Empirical investigation of nuclear correlation function distributions in lattice QCD

Cagin Yunus; Rohan Kanchana; William Detmold

REVIEW 4 major objections 5 minor 29 references

Empirical investigation of nuclear correlation function distributions in lattice QCD

T0 review · 4 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read Lattice QCD correlation functions for nuclei follow an O(N)-model distribution, with effective N≈2/B.

desk verdict The O(N) distribution really does describe these QCD histograms, but the B=4 anchor of the N~2/B scaling is not verifiable as written because the alpha fit table duplicates the deuteron table. read the letter →

arxiv 2508.20378 v1 pith:M5VLVYIC submitted 2025-08-28 hep-lat

William Detmold , Rohan Kanchana , Cagin Yunus This is my paper · ORCID

classification hep-lat

keywords latticeQCDnuclearcorrelationfunctionsO(N)modelsignal-to-noiseproblembaryonnumberMonteCarlodistributionstwo-pointvonMisesdistribution

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

Lattice QCD correlation functions for the proton, deuteron, triton, and helium-4 are dominated at large times by Monte-Carlo noise, but this paper finds that the full probability distribution of that noise has a simple shape. The paper fits the empirically sampled histograms of the zero-momentum two-point correlation functions to the analytic distribution derived for the O(N) model—a field theory with N equivalent components—and reports that the fit works for all four baryon numbers at every studied temporal separation, with typical normalized residuals below 0.2. The fitted value of the parameter N is non-integer and, at large separations, asymptotes to about 2/B, so a deuteron behaves like an O(1) distribution, a triton like O(0.6), and an alpha like O(0.4). If the result holds, larger nuclei would push the effective description toward the O(N→0) limit, whose statistical mechanics is connected to self-avoiding random walks.

What carries the argument

The object doing the work is the analytic distribution P(x;ω+,ω−,N) of Eq. (2) for the zero-momentum O(N)-invariant two-point function, written in terms of a modified Bessel function of the second kind. It has two scale parameters ω± that become equal at late times and an effective field count N that controls the shape near x=0 and the tails. The paper turns this into a fitting template: Monte-Carlo samples are concatenated, the lowest and highest 10% are removed, a 1600-bin histogram is built, and an L1-norm fit with a free normalization extracts N and ω± for each B and t. The extracted N is the quantity that carries the result—its large-time value is the 2/B scaling.

What would settle it

Discard the 10% tail cut or fix the histogram normalization to unit integral and refit N_B(t) at late t; the 2/B claim fails if the asymptotic N values change by more than the quoted 17–83% bootstrap uncertainties, or if normalized residuals grow above the reported 0.2 level.

Watch

Extended reading notes

Core claim

The central claim is that the real part of the zero-momentum QCD two-point correlation function, sampled by lattice Monte Carlo, obeys the same probability law as the simplest two-point function of the O(N) model. For each baryon number B∈{1,2,3,4} and each time separation, the empirical histogram is fit to Eq. (2), the analytic distribution with parameters N, ω+, and ω−, and the fits reproduce the histogram's shape and magnitude at all studied times. The paper's second claim is that the best-fit N at large time saturates to N≈2/B, while N(t) approaches that value exponentially with a scale near the pion mass or the QCD scale. The phase of the correlation function is separately described by

Load-bearing premise

The load-bearing premise is that the empirical histogram produced by trimming 10% of samples, binning into 1600 bins, and allowing a free normalization in the L1 fit preserves the true baryon-dependent shape of the correlation-function distribution; if the trimmed tails or the fitted normalization carry that dependence, the O(N) descriptions and the N≈2/B scaling could be artifacts.

Editorial extensions

If this is right

If the O(N) form is correct, the Monte-Carlo noise in nuclear correlation functions has a known analytic model, so estimators for the underlying path integrals can be designed using the distribution rather than only the mean.
The 2/B asymptotic makes multi-baryon correlators approach the O(N→0) model as B grows, linking nuclear correlation-function statistics to the self-avoiding-walk universality class.
The exponential approach of N_B(t) to its asymptote gives a data-driven way to identify when the signal-to-noise problem has entered its asymptotic regime for each baryon number.
Because the distributions broaden and become strongly non-Gaussian with t, any error propagation for nuclear correlators should use the full distribution or its fitted form.

Reading between the lines

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

Testable extension: apply the same fit to meson (B=0) correlation functions; if N≈2/B continues to B=0, the effective N would diverge or lose meaning, which would distinguish the scaling from an accidental fit.
The results come from a single lattice ensemble at mπ≈450 MeV; a check at physical quark masses and different lattice spacings would show whether N≈2/B is a universal feature of QCD or an artifact of this ensemble.
The fitted normalization and 10% tail cut are analysis choices that could affect N; an unbinned likelihood fit on the complete sample would either confirm or weaken the 2/B relation.
Moment ratios (for example variance and kurtosis) implied by the asymptotic O(N) distribution could be compared directly with the raw samples, avoiding histogram resolution entirely.

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.

Desk Editor's Note

read the letter

Quick read: the paper does something genuinely useful—it takes the O(N)-model correlation-function distribution from Yunus & Detmold and actually compares it to empirical lattice QCD histograms for baryon number B=1..4. The fit quality is impressive: the model captures the shape of the distributions across t, the residuals are small, and the phase analysis with von Mises fits is a nice extra. That descriptive result alone is worth knowing, and the comparison to earlier proposed distributions is fair and useful.

The central claim, however, is the N(t->infty) ~ 2/B scaling, and here the manuscript has a serious problem. Table V, labelled 'Fit parameters for the alpha', is numerically identical to Table III for the deuteron. I checked: every entry matches, including N=2.49 at t=1 and N=1.053 at t=48. So the paper does not actually contain the alpha fit parameters shown in Fig. 5. The B=4 point is exactly what makes the inverse-linear relation look well resolved, and it is currently unsupported by the tabulated data. Either the table was copied in error or the figure uses a different set of fits; either way, a referee cannot verify the anchor of the scaling claim.

There is also a smaller internal inconsistency about the normalization alpha. Section IIIB says alpha is determined in the L1 fit (Eq. 8), but then later says alpha is not a free parameter and is set by equivalent normalization. Those two statements cannot both be true. If alpha is free, shape mismatch can be absorbed, which weakens the interpretation of the fitted N. The authors need to clarify which procedure they actually used.

The acceptance criterion (|residual| < 0.3 in most bins) is loose, and the robustness of the 10% truncation and 1600-bin choice is only qualitatively asserted in footnote 1. These are not fatal on their own, but they matter because the asymptotic N is extracted from the same histograms that are being fit. None of this kills the qualitative claim that the O(N) distribution describes the QCD histograms—that seems robust from the figures—but the N~2/B relation is a fitted scaling, not an independent check, and its B=4 value is presently unverifiable.

The paper deserves peer review: the empirical observation is interesting, the errors are fixable, and the field would benefit from seeing the corrected version. But it should not be accepted as is; the Table V duplication in particular must be fixed and the alpha normalization clarified.

Referee Report

4 major / 5 minor

Summary. The paper analyzes the empirical probability distributions of zero-momentum two-point correlation functions computed in lattice QCD for systems with baryon number B = 1, 2, 3, 4 (proton, deuteron, triton, alpha). Using a single NPLQCD ensemble, the authors build histograms of Monte-Carlo samples and fit them to the O(N)-model distribution of Eq. (2), with parameters N, omega_+, omega_-. They report that the O(N) form describes the QCD histograms well across all temporal separations, and that the extracted parameter N_B(t) saturates at large t to values approximately scaling as 2/B. The paper also analyzes phase distributions via von Mises fits and speculates on a connection to O(N -> 0) self-avoiding walk behavior.

Significance. If the central empirical claim is robust, this is an intriguing and potentially useful observation: it would show that the noise distributions of QCD correlation functions in the multi-baryon sector are quantitatively captured by a one-parameter family of O(N) distributions, and it would suggest a specific, testable scaling N ~ 2/B at large B. The paper uses genuine lattice data and presents a new comparison that goes beyond previous signal-to-noise analyses. The strengths include the explicit presentation of the fitted distributions in extensive appendix figures, the use of phase distributions as a secondary check, and a clearly stated, falsifiable scaling law. However, the current manuscript contains an internal inconsistency in the tabulated alpha fits, and the statistical acceptance criteria are not stringent enough to fully support the 'accurate description' claim.

major comments (4)

[Appendix A, Tables III and V] Table V, labelled 'Fit parameters for the alpha,' is numerically identical to Table III for the deuteron for every entry (e.g., t=1: N=2.49, omega_+=1.43; t=48: N=1.053). This makes the tabulated support for the alpha fits unverifiable. Since the B=4 column in Fig. 5 and the N_alpha(infinity)=0.433 value in Table I are the anchor that makes the N^{-1} = c0 + c1 B relation well resolved, this duplication is load-bearing. The authors must provide the correct alpha fit tables, or at minimum state explicitly that the alpha results shown in Fig. 5 are not tabulated and why.
[Sec. III B, Eq. (8) and footnote 1] The fitting methodology removes the lowest and highest 10% of samples before histogramming and then includes a free normalization alpha in Eq. (8). This is internally inconsistent with the later statement in Sec. III B that 'alpha is not a free parameter but is set such that the model and empirical data are normalised equivalently.' If alpha is indeed free, the shape comparison is weakened because alpha absorbs overall normalization mismatch and can shift the best-fit N; if alpha is fixed, then Eq. (8) as written is not the actual loss function. The footnote's qualitative claim of robustness to Delta_cut and Nbins is not a substitute for a systematic scan, especially because the extracted N_B(t) values feed directly into the central N ~ 2/B result. Please clarify the role of alpha and provide a quantitative robustness study for the truncation/binning choices, at least for the asymptotic
[Sec. III B and Fig. 3] The acceptance criterion for the fits is |delta_{B,t}(x)| < 0.3 in 'most bins' (footnote 4), and the reported average residual is typically below 0.2. This is a weak goodness-of-fit measure, particularly because the bins are highly correlated and the zero-bin singularity is excluded from the criterion. The claim that the O(N) distribution 'accurately describes' the QCD distributions over all t deserves a more quantitative test, such as a chi-square per degree-of-freedom, a Kolmogorov-Smirnov test, or a likelihood-based comparison, with the number of effective degrees of freedom clearly stated. Without such a test, the fitted N values could be absorbing systematic shape mismatches rather than representing a physically meaningful parameter.
[Sec. III C, Table I] The extraction of N_B(infinity) for B=alpha uses only a constant fit over t in [25,48], because the time dependence is noisy. This is reasonable, but it makes the B=4 point particularly sensitive to the quality of the underlying histograms and to the binning/truncation choices. Given the duplication in Table V and the lack of a systematic binning study, the B=4 value of 0.433 is currently not independently supported. Please either provide the correct alpha table, or present the alpha fits in a form that allows the reader to verify N_alpha(t -> infinity).

minor comments (5)

[Sec. III B, step 3] Only 14 bootstrap resamples are used to estimate uncertainties. While the paper notes that doubling the number of bootstraps gives negligible differences, the 17%/83% quantiles from 14 samples are noisy. It would be helpful to state the seed/protocol and, if possible, increase to at least 100 resamples for the final fits.
[Sec. III C, Fig. 5] The lower panel of Fig. 5 shows log10(omega_+/-), but the text describes the parameters as omega_+ and omega_-. The axis label should be explicit about which parameter is plotted and in what units; currently it is easy to misread the two curves.
[Sec. III D] The phase analysis using von Mises distributions is a nice complement, but the R^2 values >0.99 for t>=4 should be supplemented by the fitted kappa uncertainties or a residual plot. The current Fig. 7 shows only the central values.
[General] The paper would benefit from a short description of how the correlation function signs are handled in the histogram: the text says the real part is used, but the effective mass discussion refers to 'absolute values' and 'phase of the correlation function.' Clarify exactly what quantity is histogrammed for the zero-bin region.
[Captions of Tables III-V] The tables are large and the uncertainty notation is compact. A note defining the 17%/83% quantile convention (as in Table I) would help readers interpret the many asymmetric uncertainties.

Circularity Check

1 steps flagged · score 6.0 of 10

Headline N~2/B claim is a post-fit summary of the free parameter N, not an independent prediction; Table V duplicates Table III, undermining the B=4 anchor.

fitted input called prediction [Abstract; Sec. III B (Eq. (8)); Sec. III C (Eq. (9), Fig. 6)]
"the behaviour of the baryon number $B$ QCD correlation function at large temporal separation is well-reproduced by the $O(N \sim 2/B)$ model distribution. ... To extract the asymptotic value, $N_B = N_B(t \to \infty)$, fits to the above exponential model and to a constant ... are performed. ... Qualitatively that suggests the intriguing scaling, $N_B(\infty) \sim 2/B$."

N is a free parameter in the L1 fit of Eq. (8) to each empirical histogram; the extracted $N_B(t)$ curves are then fit with Eq. (9) to obtain $N_B(\infty)$. The abstract's 'well-reproduced by $O(N \sim 2/B)$' is therefore a restatement of those fitted values, not a test with N fixed to 2/B. The validation uses a loose residual criterion ($|\delta|<0.3$ in most bins) that does not distinguish the $N\sim2/B$ relation from the fit simply returning its best N. The 2/B scaling is fit to the same extracted values it is then said to reproduce, so no independent prediction is tested.

full rationale

The paper is mostly a transparent fitting study, and the O(N) distribution in Eq. (2) is an external theoretical input from Refs. [17,18]; testing it against QCD data is not circular. However, the headline claim that QCD correlators are 'well-reproduced by the O(N ~ 2/B) model distribution' is a post-fit summary: N is fitted freely in Eq. (8), and the asymptotic values $N_B(\infty)$ are extracted from those same fits, with the 2/B relation then fit to those extracted values. The only independent content is the goodness-of-fit residual, which is reported with a loose threshold (|δ|<0.3 in most bins). I therefore flag one partial circularity of the 'fitted input called prediction' type. Separately, there is a serious internal-consistency failure that is not circularity but weakens the scaling claim: Table V ('Fit parameters for the α') is numerically identical to Table III ('Fit parameters for the deuteron') at every listed t (e.g., t=1: N=2.49; t=48: N=1.053), removing the tabulated support for the B=4 anchor. The paper's own footnote 1 gives only a qualitative robustness statement, and Sec. III C admits that for α 'the saturation to a constant is less clear and there are significant fluctuations as t varies.' These issues lower confidence in the central scaling but are not additional circular steps.

Assumptions & free parameters 8 free parameters · 6 assumptions · 0 invented entities

The central claim rests on the O(N) model distribution as an assumed functional form, on the validity of histogram truncation and binning, on the free normalization in the fit, and on fitting N_B(t) to extract the asymptotic N~2/B scaling. No new entities are introduced.

free parameters (8)

N_B(t) (O(N) parameter per baryon and time) = e.g., N_p(inf)=2.12, N_d(inf)=0.956, N_t(inf)=0.588, N_alpha(inf)=0.433 (Table I)
Fitted to empirical histograms; the central claim N~2/B depends on these values.
omega+_B(t), omega-_B(t) = Tables II-V
Fitted scale parameters of the O(N) distribution; at large t they approach each other.
alpha_B(t) (histogram normalization) = not tabulated
Determined in the L1 fit, Eq. (8); used to match model and empirical normalizations.
N_B, C_B, M_B in Eq. (9) = N_B values in Table I; C_B and M_B less constrained
Exponential fits to extracted N_B(t) used to define asymptotic N_B(t -> infinity).
c0, c1 in N^{-1} = c0 + c1 B = c1=0.55+/-0.02 (c0=0) or c0=-0.14+/-0.02, c1=0.61+/-0.01
Linear fit to the four asymptotic N_B values; this fit produces the N~2/B statement.
Delta_cut = 0.1 = 10%
Hand-chosen truncation of extreme samples; paper says robustness was checked but not shown.
Nbins = 1600 = 1600
Hand-chosen bin count; affects resolution and fit results.
chi2_max/dof = 1.8 = 1.8
Selection threshold for acceptable fits when extracting N_B(t -> infinity); choice affects uncertainty envelope.

assumptions (6)

domain assumption The O(N) model correlation function distribution Eq. (2) from Ref. [18] applies to QCD correlation function samples.
The paper assumes the functional form derived for the O(N) model carries over to QCD without a QCD derivation; Ref. [18] is by two of the present authors.
domain assumption At large t, omega+ = omega- in QCD as in the O(N) model.
Used to justify the simplified large-time distribution and to interpret the asymptotic fits (Sec. II, Eq. (3); Sec. III C).
ad hoc to paper Truncating the lowest and highest 10% of samples and binning into 1600 bins preserves the information needed for unbiased fits.
Sec. III B step 1; the paper asserts robustness to variations but does not show the tails are irrelevant, and large-t fits are sensitive to binning.
domain assumption All Monte-Carlo samples (configurations and source locations) can be treated as independent draws from a common distribution.
Samples from the same gauge configuration are correlated; the histogram treats each sample as independent, which can underestimate uncertainties.
ad hoc to paper The fitted normalization alpha in Eq. (8) does not compromise the shape comparison.
Alpha absorbs overall scale; the text says it is not free but it is determined in the fit, so part of the agreement may come from this freedom.
ad hoc to paper Accepting fits with |delta_B,t(x)| < 0.3 in most bins is a valid goodness-of-fit measure.
No p-value or comparison to alternative distributions is given for this criterion.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Empirical investigation of nuclear correlation function distributions in lattice QCD." pith.science (2026). https://pith.science/paper/M5VLVYIC

@misc{pith2026250820378,
  author       = {Pith},
  title        = {Pith review of: Empirical investigation of nuclear correlation function distributions in lattice QCD},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M5VLVYIC}},
  note         = {Machine review of arXiv:2508.20378}
}

abstract

Two-point correlation functions of systems with baryon number $B \in \{1,2,3,4\}$ are investigated using lattice Quantum Chromodynamics (QCD). In particular, the empirical distributions of importance-sampling Monte-Carlo samples of these correlation functions are examined as a function of the spacetime separation between the two points and the baryon number. While the exact forms of these distributions are not known for QCD, recent work has determined asymptotic expressions for analogous correlation function distributions in simpler theories such as scalar field theory and the disordered phase of the $O(N)$ model. The theoretical O(N) model distributions are found to provide an accurate description of the empirical QCD distributions at zero momentum over a wide range of temporal separations for each baryon number when assessed with a range of different statistical tests. In particular, the behaviour of the baryon number $B$ QCD correlation function at large temporal separation is well-reproduced by the $O(N \sim 2/B)$ model distribution.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

29 extracted references · 18 canonical work pages

[1]

,2NcfgN src} indexing both the configura- tion and source location (appropriately translated) used for the correlation function

Forward and backward correlation functions are concatenated to produce a set of samples {Ci} with i ∈ {1, . . . ,2NcfgN src} indexing both the configura- tion and source location (appropriately translated) used for the correlation function. For each times- lice, the data are ordered and truncated by remov- ing the lowest and highest ∆ cut = 10% in order t...

work page
[2]

(2) with parameters θ = {N, ω+, ω−} (since the normali- sation of the histogram changes with the number of bins and data cut, the normalisation α is also determined in the fit)

Each empirical distribution EB,t(x) is fit by min- imising the L1-norm (total variation)2 LB,t(θ, α) = X i |EB,t(xi) − α MB,t(xi, θ)|, (8) where MB,t(x, θ) ≡ P (x; ω+, ω−, N) is the asym- metric O(N )-model distribution in Eq. (2) with parameters θ = {N, ω+, ω−} (since the normali- sation of the histogram changes with the number of bins and data cut, the ...

work page
[3]

This procedure is performed for each of the Nboot = 14 bootstrap resamplings of the original data and 1 Variations of the cuts used in data truncation and the number of bins used to produce the histograms have been investigated. For variations around the values of ∆ cut and Nbins chosen here, the analysis is robust provided thatNbins increases as ∆cut dec...

work page
[4]

(8) is com- plicated, the search is repeated Niter = 10 times, with the best-fit parameter values for each B, t and data bootstrap updated if the loss decreases

Since the loss landscape defined by Eq. (8) is com- plicated, the search is repeated Niter = 10 times, with the best-fit parameter values for each B, t and data bootstrap updated if the loss decreases. The resulting fits are shown for allB and for exemplary values of t in Fig. 2, with the full set of results shown in Figs. 9–16 in Appendix A. The extracte...

work page 2000
[5]

S. R. Beane, W. Detmold, K. Orginos, and M. J. Savage, Prog. Part. Nucl. Phys. 66, 1 (2011), arXiv:1004.2935 [hep-lat]

arXiv 2011
[6]

Parisi, Phys

G. Parisi, Phys. Rept. 103, 203 (1984). [2] G. P. Lepage, in Theoretical Advanced Study Institute in Elementary Particle Physics(1989). 9 -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π FIG. 8. Phase distributions for different time sepa...

work page 1984
[7]

S. R. Beane, W. Detmold, T. C. Luu, K. Orginos, A. Par- reno, M. J. Savage, A. Torok, and A. Walker-Loud, Phys. Rev. D 80, 074501 (2009), arXiv:0905.0466 [hep-lat]

work page Pith review arXiv 2009
[8]

S. R. Beane, W. Detmold, H.-W. Lin, T. C. Luu, K. Orginos, M. J. Savage, A. Torok, and A. Walker- Loud (NPLQCD), Phys. Rev. D 81, 054505 (2010), arXiv:0912.4243 [hep-lat]

work page Pith review arXiv 2010

Show all 29 references

[9]

M. L. Wagman, F. Winter, E. Chang, Z. Davoudi, W. Detmold, K. Orginos, M. J. Savage, and P. E. Shana- han, Phys. Rev. D 96, 114510 (2017), arXiv:1706.06550 [hep-lat]

2017 arXiv
[10]

S. R. Beane, E. Chang, W. Detmold, H. W. Lin, T. C. Luu, K. Orginos, A. Parreno, M. J. Savage, A. Torok, and A. Walker-Loud (NPLQCD), Phys. Rev. D 85, 054511 (2012), arXiv:1109.2889 [hep-lat]

2012 arXiv
[11]

S. R. Beane, E. Chang, S. D. Cohen, W. Detmold, H. W. Lin, T. C. Luu, K. Orginos, A. Parreno, M. J. Sav- age, and A. Walker-Loud (NPLQCD), Phys. Rev. D 87, 034506 (2013), arXiv:1206.5219 [hep-lat]

2013 arXiv
[12]

Orginos, A

K. Orginos, A. Parre˜ no, M. J. Savage, S. R. Beane, E. Chang, and W. Detmold, Phys. Rev. D 92, 114512 (2015), [Erratum: Phys.Rev.D 102, 039903 (2020)], arXiv:1508.07583 [hep-lat]

2015 arXiv
[13]

M. L. Wagman and M. J. Savage, Phys. Rev. D 96, 114508 (2017), arXiv:1611.07643 [hep-lat]. 10

2017 arXiv
[14]

M. G. Endres, D. B. Kaplan, J.-W. Lee, and A. N. Nicholson, PoS LA TTICE2011, 017 (2011), arXiv:1112.4023 [hep-lat]

2011 arXiv
[15]

DeGrand, Phys

T. DeGrand, Phys. Rev. D 86, 014512 (2012), arXiv:1204.4664 [hep-lat]

2012 arXiv
[16]

Grabowska, D

D. Grabowska, D. B. Kaplan, and A. N. Nicholson, Phys. Rev. D 87, 014504 (2013), arXiv:1208.5760 [hep-lat]

2013 arXiv
[17]

Yunus and W

C. Yunus and W. Detmold, Phys. Lett. B 840, 137890 (2023), arXiv:2210.15789 [hep-lat]

2023 arXiv
[18]

M. L. Wagman, Statistical Angles on the Lattice QCD Signal-to-Noise Problem, Ph.D. thesis, U. Washington, Seattle (main) (2017), arXiv:1711.00062 [hep-lat]

2017 arXiv
[19]

M. L. Wagman and M. J. Savage, (2017), arXiv:1704.07356 [hep-lat]

2017 arXiv
[20]

Davoudi, W

Z. Davoudi, W. Detmold, P. Shanahan, K. Orginos, A. Parre˜ no, M. J. Savage, and M. L. Wagman, Phys. Rept. 900, 1 (2021), arXiv:2008.11160 [hep-lat]

2021 arXiv
[21]

Duane, A

S. Duane, A. D. Kennedy, B. J. Pendleton, and D. Roweth, Phys. Lett. B 195, 216 (1987)

1987
[22]

Yunus and W

C. Yunus and W. Detmold, Phys. Rev. D 109, 034512 (2024), arXiv:2304.03820 [hep-lat]

2024 arXiv
[23]

with one level of stout [24] smearing. The lattice volume was L3 × T = 323 × 96 and the bare coupling and light (up and down) and strange quark mass parameters correspond to a lattice spacing of a ∼ 0.145 fm, a pion mass, mπ ∼ 450 MeV, and a kaon mass, mK ∼ 596 MeV. In what fo...
[24]

P. G. de Gennes, Phys. Lett. A 38, 339 (1972)

1972
[25]

Parre˜ no, P

A. Parre˜ no, P. E. Shanahan, M. L. Wagman, F. Winter, E. Chang, W. Detmold, and M. Illa (NPLQCD), Phys. Rev. D 103, 074511 (2021), arXiv:2102.03805 [hep-lat]

2021 arXiv
[26]

Sheikholeslami and R

B. Sheikholeslami and R. Wohlert, Nucl. Phys. B 259, 572 (1985)

1985
[27]

L¨ uscher, and P

M. L¨ uscher, and P. Weisz, Commun. Math. Phys.98, 433 (1985), [Erratum: Commun.Math.Phys. 98, 433 (1985)]

1985
[28]

Morningstar and M

C. Morningstar and M. J. Peardon, Phys. Rev. D 69, 054501 (2004), arXiv:hep-lat/0311018

2004 arXiv
[29]

Detmold and K

W. Detmold and K. Orginos, Phys. Rev. D 87, 114512 (2013), arXiv:1207.1452 [hep-lat]. 11 Appendix A: Fitted distributions t N ω + ω− 1 88 .7+1.3 −1.0 44.6+0.8 −0.4 9.2+0.8 −1.0 × 104 2 57 .1+1.1 −32. 28.9+0.5 −16. 8.1+1.7 −0.9 × 104 3 36 .8+0.9 −11. 18.6+0.5 −6. 8.3+1.7 −8. × ...

2013 arXiv

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.

[1] [1]

,2NcfgN src} indexing both the configura- tion and source location (appropriately translated) used for the correlation function

Forward and backward correlation functions are concatenated to produce a set of samples {Ci} with i ∈ {1, . . . ,2NcfgN src} indexing both the configura- tion and source location (appropriately translated) used for the correlation function. For each times- lice, the data are ordered and truncated by remov- ing the lowest and highest ∆ cut = 10% in order t...

work page

[2] [2]

(2) with parameters θ = {N, ω+, ω−} (since the normali- sation of the histogram changes with the number of bins and data cut, the normalisation α is also determined in the fit)

Each empirical distribution EB,t(x) is fit by min- imising the L1-norm (total variation)2 LB,t(θ, α) = X i |EB,t(xi) − α MB,t(xi, θ)|, (8) where MB,t(x, θ) ≡ P (x; ω+, ω−, N) is the asym- metric O(N )-model distribution in Eq. (2) with parameters θ = {N, ω+, ω−} (since the normali- sation of the histogram changes with the number of bins and data cut, the ...

work page

[3] [3]

This procedure is performed for each of the Nboot = 14 bootstrap resamplings of the original data and 1 Variations of the cuts used in data truncation and the number of bins used to produce the histograms have been investigated. For variations around the values of ∆ cut and Nbins chosen here, the analysis is robust provided thatNbins increases as ∆cut dec...

work page

[4] [4]

(8) is com- plicated, the search is repeated Niter = 10 times, with the best-fit parameter values for each B, t and data bootstrap updated if the loss decreases

Since the loss landscape defined by Eq. (8) is com- plicated, the search is repeated Niter = 10 times, with the best-fit parameter values for each B, t and data bootstrap updated if the loss decreases. The resulting fits are shown for allB and for exemplary values of t in Fig. 2, with the full set of results shown in Figs. 9–16 in Appendix A. The extracte...

work page 2000

[5] [5]

S. R. Beane, W. Detmold, K. Orginos, and M. J. Savage, Prog. Part. Nucl. Phys. 66, 1 (2011), arXiv:1004.2935 [hep-lat]

arXiv 2011

[6] [6]

Parisi, Phys

G. Parisi, Phys. Rept. 103, 203 (1984). [2] G. P. Lepage, in Theoretical Advanced Study Institute in Elementary Particle Physics(1989). 9 -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π -π 0 π FIG. 8. Phase distributions for different time sepa...

work page 1984

[7] [7]

S. R. Beane, W. Detmold, T. C. Luu, K. Orginos, A. Par- reno, M. J. Savage, A. Torok, and A. Walker-Loud, Phys. Rev. D 80, 074501 (2009), arXiv:0905.0466 [hep-lat]

work page Pith review arXiv 2009

[8] [8]

S. R. Beane, W. Detmold, H.-W. Lin, T. C. Luu, K. Orginos, M. J. Savage, A. Torok, and A. Walker- Loud (NPLQCD), Phys. Rev. D 81, 054505 (2010), arXiv:0912.4243 [hep-lat]

work page Pith review arXiv 2010