Pith. sign in

REVIEW 3 major objections 5 minor 53 references

A simple local nonlinear rescaling of the matter density field recovers linear power-spectrum information and makes the power spectrum directly sensitive to local primordial non-Gaussianity.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-03 16:41 UTC pith:N5ISPL4Z

load-bearing objection A plausible Fisher gain for f_NL from transformed density power spectra, but the abstract/body mismatch and an unquantified step in the analytic derivation need fixing before I'd trust the factor-237. the 3 major comments →

arxiv 2512.12304 v2 pith:N5ISPL4Z submitted 2025-12-13 astro-ph.CO

Local Nonlinear Transforms Effectively Extract Cosmological Information From Large-Scale Structure

Yun Wang , Hao-Ran Yu , Yu Yu , Ping He This is my paper
classification astro-ph.CO
keywords primordial non-Gaussianitylarge-scale structureZel'dovich approximationlocal nonlinear transformpower spectrumFisher forecastsqueezed bispectrumneutrino mass
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper claims that a simple local rescaling of the late-time matter density field—the Zel'dovich-κ (Z-κ) transform, derived as the inverse of the Zel'dovich collapse mapping—can undo much of the non-Gaussianity produced by gravitational evolution. When applied to N-body simulations, the transform Gaussianizes the density distribution and yields power spectra that closely track the linear spectrum. The paper's central claim is that these transformed power spectra are not just a cleaner version of the usual spectrum: because the transform couples long and short wavelength modes, its power spectrum contains an integral over squeezed triangles, making it directly sensitive to local primordial non-Gaussianity. Fisher forecasts from the Quijote simulations show that the combination of the nonlinear and Z-6 power spectra tightens the marginalized constraint on f_NL^local to σ≈13.9, a 237-fold improvement, and on summed neutrino mass to σ≈0.15 eV, a 7.7-fold improvement, for a 1 (h^-1 Gpc)^3 volume. If correct, this offers a computationally cheap two-point route to early-universe physics that Stage-IV surveys could exploit.

Core claim

The authors introduce the Z-κ transform, δ_{Z-κ} ≡ κ[1 − (1+δ_nl)^{-1/κ}], and show it is the analytic inverse of the Zel'dovich mapping for planar, filamentary, and spherical collapse (κ=1,2,3), with κ→∞ giving the log transform. Using Quijote simulations, they find κ≈6 optimally Gaussianizes the density field, and the power spectra P_log and P_{Z-6} track the linear spectrum. Expanding the transform to second order, they derive P_{Z-κ}(k) ≈ P_nl(k) − (1+κ^{-1})∫ B_nl(k,p,|k−p|)d^3p/(2π)^3, where the integral is dominated by squeezed triangles. Approximating the nonlinear bispectrum there by the primordial local-PNG bispectrum, the derivative ∂P_{Z-κ}/∂f_NL^local becomes proportional to −P_

What carries the argument

The Z-κ transform (Eq. 5) is the central object: a local, parameterized nonlinear map from the observed density to a 'de-evolved' field that emulates the inverse of Zel'dovich gravitational collapse. The load-bearing identity is the second-order expansion of the transformed field, δ_{Z-κ} ≈ δ_nl − (1+κ^{-1})δ_nl^2/2, which converts the power spectrum of the transformed field into the nonlinear power spectrum minus an integral over the bispectrum; because the integral's phase space is dominated by squeezed triangles, the transformed power spectrum inherits sensitivity to the squeezed bispectrum, exactly where local PNG peaks. The analytic proportionality ∂P_{Z-κ}/∂f_NL^local ∝ −P_lin(k)/M(k)

Load-bearing premise

The analytic link between the transformed power spectrum and primordial non-Gaussianity assumes that, in the squeezed-triangle integral governing P_{Z-κ}, the full nonlinear bispectrum can be replaced by the primordial local-PNG bispectrum; if gravitational mode coupling produces a comparable squeezed bispectrum at z=0, the measured derivative ∂P_{Z-κ}/∂f_NL^local mixes primordial and gravitational signals.

What would settle it

Take the Quijote simulations with Gaussian initial conditions (f_NL=0) and measure the squeezed-triangle integral ∫ B_nl(k,p,p)d^3p at z=0 for the k-range used (k≲0.25 h/Mpc). If that gravity-only integral is comparable to or larger than the primordial signal at the forecasted f_NL sensitivity (e.g., σ≈14), the claim that the transformed power spectrum 'directly captures' local PNG fails. Alternatively, check whether ∂P_{Z-6}/∂f_NL^local measured on f_NL=0 simulations is non-negligible.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • A survey volume of ~8 (h^-1 Gpc)^3 — within Stage-IV reach — would suffice to reach σ(f_NL^local) ≲ 5, and combining with CMB data could push uncertainty to ~1, enabling tests of inflationary models that predict enhanced local PNG.
  • The method makes squeezed-bispectrum information accessible via two-point statistics, avoiding expensive higher-order estimators while adding negligible computational cost to survey analyses.
  • The Z-κ framework supplies a physical origin for the widely used Box-Cox and log transforms, clarifying why they work: they emulate the inverse of nonlinear gravitational evolution.
  • The combination P_nl + P_{Z-6} is forecast to outperform P_nl + P_log on f_NL^local, f_NL^ortho, and Mν by roughly 2–25%, with slightly worse performance on equilateral PNG and some ΛCDM parameters.
  • The improvements arise from generic features of nonlinear evolution and squeezed mode coupling, so the authors argue they should persist once biased tracers, redshift space, and survey systematics are modeled.

Where Pith is reading between the lines

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

  • If the transform's squeezed-sensitivity is as clean as argued, the same trick might be used to probe other squeezed-limit signals, such as those from massive neutrinos or modified gravity, by choosing transforms that selectively weight underdense or overdense regions.
  • A direct test of the analytic link would be to compute the gravity-only squeezed bispectrum (from Gaussian-initial-condition simulations) and check that its integrated contribution in Eq. (7) is negligible against the primordial term at z=0; this is not quantified in the paper.
  • The abstract's quoted improvements (290x for f_NL^local, 107x for Mν) use a different statistic, P_ZI, that is not defined in the body; reconciling these numbers would clarify whether the stronger claim is reproducible from the body's procedure.
  • The method's real-space, matter-only scope means forecast gains will degrade with redshift-space distortions and galaxy bias; a natural extension is to test whether a redshift-space version retains the squeezed-bispectrum coupling.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper introduces a local nonlinear transform of the matter density field, the Z-κ transform δ_{Z-κ} ≡ κ[1 − (1+δ_nl)^{-1/κ}], motivated by the inverse of the Zel'dovich approximation, and shows that with κ ≈ 6 (or κ → ∞, the log transform) the density PDF is Gaussianized and the transformed power spectrum tracks the linear spectrum. Using the Quijote and Quijote-PNG simulations and a Fisher forecast over ΛCDM, neutrino mass, and three PNG shapes, the authors report that combining the nonlinear power spectrum with the Z-6 or log power spectrum substantially tightens constraints, especially for local PNG: σ(f_NL^local)=13.88 (a 237× improvement over P_nl alone) and σ(M_ν)=0.1516 eV (a 7.74× improvement) for P_nl+P_{Z-6} at k_max=0.5 h/Mpc. The central interpretive claim is that the transformed power spectrum directly captures the squeezed primordial bispectrum, with ∂P_{Z-κ}/∂f_NL^local ∝ −P_lin(k)/M(k).

Significance. If correct, the result is significant: a cheap, local, two-point statistic would recover a large fraction of the squeezed-bispectrum information that is normally associated with higher-order statistics, with immediate implications for Stage-IV surveys targeting primordial non-Gaussianity and neutrino mass. The paper uses a standard Fisher pipeline with public simulations, and the numerical body is largely internally consistent. The claimed 237× gain in f_NL and 7.7× gain in M_ν are striking and warrant scrutiny. However, the analytic mechanism that underpins the headline result is asserted rather than demonstrated, and the abstract's headline numbers are not the same as those in the body, so the paper in its current form does not yet establish the central claim at the level required.

major comments (3)
  1. [Abstract vs. Table I; Eq. (5)] The abstract reports improvement factors of 290× for f_NL^local and 107× for M_ν using a joint data vector of three transformed-field power spectra with {η=∞,6,3} denoted P_ZI. The body defines no P_ZI statistic and no parameter η; the transform in Eq. (5) depends on κ, and Table I reports 237.0× and 7.74× for f_NL and M_ν respectively from P_nl+P_{Z-6}. These are very different numbers and the 107× factor does not appear anywhere in the main text or tables. The reader cannot determine what statistic the abstract advertises or how the factors were obtained. This ambiguity must be resolved before the paper can be evaluated as a coherent contribution.
  2. [Supplemental Eq. (9)] The load-bearing analytic step is the replacement of the full nonlinear bispectrum B_nl(k,p,|k−p|) in Eq. (7) by the primordial local-PNG bispectrum B_PNG(k,p,p) in Eq. (9). At z=0 the nonlinear bispectrum is dominated by gravitational mode coupling, not by the primordial component. While the f_NL-independent gravitational contribution cancels in the derivative ∂P_{Z-κ}/∂f_NL, the f_NL-dependent part of the late-time bispectrum—arising, e.g., from nonlinear evolution of density perturbations seeded by the f_NL component of the initial potential—does not cancel and is not bounded anywhere in the paper. The claimed proportionality ∂P_{Z-κ}/∂f_NL ∝ −P_lin(k)/M(k), and hence the predicted 237× gain, requires that this gravitational/PNG cross term be subdominant in the squeezed integral over the k range where the gain saturates (k ≲ 0.25 h/Mpc; Fig. 3 right). This is asserted, not demonstrate
  3. [Fig. 2, right panel; Eq. (10)] The validation of Eq. (10) uses a CAMB P_lin(k)/M(k) curve rescaled by a constant factor to match ∂P_log/∂f_NL. A shape agreement after an arbitrary amplitude rescaling does not validate the mechanism; the amplitude of the derivative is exactly what matters for the Fisher gain. The paper should compare the measured ∂P_{Z-6}/∂f_NL with the prediction of Eq. (10) including the coefficient (1+κ^{-1}) and the squeezed integral, using measured or simulated P_lin and M(k), without an adjustable normalization. Without this, the analytic link between the transform and local PNG remains only qualitative.
minor comments (5)
  1. [Table I caption] Typo: 'combiantions' should be 'combinations'.
  2. [Supplemental page 2] Typo: 'On the large sales' should be 'On the large scales'.
  3. [Section 'Results', Fig. 3 discussion] The text says constraints from P_log and P_{Z-6} 'nearly coincide', but Table I shows a 14% difference for f_NL^local (55.92 vs 47.85) and a 9% difference for M_ν (0.7533 vs 0.6864). Please clarify whether this is considered coincident and whether the differences affect the comparison in the following sentence.
  4. [Fisher formalism] The paper does not state the number of k-bins, the covariance estimation procedure (e.g., whether the sample covariance from 15,000 realizations is inverted with a Hartlap factor), or the finite-difference step sizes used for derivatives. These details are needed for reproducibility of the Fisher numbers.
  5. [References and notation] The transform is referred to as Z-κ, Box-Cox, and log; the abstract introduces η as the transform parameter while the body uses κ. Use a single notation throughout to avoid confusion.

Circularity Check

0 steps flagged

No significant circularity: the Fisher gains are forecasts from Quijote simulation derivatives and covariance, not fits to f_NL or M_nu; the main concern is an unquantified approximation, not a circular reduction.

full rationale

The paper's core derivation is self-contained. The Z-kappa transform is defined by Eq. (5) and motivated by the Zel'dovich mapping of Eqs. (1)-(4); the Box-Cox/log relation (Eq. 6) is explicitly acknowledged. kappa=6 is chosen to Gaussianize the density PDF, but the f_NL and M_nu constraints are not fitted to kappa, and the log limit with no tuned parameter gives nearly the same Fisher gains (Table I), so the headline improvements do not reduce to the tuned parameter. The Fisher forecasts use numerical derivatives from the Quijote-PNG suite and covariance from fiducial realizations; they are forecasts, not retrospective predictions. No load-bearing self-citation appears: Quijote and CAMB are external references, and no author-specific uniqueness theorem is invoked. The only questionable analytic step is Supplemental Eq. (9), where the full nonlinear bispectrum in Eq. (7) is replaced by the primordial local-PNG bispectrum without quantifying the gravitational squeezed contribution. This is an unquantified physical approximation and validation gap, not a circular reduction, because the same scale-dependent response is checked against numerical derivatives and the log/Z-6 curves match. The abstract's undefined P_ZI and mismatched improvement factors are internal-consistency issues, not circularity.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The Z-κ transform is effectively a reparameterized Box-Cox/log transform, so the main burden lands on the free parameters (κ, k_max, and the undefined abstract η) and on three domain assumptions: the extrapolation of the Zel'dovich inverse to the fully evolved field, the Gaussian/cosmology-independent covariance in the Fisher matrix, and the replacement of the nonlinear bispectrum by the primordial PNG bispectrum in the squeezed limit. There are no new physical entities beyond the transform itself.

free parameters (3)
  • κ (Z-κ transform order) = κ ≈ 6; κ → ∞ (log) also reported
    κ is a free parameter in the transform (Eq. 5). It is set by minimizing the mean absolute deviation between the transformed PDF and a Gaussian in the same Quijote simulations later used for the Fisher forecasts (Fig. 1, top panel). The Zel'dovich derivation fixes κ=1,2,3 only for special collapse geometries before shell crossing; κ=6 is an empirical fit.
  • k_max (analysis cutoff) = 0.5 h Mpc^-1
    Maximum wavenumber for all power spectra, chosen 'to avoid numerical artifacts'; constraints and improvement factors are quoted for this cutoff. A hand-chosen analysis boundary, not fitted to the data.
  • η (abstract transform parameter) = ∞, 6, 3 (abstract only)
    The abstract defines a ZI transform with η controls weighting and a joint data vector P_ZI from three transformed spectra {η=∞,6,3}; the full text never defines η or P_ZI. If η is a parameter, it is undefined in the main text, creating an ambiguity in the central claim.
axioms (4)
  • domain assumption The Zel'dovich approximation relation 1+δ = ∏(1-Dα_i)^{-1}, and the special collapse mapping 1+δ_nl = (1-δ_lin/κ)^{-κ} for κ=1,2,3, remains a valid local inverse for the fully evolved field when κ is extended to a free parameter.
    Invoked to motivate Eq. (5); exact only before shell-crossing and for symmetric collapses, then extended to the fully evolved nonlinear density field without proof.
  • domain assumption The likelihood for the power spectrum data vector is Gaussian, and the covariance matrix is independent of cosmology.
    The Fisher formalism in Eq. (7) relies on this; covariance is estimated from 15,000 fiducial Quijote simulations and assumed fixed, with refs [23-25] cited in support but not verified for the transformed fields.
  • ad hoc to paper In the squeezed limit, the full nonlinear bispectrum B_nl can be replaced by the primordial local-PNG bispectrum B_PNG.
    Supplemental Eq. (9) substitutes B_PNG for B_nl without quantifying the gravitational bispectrum contribution; this is the load-bearing step in the analytic claim that transformed power spectra directly probe local PNG.
  • standard math The primordial potential bispectrum follows from Wick's theorem, and the Poisson equation δ(k)=M(k)Φ(k) with M(k)=2 k^2 T(k)/(3Ωm H0^2) connects potential to density.
    Used in Supplemental Eqs. (3)-(5) to express the local-PNG bispectrum; standard background, not independently verified in this paper.

pith-pipeline@v1.3.0-alltime-deepseek · 10312 in / 20978 out tokens · 210882 ms · 2026-08-03T16:41:40.338464+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of Local Nonlinear Transforms Effectively Extract Cosmological Information From Large-Scale Structure." pith.science (2026). https://pith.science/paper/N5ISPL4Z

@misc{pith2026251212304,
  author       = {Pith},
  title        = {Pith review of: Local Nonlinear Transforms Effectively Extract Cosmological Information From Large-Scale Structure},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/N5ISPL4Z}},
  note         = {Machine review of arXiv:2512.12304}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Extracting cosmological information from nonlinear and non-Gaussian large-scale structure remains a major challenge. We introduce the (Zel'dovich-inspired) ZI transform, a simple one-parameter local nonlinear transform in which $\eta$ controls the weighting of higher-order information in the transformed density field. For $\eta\geq3$, the transform can substantially suppress gravitational non-Gaussianity. Using the \textsc{Quijote} suite and Fisher information analysis over summed neutrino mass $M_\nu$, primordial non-Gaussianity $f_\mathrm{NL}$ with different shapes, and $\Lambda$CDM parameters, we find that the joint data vector of three transformed-field power spectra with $\{\eta=\infty, 6, 3\}$, denoted $P_\mathrm{ZI}$, tightens all constraints relative to the ordinary power spectrum, especially boosting the constraining power by factors of $290$ for $f_\mathrm{NL}^\mathrm{local}$ and $107$ for $M_\nu$, while yielding nearly unbiased parameter estimates. Compared further with other statistics beyond the ordinary power spectrum, $P_\mathrm{ZI}$ stands out as a powerful cosmological probe.

Figures

Figures reproduced from arXiv: 2512.12304 by Hao-Ran Yu, Ping He, Yun Wang, Yu Yu.

Figure 1
Figure 1. Figure 1: Gaussianization under different transformations. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Correlation matrix and parameter responses of power spectra. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Fisher forecast constraints from power spectra. [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 1
Figure 1. Figure 1: FIG. 1: Numerical partial derivatives of [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2: Illustration of Zel’dovich- [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3: Fisher forecast constraints on the parameters [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

53 extracted references · 1 canonical work pages · 1 internal anchor

  1. [1]

    Gaussianization under different transformations

    6 Pi(k)/P in, smooth(k) nl log Z-6 Pin/P in, smooth 0 1 2 3 4nl Figure 1. Gaussianization under different transformations. T op: mean absolute deviation between the transformed density PDF and the Gaussian distribution. The blue dashed line marks the deviation in the log limit, equal to 0.0248. Middle: PDFs of the normalized density fields yi = ( δi − µi)/...

  2. [2]

    8 PDF(yi) log Z-6 Gaussian 10− 2 10− 1 k [hMpc− 1]

  3. [3]

    into Eq. ( 2) yields the analytic mapping from the linear density δlin to the nonlinear density δnl as below 1 + δnl = [1 − δlin/κ]−κ with κ = 1, 2, 3, (4) arXiv:2512.12304v1 [astro-ph.CO] 13 Dec 2025 2 0 5 10 15 20 25 30 κ 10− 2 10− 1 ⟨|PDF(δZ− κ)− G(δZ− κ)|⟩ log − 3 − 2 − 1 0 1 2 3 yi

    Pith/arXiv arXiv 2025

  4. [4]

    A. Repp, I. Szapudi, J. Carron, and M. Wolk, MNRAS 454, 3533 (2015)

    2015

  5. [5]

    i” denotes “nl

    will take the form of Box-Cox transform [8, 9], δZ−κ λ=−1/κ − − − − − → { [(1 + δnl)λ − 1]/λ if λ ̸= 0, ln (1 + δnl) if λ = 0. (6) We see that the Box-Cox transform and its log limit Gaus- sianize the density field because the former is algebraically equivalent to a reparameterization of the Z-κ transform (with the log case corresponding to its κ → ∞ limit...

  6. [6]

    00 0. 25 0. 00 0. 25 0. 00 0. 25 0. 50 k [hMpc− 1]

  7. [7]

    50 k [hMpc− 1] Pnl Plog PZ− 6 − 0. 2

  8. [8]

    f local NL

    0 Cij/ √ CiiCjj 10− 2 10− 110− 2 10− 1 100 101 102 ∂P i/∂θ j f local NL Plin(k) M(k) 10− 2 10− 1 10− 2 10− 1 100 nl log Z-6 f equil NL 10− 2 10− 1 k [hMpc− 1] 10− 3 10− 2 10− 1 100 ∂P i/∂θ j f ortho NL 10− 2 10− 1 k [hMpc− 1] 101 102 103 Mν Figure 2. Correlation matrix and parameter responses of power spectra. Left: auto and cross correlation coefficients ...

  9. [9]

    1 0. 2 0. 3 0. 4 0. 5101 102 103 104 σ(θi) f local NL

  10. [10]

    1 0. 2 0. 3 0. 4 0. 5 103 104 f equil NL

  11. [11]

    1 0. 2 0. 3 0. 4 0. 5 kmax [hMpc− 1] 103 104 σ(θi) f ortho NL

  12. [12]

    1 0. 2 0. 3 0. 4 0. 5 kmax [hMpc− 1] 100 Mν kmax = 0. 5 hMpc− 1 nl log Z-6 nl+log nl+Z-6 Figure 3. Fisher forecast constraints from power spectra. Left: joint and marginalized constraints on the parameters {f local NL , f equil NL , f ortho NL , Mν } at z = 0 , using Pnl, Plog, PZ−6, and their combinations up to kmax = 0 .5 hMpc−1. Right: dependence of th...

  13. [13]

    DESI Collaboration, arXiv e-prints , arXiv:1611.00036 (2016)

    Pith/arXiv arXiv 2016

  14. [14]

    Amendola, S

    L. Amendola, S. Appleby, A. Avgoustidis, , and et al., Living Rev. Relativ. 21, 2 (2018)

    2018

  15. [15]

    Bernardeau, S

    F. Bernardeau, S. Colombi, E. Gazta ˜naga, and R. Scoccimarro, Phys. Rep. 367, 1 (2002)

    2002

  16. [16]

    M. C. Neyrinck, I. Szapudi, and A. S. Szalay, ApJL 698, L90 (2009)

    2009

  17. [17]

    M. C. Neyrinck, ApJ 742, 91 (2011)

    2011

  18. [18]

    M. C. Neyrinck, I. Szapudi, and A. S. Szalay, ApJ 731, 116 (2011)

    2011

  19. [19]

    Joachimi and A

    B. Joachimi and A. N. Taylor, MNRAS 416, 1010 (2011)

    2011

  20. [20]

    Joachimi, A

    B. Joachimi, A. N. Taylor, and A. Kiessling, MNRAS 418, 145 (2011)

    2011

  21. [21]

    Simpson, J

    F. Simpson, J. B. James, A. F. Heavens, and C. Heymans, PRL 107, 271301 (2011)

    2011

  22. [22]

    Simpson, A

    F. Simpson, A. F. Heavens, and C. Heymans, PRD 88, 083510 (2013)

    2013

  23. [23]

    Y . B. Zel’dovich, A&A 5, 84 (1970)

    1970

  24. [24]

    S. F. Shandarin and Y . B. Zeldovich, RMP 61, 185 (1989)

    1989

  25. [25]

    Y oshisato, M

    A. Y oshisato, M. Morikawa, N. Gouda, and H. Mouri, ApJ 637, 555 (2006)

    2006

  26. [26]

    Rampf and G

    C. Rampf and G. Rigopoulos, MNRAS:Letters 430, L54 (2013)

    2013

  27. [27]

    White, MNRAS 439, 3630 (2014)

    M. White, MNRAS 439, 3630 (2014)

    2014

  28. [28]

    Hidding, S

    J. Hidding, S. F. Shandarin, and R. van de Weygaert, MNRAS 437, 3442 (2014)

    2014

  29. [29]

    Tassev, JCAP 2014 (06), 012

    S. Tassev, JCAP 2014 (06), 012

    2014

  30. [30]

    White, MNRAS 450, 3822 (2015)

    M. White, MNRAS 450, 3822 (2015)

    2015

  31. [31]

    Kokron, S.-F

    N. Kokron, S.-F. Chen, M. White, and et al., JCAP 2022 (9), 059

    2022

  32. [32]

    R. A. Fisher, RSPTA 222, 309 (1922)

    1922

  33. [33]

    Tegmark, A

    M. Tegmark, A. N. Taylor, and A. F. Heavens, ApJ 480, 22 (1997)

    1997

  34. [34]

    R. E. Upham, M. L. Brown, and L. Whittaker, MNRAS 503, 1999 (2021)

    1999

  35. [35]

    Paillas, C

    E. Paillas, C. Cuesta-Lazaro, P . Zarrouk, and et al., MNRAS 522, 606 (2023)

    2023

  36. [36]

    Cosmological information content of Betti curves and $k$-nearest neighbor distributions

    A. Ouellette and G. Holder, arXiv e-prints 10.48550/arXiv.2502.09709 (2025)

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2502.09709 2025

  37. [37]

    Villaescusa-Navarro, C

    F. Villaescusa-Navarro, C. Hahn, E. Massara, and et al., ApJS 250, 2 (2020)

    2020

  38. [38]

    G. Jung, D. Karagiannis, M. Liguori, and et al., ApJ 940, 71 (2022)

    2022

  39. [39]

    W. R. Coulton, F. Villaescusa-Navarro, D. Jamieson, and et al., ApJ 943, 64 (2023)

    2023

  40. [40]

    Sefusatti, M

    E. Sefusatti, M. Crocce, R. Scoccimarro, and et al., MNRAS 460, 3624 (2016)

    2016

  41. [41]

    https://pylians3.readthedocs.io/en/master/

  42. [42]

    L. Blot, P . S. Corasaniti, J.-M. Alimi, and et al., MNRAS 446, 1756 (2015)

    2015

  43. [44]

    P . D. Meerburg, D. Green, R. Flauger, and et al., BAAS 51, 107 (2019)

    2019

  44. [48]

    Massara, F

    E. Massara, F. Villaescusa-Navarro, M. Viel, and et al., JCAP 2015 (11), 018

    2015

  45. [49]

    D. H. Weinberg, M. J. Mortonson, D. J. Eisenstein, and et al., Phys. Rep. 530, 87 (2013)

    2013

  46. [50]

    Local Nonlinear Transforms effectively Reveal Primordial Information in Large-Scale Structure

    F. Petracca, F. Marulli, L. Moscardini, and et al., MNRAS 462, 4208 (2016). Supplemental Material for “Local Nonlinear Transforms effectively Reveal Primordial Information in Large-Scale Structure” Yun Wang ( 王云),1, ∗ Hao-Ran Yu ( 于浩然),2 Yu Yu ( 余瑜),3, 4 and Ping He ( 何平)5, 6 1Key Laboratory of Material Simulation Methods & Software of Ministry of Educatio...

    2016

  47. [51]

    5 ns 0 1h − 1 0 1 Mν − 3000 0 3000 f ortho NL − 6000 0 6000 f equil NL − 6000 0 6000 f equil NL − 3000 0 3000 f ortho NL − 1 0 1 Mν 0 1 h

  48. [52]

    0 σ8 nl log Z-6 nl+log nl+Z-6 FIG

    5 1. 0 σ8 nl log Z-6 nl+log nl+Z-6 FIG. 3: Fisher forecast constraints on the parameters {f local NL , f equil NL , f ortho NL , Mν, h, ns, Ωm, Ωb, σ8} at z = 0, using Pnl (gray), Plog (blue), PZ−6 (red), Pnl +Plog (green), and Pnl +PZ−6 (orange) up to kmax = 0.5 hMpc−1. Diagonal panels show the one-dimensional marginalized distributions, and the lower pa...

  49. [53]

    Takahashi, Prog

    T. Takahashi, Prog. Theor. Exp. Phys. 2014, 06B105 (2014)

    2014

  50. [54]

    P. D. Meerburg, D. Green, R. Flauger, and et al., BAAS 51, 107 (2019)

    2019

  51. [55]

    Achúcarro, M

    A. Achúcarro, M. Biagetti, M. Braglia, and et al., arXiv e-prints , arXiv:2203.08128 (2022)

    Pith/arXiv arXiv 2022

  52. [56]

    Lewis, A

    A. Lewis, A. Challinor, and A. Lasenby, ApJ 538, 473 (2000)

    2000

  53. [57]

    https://camb.readthedocs.io/en/latest/