Pith. sign in

REVIEW 4 major objections 4 minor 3 cited by

Approximate Ricci-flat Metrics for Calabi-Yau Manifolds

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Analytic Kähler potentials approximate Ricci-flat Calabi-Yau metrics to about one percent.

desk verdict Real analytic approximations for two CY families, but the few-percent Ricci-flat claim is not actually tested on the final metrics. read the letter →

arxiv 2506.15766 v1 pith:TU3HNA2T submitted 2025-06-18 hep-th cs.LGmath.DG

Seung-Joo Lee , Andre Lukas This is my paper
classification hep-thcs.LGmath.DG MSC 53C2553C5514J32
keywords Calabi-YaumanifoldRicci-flatmetricKählerpotentialDonaldsonAnsatzDworkquinticbi-cubicmachinelearningmetricscomplexstructuremoduli
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 claims that simple analytic formulas for Kähler potentials can approximate Ricci-flat metrics on two one-parameter families of Calabi-Yau three-folds: the Dwork quintic in $P^{4}$ and a bi-cubic hypersurface in $P^{2}$ x $P^{2}$. The potentials are assembled from Donaldson's Ansatz, with only a handful of symmetry-invariant terms, and their dependence on the complex structure modulus psi is captured by fitted exponentials of |psi|. For the quintic with degree-three sections, the fit reproduces the machine-learned Ricci-flat Kähler potential to below one percent for most values of |psi|, and the bi-cubic fits to a few percent, degrading to about ten percent at large |psi|. A sympathetic reader would care because these are analytic, not just numerical, metrics whose moduli dependence is explicit and surprisingly independent of the phase of psi.

What carries the argument

The central object is Donaldson's Ansatz K = (1/(pi k)) log(H_{I\bar J} s^I \bar $s^{{\bar J}}$) for the Kähler potential from sections of a line bundle, restricted to the symmetry-invariant subspace. Symmetry reduction leaves one, two, or three invariant monomial combinations I_r, and the machinery is a direct least-squares fit of these few coefficients to the numerically learned Kähler potential, then a second fit of the coefficient ratios as exponentials of |psi|.

What would settle it

Generate an independent Ricci-flat metric for the same quintic or bi-cubic using a high-k Donaldson balanced metric or an independent solver, then compare the fitted Kähler potential (2.17) or (3.9) against it at many points; if the relative deviation in K or the Monge-Ampere residual systematically exceeds the reported few percent, the central claim fails. A second check: compute K at fixed |psi| for phases differing within the fundamental domain and test whether the difference is actually zero beyond numerical noise.

Watch

Extended reading notes

Core claim

For the one-parameter Dwork quintic, the Kähler potential K = t/(2 pi) log(I0 + f1(|psi|) I1 + f2(|psi|) I2), with I0, I1, I2 the three G-invariant products of coordinate norms and f1, f2 exponentials in |psi|, approximates the Ricci-flat Kähler potential to within about one percent for most complex structure values. For the bi-cubic family along the diagonal in Kähler moduli space, K = t/pi log(I0 + f(|psi|) I1) with f a sum of two exponentials gives a few-percent fit. In both families the fitted coefficients, after removing Kähler-transformation ambiguities, depend only on |psi|, not on the phase of psi, within numerical accuracy.

Load-bearing premise

The machine-learned numerical Ricci-flat Kähler potentials are accurate enough that fitting a simple ansatz to them at the one-to-few-percent level really yields a one-to-few-percent approximation to the true Ricci-flat metric, rather than a fit to the network's own errors.

Editorial extensions

If this is right

  • The quintic Kähler potential (2.17) gives an analytic approximately Ricci-flat metric whose complex-structure dependence is explicit and free of any phase dependence.
  • The same method yields an analytic approximate metric for the bi-cubic along the line t1 = t2, with an asymptotic form that reduces to the sum of the two ambient Fubini-Study potentials at large |psi|.
  • For the quintic, the k=2 form already attains percent-level accuracy, so analytic approximations do not require high-degree bundle sections.
  • The known conifold singularity at psi=1 leaves a visible feature in the fitted ratio a1/a0 that the exponential fit does not capture, marking the limit of the simple analytic form.
  • Kähler-modulus dependence is fully captured for one-parameter Kähler spaces by overall scaling, but for h^{1,1}>1 the approach fixes a line in Kähler moduli space.

Reading between the lines

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

  • If the modulus-only dependence found here extends to other one-parameter Dwork-like families, it would suggest that simple log-of-invariant expressions are a general template for approximate Ricci-flat metrics on highly symmetric Calabi-Yau spaces.
  • The same fitting pipeline could be applied to other complete intersection Calabi-Yau three-folds with a single complex structure modulus, and its success there would test whether the exponential coefficient functions are a generic phenomenon.
  • The phase-independence claim could be checked more stringently by evaluating the fitted Kähler potential at pairs of psi values with identical |psi| and different phases well inside the fundamental domain, using independent high-accuracy numerical metrics not used in the fit.
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 / 4 minor

Summary. The paper proposes a two-step procedure for obtaining analytic approximations to Ricci-flat Kähler potentials on two one-parameter Calabi-Yau families: the Dwork quintic and a bi-cubic in P^2 x P^2. For each complex structure parameter psi, the authors use the machine-learning package cymetric to obtain numerical Ricci-flat Kähler potentials, then fit these to Donaldson-type ansätze built from G-invariant sections of low-degree line bundles. After removing Kähler-gauge redundancy by working with ratios of coefficients, they fit the psi-dependence by simple exponentials and arrive at closed-form expressions, Eqs. (2.15), (2.17), and (3.9), whose Kähler potentials depend only on |psi|. The paper claims these expressions are approximately Ricci-flat to within a few percent. The manuscript is clearly written and honestly discloses several caveats, including the non-freely-acting symmetry assumption and the conifold-region breakdown, but the central quantitative claim is not supported by the error measures actually reported.

Significance. If the final analytic potentials were validated directly, the paper would provide useful, simple approximate metrics for two standard Calabi-Yau families, with explicit complex-structure dependence—something currently lacking. The symmetry-group analysis and the reduction of Donaldson's ansatz to a few invariant functions are clean and of independent interest. The explicit formulas (2.15), (2.17), and (3.9) are concrete deliverables that others could test. However, the paper does not currently demonstrate that these final metrics are approximately Ricci-flat; it demonstrates only that per-psi fits to ML-generated Kähler potential values achieve percent-level pointwise accuracy. That distinction is essential, because the Kähler potential is gauge-dependent and pointwise errors do not control metric or curvature errors. The significance of the paper therefore hinges on a validation step that is missing.

major comments (4)
  1. [Sec. 2.3, Fig. 1 and Eq. (2.17)] The accuracy measure used throughout is the mean relative deviation |(K_i - K(p_i))/K(p_i)| of the Kähler potential values. This does not control the associated metric g_{α\bar{β}} = ∂_α ∂_{\bar{β}} K, nor does it quantify Ricci-flatness. The paper's own k=1 example illustrates the problem: the Fubini-Study ansatz (2.7) attains only about 2% K-error at large |ψ| (Fig. 1, blue) while being far from Ricci-flat. Therefore the statement that the metric from Eq. (2.17) is 'approximately Ricci-flat to within a few percent' is not supported by the reported data. The authors should evaluate the Monge-Ampere σ-loss (or another invariant measure such as scalar curvature or volume) directly for the final analytic potentials (2.15), (2.17), and (3.9), not only for the neural-network metrics.
  2. [Sec. 2.3, Eqs. (2.14), (2.16), and Sec. 3.3, Eq. (3.8)] The fit errors shown in Figs. 1 and 5 are for per-ψ best-fit parameters (c, a_r) for each data set. The final closed-form expressions, with f, f1, f2 given by the fitted exponentials, are never re-evaluated against the numerical Kähler potential data. Since the paper's central deliverable is the explicit analytic ψ-dependence, the paper should provide residuals for the final expressions and error bars or confidence intervals for the fitted coefficients. As it stands, the quality of the final analytic forms is not quantified.
  3. [Sec. 2.3 and Sec. 3.3] The numerical target K is produced by the cymetric package, which is authored in part by one of the present authors, and the ansätze are validated against the same data used for the fits. Because the cymetric metrics are themselves approximate (σ-loss in the range 0.01–0.025 for the quintic, Sec. 2.3), a few-percent fit accuracy does not by itself establish closeness to the true Ricci-flat metric. An independent cross-check is needed—for example, comparison with a balanced Donaldson metric at higher k, or evaluation of geometric invariants such as volumes or curvature norms—to confirm that the fitting target is itself accurate. Without such a check, the central claim rests on the ML proxy.
  4. [Sec. 3.3, Fig. 5] For the bi-cubic, the Kähler potential fit errors grow to about 10% at large |ψ|, and the authors attribute this to increased training loss (Sec. 3.3). The abstract and conclusion state the results are accurate to 'a few percent,' which is not representative of the bi-cubic over the full parameter range. The final claim should be qualified accordingly, or the bi-cubic training and fitting should be improved before making a uniform accuracy statement.
minor comments (4)
  1. [Sec. 3.1, Eq. (3.4)] The invariant P5 is written as x_0 x_1 x_1 y_0 y_1 y_2, which appears to be a typo for x_0 x_1 x_2 y_0 y_1 y_2. Please correct this, since the displayed polynomial is part of the definition of the invariant subspace.
  2. [Sec. 2.3, Fig. 4] The discussion of the irregular behavior of a1/a0 near the conifold point ψ=1 is qualitative. It would be useful to quantify the deviation from the fitted exponential (2.16) in that region, for instance by quoting a residual norm, so that readers can judge the size of the effect.
  3. [Sec. 3.3, Eq. (3.8)] The function f in Eq. (3.8) is chosen to enforce the asymptotic value 1, as stated in the text. This assumption should be stated already where the fit is introduced, not only after Eq. (3.8), and the sensitivity of the earlier fit quality to this asymptotic constraint should be reported.
  4. [Sec. 1, Introduction] The phrase 'analytic expressions for metrics' in the introduction could be misunderstood as exact closed-form Ricci-flat metrics. Since the final objects are empirical fits to numerical data, a more precise phrase such as 'simple empirical approximations' would better reflect the content of the paper.

Circularity Check

1 steps flagged · score 6.0 of 10

Central 'few-percent Ricci-flat' validation is in-sample: coefficients are fit to the cymetric data and the same data are used to report the error, so the agreement is by construction and the final analytic Kähler potentials are not independently tested.

  1. fitted input called prediction [Section 2.3 (quintic), Eqs. (2.10), (2.12), Fig. 1; also Section 3.3 (bi-cubic), Eq. (3.7), Fig. 5; Conclusion]
    "The mean error between the data and the Ansatz (2.12) with these best-fit values inserted is shown in Fig. 1 on the right (green). ... The quality of the fits is quite convincing, typically reproducing the data within a few percent. This indicates that the Kähler potentials obtained provide a good approximation to the Ricci-flat ones."

    The coefficients c and a_r are determined for each ψ by fitting the Ansatz to the pairs (p_i,K_i) read out from the cymetric network. The error plotted in Fig. 1 is then evaluated from the same Ansatz with those best-fit values inserted, on the same pairs, so it is the in-sample residual of the least-squares fit, not an independent test. The conclusion that the fitted potentials 'provide a good approximation to the Ricci-flat ones' therefore uses an agreement that is minimized by construction against the very data used as ground truth. The final analytic forms (2.17)/(3.9), assembled from the fitted exponentials (2.16)/(3.8), are never re-evaluated on held-out points or by their own Monge-Ampere loss, so the 'few percent' Ricci-flatness claim is not independently demonstrated.

full rationale

The central derivation chain is: train a cymetric network, read off (p_i, K_i), fit the coefficients c and a_r of Donaldson's Ansatz for each ψ, plot the training residual, and conclude that the fitted Ansatz is 'approximately Ricci-flat to within a few percent'; then fit exponentials to the ratios a_1/a_0 and write the global Kähler potentials (2.17)/(3.9). The only quantitative evidence for the Ricci-flatness claim is the in-sample fit error, which is computed on the same data used to fix the coefficients, so the agreement is a training residual, not an independent verification. The final closed forms are never checked against held-out points or by their own Monge-Ampere loss; small relative error in K does not control the metric or Ricci-flatness, a further correctness gap but not itself a circularity. I did not count the cymetric self-citation (Refs. [14,16], one by a present author) as a separate circular step, because the package is published and code-based and the paper reports its networks' Monge-Ampere losses; the weakness there is lack of independent cross-validation rather than by-construction reduction. The G-invariance assumption is explicitly flagged as an assumption, not disguised as a result. Overall, the central validation reduces to a training error, while the Donaldson Ansatz and the numerical Ricci-flat computation provide independent content, giving partial circularity.

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

The central output is an empirical fit: numerical Kähler potentials from the authors' own cymetric package are fit to a symmetry-reduced Donaldson Ansatz, and the resulting coefficient ratios are fit to exponentials in |psi|. All exponential constants and the raw per-psi coefficients are fitted to data. No new physical entities are introduced.

free parameters (4)
  • Exponential fit constants, quintic k=2 f(|psi|) = c0=0.141, c1=1.453, lambda=0.348 (Eq. 2.14)
    Fitted to the ratio a1/a0 obtained from per-psi least-squares fits to the cymetric Kähler potentials.
  • Exponential fit constants, quintic k=3 f1(|psi|), f2(|psi|) = c0=0.121, c1=0.605, lambda=0.264; c0=0.106, c1=0.783, lambda=0.352 (Eq. 2.16)
    Fitted to ratios a1/a0 and a2/a0.
  • Exponential fit constants, bi-cubic f(|psi|) = 1.0 (imposed), 0.675, 2.959, 0.238, 0.122 (Eq. 3.8)
    Fitted to a1/a0; the constant 1 is imposed by the |psi| to infinity limit (Eq. 3.10).
  • Per-psi Donaldson Ansatz coefficients a0(psi), a1(psi), a2(psi) and c = not tabulated; c approximately 1/(pi k)
    For each of 301 psi values the cymetric Kähler potential is fit to Eqs. (2.10) or (2.12); these raw fitted parameters generate the ratios used to define f.
assumptions (4)
  • domain assumption The Ricci-flat Kähler potential is G-invariant for the non-freely-acting symmetry group G (Z5^1 x S5 for the quintic, Z3^x x Z3^y x S3 for the bi-cubic).
    Assumed in Sec. 2 (p.3 of the preprint) for the quintic; the authors note the free-action argument does not apply and justify it only a posteriori by fit success.
  • domain assumption The cymetric neural-network Kähler potentials approximate the true Ricci-flat Kähler potentials well enough (Monge-Ampere loss 0.01 to 0.025) that least-squares fits at the few-percent level inherit the same accuracy.
    Invoked in Sec. 2.3 and 3.3; the paper provides no independent verification of the ML metric against other methods.
  • domain assumption Donaldson's Ansatz with low line-bundle powers k=2 and k=3 (i.e., a span of two or three G-invariant monomials) is sufficiently expressive to represent the Ricci-flat Kähler potential to percent-level accuracy.
    The basis of Eqs. (2.10) and (2.12); the fit errors support it but it is not proven and the paper notes the conifold region is not captured.
  • standard math Yau's theorem and Donaldson's convergence theorem for balanced metrics.
    Standard background results cited in the Introduction, not proved in the paper.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Approximate Ricci-flat Metrics for Calabi-Yau Manifolds." pith.science (2026). https://pith.science/paper/TU3HNA2T

@misc{pith2026250615766,
  author       = {Pith},
  title        = {Pith review of: Approximate Ricci-flat Metrics for Calabi-Yau Manifolds},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/TU3HNA2T}},
  note         = {Machine review of arXiv:2506.15766}
}
abstract

We outline a method to determine analytic K\"ahler potentials with associated approximately Ricci-flat K\"ahler metrics on Calabi-Yau manifolds. Key ingredients are numerically calculating Ricci-flat K\"ahler potentials via machine learning techniques and fitting the numerical results to Donaldson's Ansatz. We apply this method to the Dwork family of quintic hypersurfaces in $\mathbb{P}^4$ and an analogous one-parameter family of bi-cubic CY hypersurfaces in $\mathbb{P}^2\times\mathbb{P}^2$. In each case, a relatively simple analytic expression is obtained for the approximately Ricci-flat K\"ahler potentials, including the explicit dependence on the complex structure parameter. We find that these K\"ahler potentials only depend on the modulus of the complex structure parameter.

Figures

Figures reproduced from arXiv: 2506.15766 by the authors.

Figure 1
Figure 1. Left plot: Final σ-loss as a function of |ψ| for the quintic. Right plot: Mean of the relative error |(Ki − K(pi))/K(pi)| as a function of |ψ| for the K¨ahler k = 1 potential (2.7)(blue), the k = 2 K¨ahler potential (2.10) (yellow) and the k = 3 K¨ahler potential (2.12) (green). Donaldson’s Ansatz for k = 2 can already provide an approximation to the Ricci-flat K¨ahler potential at the percent level, at least for ou… view at source ↗
Figure 2
Figure 2. Left plot: Best-fit coefficients a0(ψ) (blue) and a1(ψ) (orange) and c (green) in the Ansatz (2.10) for the quintic K¨ahler potential, as a function of |ψ|. Right plot: Ratio a1(ψ)/a0(ψ) as a function of |ψ| and best-fit of the form (2.14). and its value, c ≃ 1 2π , is in good agreement with Eq. (1.1). The other features in [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Left plot: Best-fit coefficients a0(ψ) (yellow), a1(ψ) (red), a2(ψ) (green) and c (blue) in the Ansatz (2.12) for the quintic K¨ahler potential, as a function of |ψ|. Right plot: The ratios a1(ψ)/a0(ψ) (yellow points) and a2(ψ)/a0(ψ) (blue points) as a function of |ψ| and best fits of the form (2.16). inserted is shown in [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The ratios a1(ψ)/a0(ψ) (yellow) and a2(ψ)/a0(ψ) (blue) for the quintic, as a function of |ψ|, focusing on values ψ = 0.1 n, for n = 0, . . . , 30, near the conifold point ψ = 1. course not captured by our simple fit in Eq. (2.16).) It is tempting to attribute this beha…
Figure 5
Figure 5. Figure 5: Left plot: Mean of the relative deviation [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Ratio a1/a0 of the best-fit parameters in the Ansatz (3.7) for the bi-cubic K¨ahler potential, as a function of |ψ| (blue points). The red curve is a graph of the function f in Eq. (3.8), the best fit of a sum of two exponentials to the data. suggests that a1/a0 is a f…

Discussion (0). Sign in to comment.

Forward citations

Cited by 3 Pith papers

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

  1. Balanced Metrics Know About SYZ

    hep-th 2026-07 conditional novelty 6.0 of 10

    Ambient balanced metric coefficients on Calabi-Yau manifolds decay as |ψ|^{-f(α)} near the large complex structure limit, and the exponent function's Legendre transform gives the dual tropical potential expected from SYZ.

  2. Interpretable Analytic Calabi-Yau Metrics via Symbolic Distillation

    cs.LG 2026-02 conditional novelty 5.0 of 10

    The Ricci-flat metric's determinant ratio on the Dwork quintic is reproduced to R²=0.9994 by a five-term symbolic formula in two symmetric invariants, with moduli entering only through fitted coefficients.

  3. What to do with a Ricci-flat Calabi--Yau metric?

    hep-th 2026-05 unverdicted novelty 3.0 of 10

    Numerical Ricci-flat Calabi–Yau metrics turn string compactifications from topological existence statements into computable geometries, unlocking normalized couplings, spectra, and metric-level tests of mirror symmetry.

Reference graph

Works this paper leans on

20 extracted references · 5 canonical work pages · cited by 3 Pith papers

  1. [12]

    Moduli- dependent Calabi-Yau and SU(3)-structure metrics from Machine Learning,

    L. B. Anderson, M. Gerdes, J. Gray, S. Krippendorf, N. Raghuram and F. Ruehle, “Moduli- dependent Calabi-Yau and SU(3)-structure metrics from Machine Learning,” JHEP05(2021), 013 [arXiv:2012.04656 [hep-th]]

    arXiv 2021

  2. [1]

    Some numerical results in complex differential geometry,

    S. K. Donaldson “Some numerical results in complex differential geometry,” 2005

    work page 2005

  3. [2]

    Numerical solution to the hermitian Yang-Mills equation on the Fermat quintic,

    M. R. Douglas, R. L. Karp, S. Lukic and R. Reinbacher, “Numerical solution to the hermitian Yang-Mills equation on the Fermat quintic,” JHEP12(2007), 083 [arXiv:hep-th/0606261 [hep- th]]

    arXiv 2007

  4. [3]

    Numerical Calabi-Yau metrics,

    M. R. Douglas, R. L. Karp, S. Lukic and R. Reinbacher, “Numerical Calabi-Yau metrics,” J. Math. Phys.49(2008), 032302 [arXiv:hep-th/0612075 [hep-th]]

    arXiv 2008

  5. [4]

    Calabi-Yau Metrics for Quotients and Complete Intersections,

    V. Braun, T. Brelidze, M. R. Douglas and B. A. Ovrut, “Calabi-Yau Metrics for Quotients and Complete Intersections,” JHEP05(2008), 080 [arXiv:0712.3563 [hep-th]]

    arXiv 2008

  6. [5]

    Eigenvalues and Eigenfunctions of the Scalar Laplace Operator on Calabi-Yau Manifolds,

    V. Braun, T. Brelidze, M. R. Douglas and B. A. Ovrut, “Eigenvalues and Eigenfunctions of the Scalar Laplace Operator on Calabi-Yau Manifolds,” JHEP07(2008), 120 [arXiv:0805.3689 [hep- th]]

    arXiv 2008

  7. [6]

    Numerical Hermitian Yang-Mills Connec- tions and Vector Bundle Stability in Heterotic Theories,

    L. B. Anderson, V. Braun, R. L. Karp and B. A. Ovrut, “Numerical Hermitian Yang-Mills Connec- tions and Vector Bundle Stability in Heterotic Theories,” JHEP06(2010), 107 [arXiv:1004.4399 [hep-th]]

    arXiv 2010

  8. [7]

    Numerical Hermitian Yang-Mills Connections and Kahler Cone Substructure,

    L. B. Anderson, V. Braun and B. A. Ovrut, “Numerical Hermitian Yang-Mills Connections and Kahler Cone Substructure,” JHEP01(2012), 014 [arXiv:1103.3041 [hep-th]]. 14

    arXiv 2012

Show all 20 references
  1. [8]

    Energy functionals for Calabi-Yau metrics,

    M. Headrick and A. Nassar, “Energy functionals for Calabi-Yau metrics,” Adv. Theor. Math. Phys. 17(2013) no.5, 867-902 [arXiv:0908.2635 [hep-th]]

    2013 arXiv

  2. [9]

    Numerical Metrics, Curvature Expansions and Calabi-Yau Manifolds,

    W. Cui and J. Gray, “Numerical Metrics, Curvature Expansions and Calabi-Yau Manifolds,” JHEP 05(2020), 044 [arXiv:1912.11068 [hep-th]]

    2020 arXiv

  3. [10]

    Machine Learning Calabi–Yau Metrics,

    A. Ashmore, Y. H. He and B. A. Ovrut, “Machine Learning Calabi–Yau Metrics,” Fortsch. Phys. 68(2020) no.9, 2000068 [arXiv:1910.08605 [hep-th]]

    2020 arXiv

  4. [11]

    Numerical Calabi-Yau metrics from holomorphic networks,

    M. R. Douglas, S. Lakshminarasimhan and Y. Qi, “Numerical Calabi-Yau metrics from holomorphic networks,” [arXiv:2012.04797 [hep-th]]

    2012 arXiv

  5. [13]

    Neural network approximations for Calabi-Yau metrics,

    V. Jejjala, D. K. Mayorga Pena and C. Mishra, “Neural network approximations for Calabi-Yau metrics,” JHEP08(2022), 105 [arXiv:2012.15821 [hep-th]]

    2022 arXiv

  6. [14]

    Learning Size and Shape of Calabi-Yau Spaces,

    M. Larfors, A. Lukas, F. Ruehle and R. Schneider, “Learning Size and Shape of Calabi-Yau Spaces,” [arXiv:2111.01436 [hep-th]]

    arXiv

  7. [15]

    Calabi-Yau Metrics, Energy Functionals and Machine-Learning,

    A. Ashmore, L. Calmon, Y. H. He and B. A. Ovrut, “Calabi-Yau Metrics, Energy Functionals and Machine-Learning,” International Journal of Data Science in the Mathematical Sciences1(2023) no.1, 49-61 [arXiv:2112.10872 [hep-th]]

    2023 arXiv

  8. [16]

    Numerical metrics for complete intersec- tion and Kreuzer–Skarke Calabi–Yau manifolds,

    M. Larfors, A. Lukas, F. Ruehle and R. Schneider, “Numerical metrics for complete intersec- tion and Kreuzer–Skarke Calabi–Yau manifolds,” Mach. Learn. Sci. Tech.3(2022) no.3, 035014 [arXiv:2205.13408 [hep-th]]

    2022 arXiv

  9. [17]

    Attractors, Geodesics, and the Geometry of Moduli Spaces,

    F. Ruehle and B. Sung, “Attractors, Geodesics, and the Geometry of Moduli Spaces,” [arXiv:2408.00830 [hep-th]]

    arXiv

  10. [18]

    Computation of quark masses from string theory,

    A. Constantin, C. S. Fraser-Taliente, T. R. Harvey, A. Lukas and B. Ovrut, “Computation of quark masses from string theory,” Nucl. Phys. B1010(2025), 116778 [arXiv:2402.01615 [hep-th]]

    2025 arXiv

  11. [19]

    Calabi-Yau manifolds: A Bestiary for physicists

    T. Hubsch, “Calabi-Yau manifolds: A Bestiary for physicists”, World Scientific Singapore 1994

    1994

  12. [20]

    Compact Manifolds with Special Holonomy

    D. Joyce, “Compact Manifolds with Special Holonomy”, OUP 2000. 15

    2000

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.