisson equation, Journal of the London Mathematical Society113(2026), no. 1, e70426

    2026

  10. [17]

    A. D. Ionescu, B. Pausader, X. Wang and K. Widmayer,On the asymptotic behavior of solutions to the Vlasov–Poisson system,Int. Math. Res. Not. IMRN2022(2022), no. 12, 8865–8889

    2022

  11. [18]

    Kato and F

    J. Kato and F. Pusateri,A new proof of long-range scattering for critical nonlinear Schr¨ odinger equations, Differential Integral Equations24(2011), no. 9–10, 923–940

    2011

  12. [19]

    Keel and T

    M. Keel and T. C. Tao, Endpoint Strichartz estimates, Amer. J. Math.120(1998), no. 5, 955–980

    1998

  13. [20]

    Kepka and K

    B. Kepka and K. Widmayer,Modified scattering dynamics in the Vlasov–Poisson equation near an attractive point mass,arXiv:2511.04363

  14. [21]

    Lions and B

    P.-L. Lions and B. Perthame,Propagation of moments and regularity for the 3-dimensional Vlasov–Poisson system, Invent. Math.105(1991), 415–430

    1991

  15. [22]

    O’Neil, Convolution operators andL(p, q) spaces, Duke Math

    R. O’Neil, Convolution operators andL(p, q) spaces, Duke Math. J.30(1963), 129–142; MR0146673

    1963

  16. [23]

    Pausader and K

    B. Pausader and K. Widmayer,Stability of a point charge for the Vlasov–Poisson system: The radial case, Comm. Math. Phys.385(2021), no. 3, 1741–1769

    2021

  17. [24]

    Pausader, K

    B. Pausader, K. Widmayer and J. Yang,Stability of a point charge for the repulsive Vlasov–Poisson system, J. Eur. Math. Soc.28(2026), no. 7, 2751–2848

    2026

  18. [26]

    Peetre, Espaces d’interpolation et th´ eor` eme de Soboleff, Ann

    J. Peetre, Espaces d’interpolation et th´ eor` eme de Soboleff, Ann. Inst. Fourier (Grenoble)16(1966), fasc. 1, 279–317; MR0221282

    1966

  19. [27]

    Pfaffelmoser,Global classical solutions of the Vlasov–Poisson system in three dimensions for general initial data, J

    K. Pfaffelmoser,Global classical solutions of the Vlasov–Poisson system in three dimensions for general initial data, J. Differential Equations95(1992), 281–303

    1992

  20. [28]

    Schlue and M

    V. Schlue and M. Taylor,Inverse modified scattering and polyhomogeneous expansions for the Vlasov– Poisson system, Nonlinearity38(2025), no. 9, 095019

    2025

  21. [29]

    E. M. Stein,Singular integrals and differentiability properties of functions, Princeton Mathematical Series, No. 30, Princeton Univ. Press, Princeton, NJ, 1970

    1970

  22. [30]

    Van Hoose,Modified scattering for the Hartree nonlinear Schr¨ odinger equation, arXiv preprint arXiv:2407.18411 (2024)

    T. Van Hoose,Modified scattering for the Hartree nonlinear Schr¨ odinger equation, arXiv preprint arXiv:2407.18411 (2024). Brown University Email address:wenrui huang@brown.edu Yale University Email address:mengyi.xie@yale.edu

    2024 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.