Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Towards an application of fourth-order shear statistics II: Efficient estimation of fourth-order shear correlation functions and an application to the DES Y3 data

T0 review · 3 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

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

Pith's one-line read A quadratic-time estimator measures the connected fourth-order aperture mass in DES Y3 data

desk verdict Solid methodological advance in 4PCF estimation; the DES Y3 'detection' p-value is not calibrated and should not be quoted as a headline number as published. read the letter →

arxiv 2509.07974 v1 pith:Q5UM3BKL submitted 2025-09-09 astro-ph.CO

Lucas Porth , Elena Silvestre-Rosello , Peter Schneider , Martina Larma This is my paper
classification astro-ph.CO
keywords weakgravitationallensingcosmicshearfourth-ordercorrelationfunctionaperturemassstatisticsmultipoledecompositionDESY3non-Gaussianitylarge-scalestructure
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

The paper claims that the four-point shear correlation function can be estimated efficiently by expanding it in multipoles, reducing the cost from counting galaxy quadruplets to quadratic scaling. It validates this estimator on Gaussian random fields and on N-body simulations, finding percent-level agreement, and then applies it to the Dark Energy Survey Y3 shape catalogue. The headline result is a detection of the connected part of the fourth-order aperture mass—the non-Gaussian piece left after two-point contributions are subtracted—with a null-hypothesis p-value of about 4e-137 using the internally estimated covariance. The paper also finds that the sampling distribution of this statistic is strongly skewed, so internally estimated error bars are overconfident unless corrected. If correct, the measurement opens fourth-order shear statistics as a practical probe for Stage III and IV surveys.

What carries the argument

The estimator decomposes the angular dependence of the shear 4PCF into multipole components using a generalized projection. Each multipole is built from sums of per-galaxy discrete fields that collect weighted shears or counts in annular bins; the four-point sum factorizes into three neighbor sums plus one outer sum, giving quadratic scaling. Radial-bin permutation symmetries reduce the number of configurations that must be computed, and a low-memory implementation with hierarchical reduced catalogues controls both memory and runtime. The aperture-mass integrals from the companion paper then convert the binned 4PCF into fourth-order aperture statistics with an implicit E/B-mode decomposition

What would settle it

Compute the null-hypothesis p-value for the connected fourth-order aperture mass using the full 864-footprint sample covariance, or an independent set of DES-Y3-geometry mocks, instead of the internal covariance corrected by r_std,foot; if the p-value becomes non-significant, the claimed detection rests on that heuristic. A second check is to run the estimator on mocks with the DES Y3 best-fit cosmology and see whether the low measured amplitude persists.

Watch

Extended reading notes

Core claim

The central claim is that the multipole decomposition of the shear 4PCF, previously applied to three-point functions, extends to fourth order and turns the naive N^4 quadruplet sum into factorized per-galaxy sums that scale quadratically, with tree-based hierarchies reducing the prefactor. Using the natural components of the shear 4PCF, the estimator reconstructs the correlation function to percent accuracy with multipoles up to n_max ~ 15. Applied non-tomographically to DES Y3 and integrated into fourth-order aperture statistics, it yields a strong rejection of the null hypothesis of no connected E-mode signal (p = 4.09e-137 with the internally estimated covariance), while B-mode and parity

Load-bearing premise

The detection's quoted significance relies on a Gaussian likelihood and a covariance built from mocks plus an internal patch estimate corrected by a heuristic factor, even though the paper shows the sampling distribution is strongly skewed and the raw internal errors are overconfident.

Editorial extensions

If this is right

  • Measuring the shear 4PCF becomes practical for billions-of-galaxies surveys: the multipole estimator runs in quadratic time and the low-memory implementation keeps the footprint manageable.
  • The recommended binning (65 radial bins, n_max=15) gives about 2 percent accuracy for the fourth-order aperture mass on scales 4-30 arcmin, so the statistic can serve as a compression of the full 4PCF.
  • DES Y3 shows a connected fourth-order aperture-mass signal: the E-mode null hypothesis is rejected with p about 4e-137 under the internally estimated covariance, while B-modes and parity-violating modes are consistent with noise.
  • The DES Y3 amplitude is lower than the mock prediction, and the paper argues this mirrors the earlier DES Y3 third-order result; understanding this offset is a prerequisite for cosmological interpretation.
  • The sampling distribution of the fourth-order aperture mass is skewed, so error bars from internal estimators are overconfident unless a correction such as r_std,foot is applied.

Reading between the lines

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

  • The same factorization extends to arbitrary order, so fifth- and sixth-order shear cumulants may become computationally accessible; the bottleneck will shift to the number of multipoles and bins rather than tuple counting.
  • The strong non-Gaussianity of the sampling distribution suggests that Gaussian likelihood analyses of fourth-order cumulants will be biased; simulation-based inference or variance-stabilizing transformations should be tested on the mocks before cosmological fits.
  • If the low DES Y3 amplitude is cosmological, combining third- and fourth-order aperture masses could break degeneracies that two-point analyses leave unresolved, because the two orders weight the density field differently; a tomographic measurement would test whether the offset is redshift-dependent.
  • Applying the estimator to tomographic bin pairs should be done with awareness of the quartic scaling with the number of bins; the paper's memory-light implementation makes this feasible, but runtime will still grow quickly.
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. This paper presents an efficient estimator for the shear four-point correlation function (4PCF) based on multipole decomposition, extending Porth et al. (2024) to fourth order and implementing it in the public orpheus package. The estimator is validated on Gaussian random fields, where the measured 4PCF is compared with a prediction built from the measured 2PCF, and on the SLICS N-body suite, where the fourth-order aperture mass obtained from the 4PCF is compared with a direct estimator. The paper then applies the estimator to DES Y3 data, measuring fourth-order aperture statistics in a non-tomographic setup. The authors report a significant detection of the connected E-mode fourth-order aperture mass (p = 4.09e-137 in Sect. 6.4) and note that the sampling distribution is strongly skewed and that internal covariance estimates are overconfident. They also find the DES Y3 amplitude to be lower than expectations from T17 mocks and discuss possible astrophysical explanations.

Significance. The main methodological contribution—a quadratic-scaling multipole estimator for the shear 4PCF with a low-memory implementation, validated against a direct estimator on N-body simulations and made publicly available—is useful and likely to be adopted for Stage III/IV analyses. The application to DES Y3 is a first step toward using fourth-order shear statistics in real data. However, the headline detection significance is not calibrated: it is based on a Gaussian likelihood and an internally estimated covariance that the paper itself shows to be overconfident, and no corrected p-value is provided. With appropriate recalibration or a more cautious claim, the paper would be a solid contribution.

major comments (3)
  1. [Sect. 6.4 and Sect. 6.3.2 / Fig. 4 / Fig. D.1] The headline p-value p = 4.09e-137 is computed with the internally estimated covariance matrix and a Gaussian likelihood. Section 6.3.2 and Fig. 4 (right panel) explicitly show that this internal covariance underestimates the true spread and overpredicts S/N, motivating the heuristic correction r_std,foot. Figure D.1 further shows that the sampling distribution of <M4_ap>_c is strongly skewed and non-Gaussian. No p-value is reported for cov_foot, cov_robust,foot, or for the r_std,foot-corrected covariance, and no simulation-based calibration of the null distribution is provided. Because the detection claim is the central result, this is a load-bearing gap. Please provide a calibrated p-value under a non-Gaussian/non-normal likelihood, or explicitly restate the result as a methodology demonstration without a quantitative detection significance.
  2. [Sect. 6.4] The sentence 'using the internally estimated covariance matrix of <M4_x>' appears to use the wrong statistic for the E-mode null test: in the notation of Sect. 5.1 and Eq. (35), <M4_x> is not the pure E-mode statistic <M4_ap>_c. If this is literal, the reported p-value is computed with the covariance of a different quantity and does not support the stated claim. If it is a typo, it must be corrected. This clarification is needed for reproducibility of the main quantitative result.
  3. [Sect. 4.2 / Fig. 2] The Gaussian-field validation compares the multipole-based 4PCF to a 4PCF built from the 2PCF measured on the same mocks, so it mainly demonstrates internal consistency rather than unbiasedness with respect to the true theoretical 4PCF. The small differences shown in the bottom row are attributed to noise, and the quoted percent-level agreement is not a fully independent test. The SLICS/direct-estimator comparison in Sect. 5.3 is stronger and more decisive; please describe the GRF test accordingly and state explicitly that the Gaussian truth is reconstructed from the same realizations.
minor comments (5)
  1. [Eq. (4)] The fourth factor in the definition of Γ^P_0 uses X3 twice: the third factor should be γ(X2; ζ2), not γ(X3; ζ2). Please fix this typo, which propagates to the index conventions.
  2. [Eq. (36)] The denominator contains 'γ_t,k γ_t,k l'; this appears to be a typo for γ_t,k γ_t,l. Please correct.
  3. [References] The companion paper SR25 (Silvestre-Rosello et al. 2025) is cited as 'submitted to A&A, and published on arXiv' but no arXiv identifier or DOI is given. Since Eq. (34) and the binning recommendations rely on SR25, please provide a complete reference.
  4. [Abstract / Sect. 1] The statement 'We make our estimator code available on GitHub as part of the orpheus package/github-square' does not contain a working link or repository identifier. Please add the actual URL or a footnote.
  5. [Sect. 6.3.1] The mock footprints use octant cuts rather than the actual DES Y3 geometry. The text notes that this may cause 'a slight misestimation' for large scales. Given that the covariance is used for the signal-to-noise assessment, please add a quantitative test of the geometry sensitivity (e.g., using the DES Y3 mask with the direct estimator) or state more explicitly the expected impact.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the 4PCF estimator is derived from first principles and validated against external N-body simulations; the flagged covariance overconfidence is a statistical calibration concern, not a circular step.

full rationale

The central derivation chain is independent of its outputs. The multipole-based 4PCF estimator (Sect. 3) is derived algebraically from the bin-averaged 4PCF definitions and the ×-projection, with no parameter fitted to the data. The Gaussian random field validation (Sect. 4) compares the estimator to the Wick-theorem 4PCF reconstructed from the measured 2PCF on the same mocks; this is a self-consistency check of the multipole expansion, not a circular derivation, because the 4PCF estimator is not defined in terms of the 2PCF and no parameter is adjusted to force agreement. The aperture-mass conversion (Sect. 5) adopts filter functions from the companion paper SR25, but these are mathematical identities rather than empirical claims, and the full pipeline is cross-validated on the independent SLICS N-body suite against a direct estimator (Sect. 5.3). The DES Y3 detection (Sect. 6.4) uses a covariance estimated from T17 mocks and internal patches; the paper itself highlights that the internal covariance is overconfident and the sampling distribution non-Gaussian (Sect. 6.3.2, Fig. D.1). That is a statistical calibration and correctness risk, not circularity: the reported p-value does not reduce to a fitted input or to a self-citation. No load-bearing self-citation is invoked to forbid alternative estimators or interpretations. Therefore the paper's core estimator and validation are self-contained against external benchmarks, and no circular step is identifiable.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The estimator itself is self-contained and introduces no new physical entities; the listed free parameters are resolution and binning choices, and the axioms are standard lensing/statistical assumptions plus the fidelity of the T17 mock covariance. The main burden is carried by the mock-based covariance and the multipole truncation.

free parameters (4)
  • n_max = 15
    Multipole truncation order for the 4PCF expansion, chosen from convergence tests on Gaussian and SLICS mocks (Sect. 4.2, 5.3); affects accuracy of the recovered 4PCF and aperture statistics.
  • r_min,Delta = 40
    Tree-mesh resolution parameter that controls the approximation in the DES Y3 measurement (Sect. 6.2); chosen to balance runtime and accuracy.
  • r_std,foot = not quoted
    Heuristic correction factor for internal covariance S/N, calibrated on the T17 ensemble (Sect. 6.3.2); not fitted to DES Y3 data but affects the reported S/N values.
  • 4PCF binning scheme = 65 log radial bins in [0.25',166.29']; 129 linear angular bins; 24/31 aperture radii in [1',32']
    Binning choices validated in SR25 and Sect. 5.3 to give ~2% accuracy at theta > 4 arcmin; they determine integration bias, especially at small radii.
assumptions (5)
  • domain assumption Flat-sky approximation on 100 overlapping patches covers DES Y3 and the T17 mocks
    Used in Sect. 6.2 for the 4PCF measurement; the paper states 'for which the flat-sky approximation holds'.
  • domain assumption Statistical homogeneity and isotropy of the shear field
    Required for the natural-component / multipole description and bin averaging (Sect. 2.2, Eq. 14).
  • domain assumption Observed ellipticities can be modelled as reduced shear plus shape noise (Seitz & Schneider 1997)
    Used to generate mock catalogues (Sect. 6.3.1, step 4); any failure of this model for real DES Y3 galaxies biases the covariance or the signal.
  • domain assumption T17 ray-tracing mocks with octant geometry provide a faithful covariance model for DES Y3
    Sect. 6.3: cosmology differs from DES Y3 best fit (Omega_m=0.279, sigma8=0.82) and geometry is simplified; the authors argue effects are small but this underpins the covariance and the amplitude comparison.
  • domain assumption The 'base catalogue' resampling of DES Y3 weights, shapes and n(z) preserves the data's noise properties
    Sect. 6.3.1: the authors rely on this to make mocks statistically equivalent to DES Y3 for covariance estimation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards an application of fourth-order shear statistics II: Efficient estimation of fourth-order shear correlation functions and an application to the DES Y3 data." pith.science (2026). https://pith.science/paper/Q5UM3BKL

@misc{pith2026250907974,
  author       = {Pith},
  title        = {Pith review of: Towards an application of fourth-order shear statistics II: Efficient estimation of fourth-order shear correlation functions and an application to the DES Y3 data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Q5UM3BKL}},
  note         = {Machine review of arXiv:2509.07974}
}
abstract

Higher-order lensing statistics contain a wealth of cosmological information that is not captured by second-order statistics. Stage-III lensing surveys have sufficient statistical power to significantly detect cumulant-based statistics up to fourth order. We derive and validate an efficient estimation procedure for the four-point correlation function (4PCF) of polar fields such as weak lensing shear. We then use our approach to measure the shear 4PCF and the fourth-order aperture mass statistics in the DES Y3 survey. We construct an efficient estimator for fourth-order shear statistics which builds on the multipole decomposition of the shear 4PCF. We then validate our estimator on mock ellipticity catalogues obtained from Gaussian random fields and on realistic $N$-body simulations. Finally, we apply our estimator to the DES Y3 data and present a measurement of the fourth-order aperture statistics in a non-tomographic setup. Due to its quadratic scaling, our estimator provides a significant speed-up over hypothetical brute force or tree-based estimation methods of the shear 4PCF. We report a significant detection of the connected part of the fourth-order aperture mass in the DES Y3 data. We find the sampling distribution of the fourth-order aperture mass to be significantly skewed. We make our estimator code available on GitHub as part of the orpheus package.

Figures

Figures reproduced from arXiv: 2509.07974 by the authors.

Figure 1
Figure 1. Parametrization of a quadruplet of shears used in this work. For some shear at position X0, we denote the connecting lines to the other shears at positions X𝑖 as 𝝑𝑖 and the enclosing angles as 𝜙12 and 𝜙13. The red dashed lines show the directions of the ×-projection Eq. (13) for which the three projection axes intersect in X0 where the associated projection kernel, 𝑊, is defined as 𝑊 (𝜒) ≡ 3Ωm𝐻 2 0 2𝑐 2 𝑓𝐾 (𝜒) 𝑎(𝜒) … view at source ↗
Figure 2
Figure 2. Estimator validation using the shear 4PCF from an ensemble of GRFs. Top row: Absolute value of 4PCF multipoles for three different natural components and different radial configurations, normalised by their largest value. The square surrounding the region with |𝑛𝑖 | ≤ 15 indicates the multipole cuts used for our default analysis. Middle row: Convergence of the 4PCF in real space for the same natural components and r… view at source ↗
Figure 3
Figure 3. Validation of the fourth-order aperture statistics on the SLICS simulation suite. Left: The independent aperture measures (35) obtained by applying the transformations (34) to the estimated shear 4PCF. In the top panel, the real (imaginary) parts of the aperture measures are displayed as solid (dashed) lines which are colour-coded according to the structurally different 4PCF components. The black line shows the disc… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Left: Joint correlation matrix of the second, third, and fourth aperture mass cumulants in the T17 ensemble obtained from the footprints (lower triangle) or from the internal covariance estimates averaged over the footprints (upper triangle). Right: Signal-to-noise of …
Figure 5
Figure 5. Figure 5: Left: The fourth-order aperture statistics in the T17 ensemble (black) and in the DES Y3 data (other). In the upper panel, the sampling distribution of the 𝐸-mode is shown as a violin shape, where for the DES Y3 data the left part of each violin shows the area-rescaled…

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Fourth-order galaxy-galaxy-lensing: Theoretical framework and direct estimation

    astro-ph.CO 2026-04 unverdicted novelty 7.0 of 10

    The authors derive the fourth-order galaxy-galaxy lensing 4PCF and aperture statistics, implement a numerical pipeline and FFT estimator, and detect the connected ⟨N³ M_ap⟩ signal at SNR ~9 in stage IV mock data over ...

Reference graph

Works this paper leans on

69 extracted references · 35 canonical work pages · cited by 1 Pith paper

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archiveprefix author booktitle chapter edition editor howpublished institution eprint journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.all := #1 ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in " " * FUNCTION format....

  3. [3]

    A., et al

    Amon , A., Gruen , D., Troxel , M. A., et al. 2022, , 105, 023514

    2022

  4. [4]

    2023, , 526, 5530

    Anbajagane , D., Chang , C., Banerjee , A., et al. 2023, , 526, 5530

    work page 2023

  5. [5]

    2021, , 645, A104

    Asgari , M., Lin , C.-A., Joachimi , B., et al. 2021, , 645, A104

    2021

  6. [6]

    J., Refregier , A

    Bacon , D. J., Refregier , A. R., & Ellis , R. S. 2000, , 318, 625

    work page 2000

  7. [7]

    & Schneider , P

    Bartelmann , M. & Schneider , P. 2001, , 340, 291

    2001

  8. [8]

    2020, , 492, 3420

    Barthelemy , A., Codis , S., Uhlemann , C., Bernardeau , F., & Gavazzi , R. 2020, , 492, 3420

    work page 2020

Show all 69 references
  1. [9]

    D., Saust , A

    Blandford , R. D., Saust , A. B., Brainerd , T. G., & Villumsen , J. V. 1991, , 251, 600

    1991

  2. [10]

    A., Porth , L., Heydenreich , S., et al

    Burger , P. A., Porth , L., Heydenreich , S., et al. 2024, , 683, A103

    2024

  3. [11]

    & Szapudi , I

    Chen , G. & Szapudi , I. 2005, , 635, 743

    2005

  4. [12]

    G., Natarajan , P., Pen , U.-L., & Theuns , T

    Crittenden , R. G., Natarajan , P., Pen , U.-L., & Theuns , T. 2002, , 568, 20

    2002

  5. [13]

    2023, , 108, 123519

    Dalal , R., Li , X., Nicola , A., et al. 2023, , 108, 123519

    2023

  6. [14]

    2017, Gravitational Lensing (Cambridge University Press)

    Dodelson, S. 2017, Gravitational Lensing (Cambridge University Press)

    2017

  7. [15]

    2023, , 675, A120

    Euclid Collaboration: Ajani , V., Baldi , M., Barthelemy , A., et al. 2023, , 675, A120

    2023

  8. [16]

    2025, A&A, 697, A1

    Euclid Collaboration: Mellier , Y., Abdurro'uf , Acevedo Barroso , J., et al. 2025, A&A, 697, A1

    2025

  9. [17]

    T., Honscheid , K., et al

    Flaugher , B., Diehl , H. T., Honscheid , K., et al. 2015, , 150, 150

    2015

  10. [18]

    2014, , 441, 2725

    Fu , L., Kilbinger , M., Erben , T., et al. 2014, , 441, 2725

    2014

  11. [19]

    2021, , 504, 4312

    Gatti , M., Sheldon , E., Amon , A., et al. 2021, , 504, 4312

    2021

  12. [20]

    Gomes , R. C. H., Sugiyama , S., Jain , B., et al. 2025, arXiv:2503.03964

    2025

  13. [21]

    M., Hivon , E., Banday , A

    G \'o rski , K. M., Hivon , E., Banday , A. J., et al. 2005, , 622, 759

    2005

  14. [22]

    2018, , 481, 1337

    Harnois-D \'e raps , J., Amon , A., Choi , A., et al. 2018, , 481, 1337

    2018

  15. [23]

    2024, , 534, 3305

    Harnois-D \'e raps , J., Heydenreich , S., Giblin , B., et al. 2024, , 534, 3305

    2024

  16. [24]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357

    2020

  17. [25]

    2022, , 667, A125

    Heydenreich , S., Br \"u ck , B., Burger , P., et al. 2022, , 667, A125

    2022

  18. [26]

    2023, , 672, A44

    Heydenreich , S., Linke , L., Burger , P., & Schneider , P. 2023, , 672, A44

    2023

  19. [27]

    Hou , J., Slepian , Z., & Cahn , R. N. 2023, , 522, 5701

    2023

  20. [28]

    Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90

    2007

  21. [29]

    M., Tyson , J

    Ivezi \'c , Z ., Kahn , S. M., Tyson , J. A., et al. 2019, , 873, 111

    2019

  22. [30]

    & Van Waerbeke , L

    Jain , B. & Van Waerbeke , L. 2000, , 530, L1

    2000

  23. [31]

    2016, , 460, 2245

    Jarvis , M., Sheldon , E., Zuntz , J., et al. 2016, , 460, 2245

    2016

  24. [32]

    1992, , 388, 272

    Kaiser , N. 1992, , 388, 272

    1992

  25. [33]

    1995, , 439, L1

    Kaiser , N. 1995, , 439, L1

    1995

  26. [34]

    & Squires , G

    Kaiser , N. & Squires , G. 1993, , 404, 441

    1993

  27. [35]

    Kaiser , N., Wilson , G., & Luppino , G. A. 2000, arXiv:0003338

    2000

  28. [36]

    2015, Reports on Progress in Physics, 78, 086901

    Kilbinger , M. 2015, Reports on Progress in Physics, 78, 086901

    2015

  29. [37]

    M., Lim , E

    Kratochvil , J. M., Lim , E. A., Wang , S., et al. 2012, , 85, 103513

    2012

  30. [38]

    2023, , 108, 123518

    Li , X., Zhang , T., Sugiyama , S., et al. 2023, , 108, 123518

    2023

  31. [39]

    2018, , 56, 393

    Mandelbaum , R. 2018, , 56, 393

    2018

  32. [40]

    2021, , 505, 4249

    Myles , J., Alarcon , A., Amon , A., et al. 2021, , 505, 4249

    2021

  33. [41]

    2011, Journal of Machine Learning Research, 12, 2825

    Pedregosa, F., Varoquaux, G., Gramfort, A., et al. 2011, Journal of Machine Learning Research, 12, 2825

    2011

  34. [42]

    Philcox , O. H. E. 2025, , 111, 123534

    2025

  35. [43]

    Philcox , O. H. E. & Ereza , J. 2025, Philosophical Transactions of the Royal Society of London Series A, 383, 20240034

    2025

  36. [44]

    Philcox , O. H. E., Slepian , Z., Hou , J., et al. 2022, , 509, 2457

    2022

  37. [45]

    2024, , 689, A227

    Porth , L., Heydenreich , S., Burger , P., Linke , L., & Schneider , P. 2024, , 689, A227

    2024

  38. [46]

    & Smith , R

    Porth , L. & Smith , R. E. 2021, , 508, 3474

    2021

  39. [47]

    Rousseeuw , P. J. & van Driessen , K. 1999, Technometrics, 41, 212

    1999

  40. [48]

    2010, , 404, 350

    Rowe , B. 2010, , 404, 350

    2010

  41. [49]

    1996, , 283, 837

    Schneider , P. 1996, , 283, 837

    1996

  42. [50]

    2005, , 431, 9

    Schneider , P., Kilbinger , M., & Lombardi , M. 2005, , 431, 9

    2005

  43. [51]

    & Lombardi , M

    Schneider , P. & Lombardi , M. 2003, , 397, 809

    2003

  44. [52]

    1998, , 296, 873

    Schneider , P., van Waerbeke , L., Jain , B., & Kruse , G. 1998, , 296, 873

    1998

  45. [53]

    2002, , 396, 1

    Schneider , P., van Waerbeke , L., Kilbinger , M., & Mellier , Y. 2002, , 396, 1

    2002

  46. [54]

    F., Jarvis , M., Jain , B., et al

    Secco , L. F., Jarvis , M., Jain , B., et al. 2022 a , , 105, 103537

    2022

  47. [55]

    F., Samuroff , S., Krause , E., et al

    Secco , L. F., Samuroff , S., Krause , E., et al. 2022 b , , 105, 023515

    2022

  48. [56]

    & Schneider , P

    Seitz , C. & Schneider , P. 1997, , 318, 687

    1997

  49. [57]

    2025, submitted to , and published on arXiv

    Silvestre-Rosello , E., Porth , L., Linke , L., et al. 2025, submitted to , and published on arXiv

    2025

  50. [58]

    & Eisenstein , D

    Slepian , Z. & Eisenstein , D. J. 2015, , 454, 4142

    2015

  51. [59]

    Sugiyama , S., Gomes , R. C. H., & Jarvis , M. 2024, arXiv:2407.01798

    2024 arXiv

  52. [60]

    2023, RAS Techniques and Instruments, 2, 62

    Sunseri , J., Slepian , Z., Portillo , S., et al. 2023, RAS Techniques and Instruments, 2, 62

    2023

  53. [61]

    2017, , 850, 24

    Takahashi , R., Hamana , T., Shirasaki , M., et al. 2017, , 850, 24

    2017

  54. [62]

    2005, arXiv:0510346

    The Dark Energy Survey Collaboration . 2005, arXiv:0510346

    2005

  55. [63]

    2016, , 460, 1270

    The Dark Energy Survey Collaboration . 2016, , 460, 1270

    2016

  56. [64]

    A., Liu , J., & Shirasaki , M

    Thiele , L., Marques , G. A., Liu , J., & Shirasaki , M. 2023, , 108, 123526

    2023

  57. [65]

    2000, , 358, 30

    Van Waerbeke , L., Mellier , Y., Erben , T., et al. 2000, , 358, 30

    2000

  58. [66]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261

    2020

  59. [67]

    M., Tyson , J

    Wittman , D. M., Tyson , J. A., Kirkman , D., Dell'Antonio , I., & Bernstein , G. 2000, , 405, 143

    2000

  60. [68]

    H., St \"o lzner , B., Asgari , M., et al

    Wright , A. H., St \"o lzner , B., Asgari , M., et al. 2025, arXiv:2503.19441

    2025 arXiv

  61. [69]

    2019, Journal of Open Source Software, 4, 1298

    Zonca, A., Singer, L., Lenz, D., et al. 2019, Journal of Open Source Software, 4, 1298

    2019

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.