Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Normalizing Flow-Assisted Nested Sampling on Type-II Seesaw Model

T0 review · 3 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Nested sampling of a seven-dimensional BSM parameter space can be accelerated by an iteratively retrained RealNVP generator plus a classifier without biasing the resulting posterior, and the paper demonstrates this on the Type-II seesaw…

desk verdict A useful ML-accelerated nested sampling engineering contribution with a new Type-II seesaw posterior, but the headline claim of an unbiased hybrid proposal is not yet backed by a converged baseline or an independent sampler check. read the letter →

arxiv 2501.16432 v2 pith:RVGJBEUL submitted 2025-01-27 hep-ph

Rajneil Baruah , Subhadeep Mondal , Sunando Kumar Patra , Satyajit Roy This is my paper · ORCID
classification hep-ph
keywords nestedsamplingnormalizingflowsRealNVPself-normalizingneuralnetworkType-IIseesawBayesianposteriorHiggsobservablesobliqueparameters
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 is trying to establish that nested sampling, the standard Bayesian evidence and posterior sampler, can be made much faster on particle-physics parameter spaces by inserting machine-learned proposals without corrupting the result. The proposed pipeline inserts two networks into the nested-sampling loop: an ensemble classifier that avoids running the expensive spectrum generator on invalid points, and a RealNVP normalizing-flow generator that is retrained as the run progresses to propose points inside the shrinking live region. The specific schedule named RNVP-SF starts with the faithful hypercube-based selection and switches to fast iso-likelihood selection once the sample rate falls; the authors report that this reaches the termination tolerance with a posterior that is not perceptibly different from the much slower faithful run. Applied to the Type-II seesaw model, the method yields what the authors describe as the first posterior estimates for all seven new parameters under 125 GeV Higgs, rho-parameter, and oblique data, and maps the favoured region for the triplet VEV and charged Higgs masses. A reader should care because the cost bottleneck is generic: any beyond-standard-model scan that must generate a spectrum at every candidate point faces the same slowdown.

What carries the argument

The load-bearing object is the RNVP-SF proposal pipeline. RealNVP is a normalizing flow built from affine coupling layers, an invertible map whose triangular Jacobian makes density evaluation cheap, and here it is trained iteratively on pooled 13-component truth vectors, namely the seven input parameters plus the SM-like Higgs mass, rho shift, three oblique parameters, and the Higgs chi-squared. An ensemble of self-normalizing networks classifies parameter points as spectrum-valid or not before the expensive generator is called. The third piece is the dataset-selection schedule: sample training points from the hypercube bounding the live points, RNVP-S, until the nested-sampling sample rate between two trainings drops below 50, then switch to selecting points near the current iso-likelihood contour, RNVP-F. This schedule is what the paper credits for preserving unbiasedness while recovering speed.

What would settle it

Run plain nested sampling on the same seven-parameter space to tolerance below 0.001 and compare the marginal posterior in lambda4; if it reveals a positive-lambda4 high-likelihood mode that RNVP-SF missed, the claim of a correct posterior fails. A cheaper diagnostic is a two-sample test between RNVP-F proposals and the exact constrained prior restricted to the live hypercube after the switch, since any systematic mismatch breaks the iid-proposal assumption.

Watch

Extended reading notes

Core claim

The paper's central claim is methodological, stated in Section 5 as: the variant called RNVP-SF "optimizes the runtime while generating the correct posterior." Correctness is established by the posterior's high-likelihood region coinciding with the concentration of live points and by comparing against a truncated slow run at tolerance ~4.09; the authors also present the posterior as the first Bayesian mapping of all seven Type-II seesaw parameters. The same section presents the physics outcome of applying the method: the high-vT Type-II seesaw parameter space, constrained by the 125 GeV Higgs data, the rho-parameter, and the oblique parameters, with the favoured scalar masses, mixing angle, and neutrino Yukawa couplings reported in the results.

Load-bearing premise

The correctness claim rests on the assumption that, after switching to the fast iso-likelihood dataset-selection phase, the generator's proposals are still independent draws from the constrained prior inside the shrinking hypercube, so the nested-sampling importance weights are unbiased; the paper checks this only by comparing against a truncated slow run and by asserting no perceptible change.

Editorial extensions

If this is right

  • The RNVP-SF scheme reduces the wall time of the Type-II seesaw scan from more than 700 hours at tolerance ~4.09 to convergence at tolerance below 0.001, without, the authors argue, altering the posterior perceptibly.
  • The resulting posterior maps the large-vT Type-II seesaw parameter space for the first time: the triplet VEV peaks near 1.2-1.4 GeV, and all seven input parameters now have central values, dispersions, and correlations.
  • The allowed exotic scalar masses are mostly 400-800 GeV, with the charged-Higgs mass splitting at most about 40 GeV and the CP-even mixing angle sin alpha around 10^-3, so the favoured region is not yet excluded by pair-produced doubly charged Higgs searches.
  • Neutrino Yukawa matrix elements in the favoured region are all near 10^-10, strongly suppressing lepton-flavour- and lepton-number-violating decays.
  • Because nested sampling produces Bayesian evidence as a by-product, the method also gives the tools to compare Type-II seesaw against other models as data accumulate.

Reading between the lines

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

  • The switch heuristic, namely switching to iso-likelihood dataset selection once the sample rate between two trainings falls below 50, is tuned on this run; a model with widely separated likelihood modes might need a different schedule, and a blended or multi-flow proposal could be more robust.
  • The authors' own admission that a high-likelihood region at positive lambda4 may be missing is exactly the signature a biased proposal would produce, so a controlled toy model with known evidence is the natural next test of the method.
  • The same architecture is portable to other beyond-standard-model scans whose bottleneck is an expensive spectrum generator, but unbiasedness is a property of the whole proposal-plus-schedule system and would have to be revalidated for each model rather than assumed.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper proposes an ML-assisted nested sampling (NS) framework for BSM parameter spaces, combining an ensemble self-normalizing neural-network classifier with an iteratively retrained RealNVP normalizing flow used as a proposal generator inside NS. Three dataset-selection strategies are compared: RNVP-S (representative prior samples from the bounding hypercube), RNVP-F (training data drawn from iso-likelihood contours just below the current live threshold), and the hybrid RNVP-SF, which switches from RNVP-S to RNVP-F once the sample rate falls below 50. The method is applied to the Type-II seesaw model with seven free parameters, using SPheno and HiggsTools to evaluate HiggsSignals, rho-parameter, and oblique-parameter likelihoods. The authors present posterior constraints on the scalar sector, predictions for charged and neutral scalar masses, doubly charged Higgs collider limits, and estimates of the neutrino Yukawa matrix. The central methodological claim is that RNVP-SF "optimizes the runtime while generating the correct posterior" (Sec. 5).

Significance. If the central correctness claim were established, the paper would be a useful practical contribution: it demonstrates a nontrivial integration of a classifier and a normalizing flow into nested sampling, with a physically interesting application and reproducible artifacts (data, figures, and trained models are provided on GitHub). The physics results are plausible and the pipeline is standard in its use of existing spectrum generators and HiggsTools. However, the methodological validation is currently insufficient to support the headline claim, and the authors themselves identify a possibly undetected high-likelihood region with positive lambda4. Because the manuscript is transparent about these gaps and the approach is promising, the correct recommendation is major revision rather than rejection.

major comments (3)
  1. [Sec. 3.3.3, Fig. 4] The claim that RNVP-SF generates the correct posterior is not supported by the provided validation. In RNVP-F mode, the training set is selected from iso-likelihood contours with L slightly below the lowest live-point likelihood (Sec. 3.3.3), and the RealNVP is used only as a generator with no reweighting by its learned density (Sec. 3.2.3). After the switch, accepted candidates with L > L* are therefore distributed according to the truncated generator density, not uniformly over the constrained prior within the current hypercube, so the NS importance weights are biased. The comparison baseline, RNVP-S, was stopped at tolerance ~4.09 after more than 700 hours, far from the target 0.001, and the agreement is judged visually as not "perceptibly" different. A non-converged baseline cannot certify unbiasedness. The authors should provide either a fully converged independent NS run, a density-corrected acceptance scheme for the generator proposals, or diagnostics demonstrating that the live points after the switch are effectively iid draws from the constrained prior.
  2. [Sec. 4.1, Fig. 5 (bottom-right)] The authors state that "there exists another high likelihood region with positive λ4, which was not detected by our algorithm." This admission directly contradicts the claim of a correct posterior: if a high-likelihood mode of the constrained prior is missing, the weighted NS sample under-represents that region and the evidence is underestimated. This is exactly the failure mode expected when the proposal is trained on iso-likelihood shells around the currently known live points. The manuscript should resolve this ambiguity, for example by performing a targeted scan or separate NS run initialized in the positive-λ4 region, or by quantifying that the posterior mass of that region is negligible within the stated credible intervals.
  3. [Sec. 3.3.1] The initial live points for the NS run are taken from a pre-collected pool of vectors with total χ² ≤ 1335, not from the declared uniform prior over the seven-dimensional parameter box. Nested-sampling theory requires the initial live points to be drawn from the prior; using a low-χ²-truncated pool changes the effective prior and therefore biases the evidence and early posterior weights. The manuscript should quantify the fraction of prior volume excluded by the χ² ≤ 1335 cutoff and demonstrate that this truncation does not affect the final posterior, for example by comparing with an initial pool drawn without that cutoff or by reweighting the initial live points.
minor comments (5)
  1. [Abstract] The PACS line "PACS-key discribring text of that key" is a placeholder and should be removed or filled with actual PACS codes.
  2. [Sec. 1] The phrase "large chuck of the parameter space" should read "large chunk of the parameter space."
  3. [Fig. 4] The legend entries such as "RN V P− SF" have inconsistent spacing and should be typeset as RNVP-SF; the vertical line marking the switching point should be defined in the caption.
  4. [Sec. 4.2.2, Eq. (13)] The Yukawa matrix is quoted with means and standard deviations, but the manuscript does not specify the PMNS parametrization, the basis in which Y is given, or the inversion formula used beyond Eq. (12); as written, the result is not reproducible.
  5. [Sec. 2, Eq. (10)] Equation (10) is typeset ambiguously; the numerator and denominator in the prefactor should be clarified, for example as m_{h±}^2 = ((2√2 μ1 − λ4 v_T) / (4 v_T)) (v_d^2 + 2 v_T^2).

Circularity Check

0 steps flagged · score 0.0 of 10

No definitional circularity: the final posterior is built from recomputed likelihoods, and the ML networks are proposal/filter aids, not substitutes for the likelihood.

full rationale

The paper's central claim, that RNVP-SF 'optimizes the runtime while generating the correct posterior' (Sec. 5), is an empirical performance claim rather than a derivation that reduces to its inputs. In the algorithm, every candidate that survives the SNN classifier is passed through SPheno and HiggsSignals to obtain the actual 13-component truth vector, and only those passing the nested-sampling criterion L > L* enter the nested sample (Sec. 3.3.2: 'We then use these truth 13-vectors to check the NS criteria (L > L*)'). The RealNVP is used only as a generator, and the paper explicitly says the learned PDF is not used for the final result (Sec. 3.2.3: 'we will not use the learned PDFs directly as our result'). Hence there is no fitted-input-called-prediction reduction: the posterior weights come from real likelihood evaluations, not from the flow's density. The iterative retraining on pooled NS output is a feedback loop, and the paper's own admission of a possible undetected positive-λ4 region (Sec. 4.1) is a genuine correctness concern, as is the validation of RNVP-SF against an RNVP-S run truncated at tolerance ~4.09 rather than the target 0.001 (Sec. 3.3.3). However, those are statistical-validity limitations, not circularity: the claim that the proposal distribution remains an unbiased constrained-prior sampler after switching to RNVP-F is asserted rather than proved, but the final posterior is not defined in terms of that proposal distribution, and no equation in the paper equates the result with the training data or with a fitted parameter. The self-citations [24, 124] describe a similar SNN setup and a broader account of ML sampling, but the architecture, training, and dataset-selection variants are described and tested in the present work, so those citations are not load-bearing and no uniqueness theorem is imported from the authors. Overall, no step in the derivation chain is circular by construction.

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

The seven physics parameters are not counted as free parameters because they are sampled, not fitted. The central method depends on several hand-chosen controls: the chi2<=1335 initial-pool cutoff, the retraining interval, the SF switching threshold, classifier and RNVP hyperparameters, and NS termination tolerance; nLive is not reported. The key ad hoc axiom is that the ML proposal distribution preserves NS unbiasedness after the SF switch, which is exactly the unvalidated premise discussed in weakest_assumption.

free parameters (6)
  • Initial dataset chi2 cutoff = 1335 (total chi2)
    Hand-chosen threshold for the starting pool; biases initial NS live points toward low-chi2 regions (Sec. 3.3.1).
  • RNVP retraining interval = every 6000 dumped points
    Hand-chosen schedule; affects generator accuracy and convergence time (Sec. 3.3.2).
  • Sample-rate switching threshold = 50 samples between trainings
    Ad hoc trigger for switching from RNVP-S to RNVP-F (Sec. 3.3.3, Fig. 4).
  • Classifier hyperparameters (nm, nw, p_alpha) = (4, 200, 0.007)
    Selected by grid search on accuracy; not derived from first principles (Sec. 3.2.2).
  • RNVP architecture (depth, coupling layers, max rounds) = (2, 2, 750)
    Selected by model selection; affects proposal quality (Sec. 3.3.2, Fig. 3).
  • NS termination tolerance and nLive = tolerance 0.001; nLive not reported
    Termination controls evidence accuracy; nLive is missing from the manuscript (Sec. 3.3.1).
assumptions (6)
  • standard math The Type-II seesaw scalar potential, tadpole equations, and mass matrices in Sec. 2 are correct and complete.
    Assumed from prior literature; the paper does not re-derive them.
  • domain assumption Vacuum stability and perturbative unitarity constraints from ref. [83] fully characterize allowed quartic couplings.
    Invoked to define the valid parameter region before spectrum generation (Sec. 3.1).
  • domain assumption SPheno, HiggsTools (HiggsBounds/HiggsSignals), and the GFitter/PDG likelihoods correctly compute the observables and uncertainties used in the 163-term chi-square.
    The entire posterior depends on these external calculations (Sec. 3.1).
  • ad hoc to paper Nested sampling with a hypercube bounding region and the ML proposal distribution yields unbiased posterior weights for RNVP-SF.
    This is the key correctness assumption; it is not demonstrated by an independent converged comparison (Sec. 3.3).
  • ad hoc to paper The starting pool of vectors with total chi2 <= 1335 is a legitimate representation of the prior for initialization.
    The initial NS live points are drawn from this pre-filtered pool, biasing the starting point toward already-favored regions (Sec. 3.3.1).
  • domain assumption The hand-chosen prior ranges for vT, lambda1, mu1, lambda2, lambda3, lambda4, lambda define the target parameter space.
    The posterior is only meaningful within these intervals (Sec. 3.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Normalizing Flow-Assisted Nested Sampling on Type-II Seesaw Model." pith.science (2026). https://pith.science/paper/RVGJBEUL

@misc{pith2026250116432,
  author       = {Pith},
  title        = {Pith review of: Normalizing Flow-Assisted Nested Sampling on Type-II Seesaw Model},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RVGJBEUL}},
  note         = {Machine review of arXiv:2501.16432}
}
abstract

We propose a novel technique for sampling particle physics model parameter space. The main sampling method applied is Nested Sampling (NS), which is boosted by the application of multiple Machine Learning (ML) networks, e.g., Self-Normalizing Network (SNN) and Normalizing Flow (specifically RealNVP). We apply this on Type-II Seesaw model to test the efficacy of the algorithm. We present the results of our detailed Bayesian exploration of the model parameter space subjected to theoretical constraints and experimental data corresponding to the 125 GeV Higgs boson, $\rho$-parameter, and the oblique parameters. All associated data, figures, and trained ML models can be found here: https://github.com/sunandopatra/MLNS-T2SS

Figures

Figures reproduced from arXiv: 2501.16432 by the authors.

Figure 1
Figure 1. Detailed flow chart of our ML-assisted nested sampling algorithm to sample the parameter space. in a specified region until it finds one point with likeli￾hood (L) greater than the lowest likelihood among the live points (L ∗ ), replaces that lowest-likelihood point with the new one, saves the replaced point as a weighted sample, shrinks the target region in a specified way and then re￾peats. This way, evaluating a … view at source ↗
Figure 2
Figure 2. Schematic diagram of the ensemble SNN used as a classifier in this work (left) and the confusion matrix (right). 3.2.2 Classification After collecting an adequate number of points from the target parameter space that pass the theoretical constraints deriving from vacuum stability and perturbative unitarity [83], we label them either ‘1’ or ‘0’, depending on whether the corresponding spectrum file is generated or not… view at source ↗
Figure 3
Figure 3. Representative learning curve for the training and model selection of the RNVP network. The vertical axis is the Mean Cross-Entropy (−3.327 ± 0.008 for the final one for this plot) and the horizontal axis is the log-scaled no. of training examples. After the initial inspection, the starting dataset is pre￾pared with > 63000 vectors of 13 components, for which the total χ 2 is not very large (≤ 1335). This dataset is… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Comparison of completion times for different algorithms used in this analysis (left). Time-series plots of (a) NS samples collected at every iteration (blue) and (b) total sample size (red) for RNV P − SF (right). The vertical line depicts the point at which the algori…
Figure 5
Figure 5. Figure 5: Two-dimensional marginal distribution of the posteriors. The contours of the same colour are from the same posterior. The solid and dot-dashed contours correspond to 68% respectively, and the similar 95% credible interval contours are dashed and dotted. Red and blue co…
Figure 6
Figure 6. Figure 6: Corner plot depicting all possible marginal distributions and highest likelihood regions of all quartic couplings [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: All points allowed by all theoretical and experimental constraints up to 95% CL: (a) in the doubly and singly charged Higgs mass plane (Top-left), (b) in the singly charged and neutral heavy Higgs mass plane (top-right) (c) The mixing angle between the CP-even Higgs st…
Figure 8
Figure 8. Figure 8: (Left) Sensitivity of the ρ-parameter and the SM-like Higgs mass to the high likelihood region of the parameter space allowed by all constraints. (Right) Comparing the allowed region in the oblique parameters S-T plane with that obtained in the high likelihood region f…
Figure 9
Figure 9. Figure 9: Corner plot depicting all possible marginal distributions and highest likelihood regions of all neutrino Yukawa [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Waterfall Plot demonstrating the additive nature of the SHAP at a high-likelihood point (above) and the distribu￾tion of the Shapley values over the whole dataset (below). training and test sets. After successfully training a Gra￾dient Boosted Trees algorithm (we coul…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Amortized Inference of Multi-Modal Posteriors using Likelihood-Weighted Normalizing Flows

    cs.LG 2025-12 reject novelty 3.0 of 10

    Training a normalizing flow on prior samples weighted by likelihood can approximate a posterior, but matching the base distribution's number of modes to the target is needed to avoid spurious bridges.

Reference graph

Works this paper leans on

137 extracted references · 20 canonical work pages · cited by 1 Pith paper

  1. [1]

    W. K. Hastings, Biometrika 57, 97 (1970), https://academic.oup.com/biomet/article- pdf/57/1/97/23940249/57-1-97.pdf

    1970

  2. [2]

    J. S. Speagle, (2019), 1909.12313

    arXiv 2019

  3. [3]

    Brooks, A

    S. Brooks, A. Gelman, G. Jones, and X. L. Meng, Hand- book of Markov Chain Monte Carlo (1st ed.)(Chapman and Hall/CRC, Boca Raton, FL, USA, 2011)

    2011

  4. [4]

    Tripuraneni, M

    N. Tripuraneni, M. Rowland, Z. Ghahramani, and R. Turner, Magnetic hamiltonian monte carlo, 2017, 1607.02738

    arXiv 2017

  5. [5]

    Albert et al., (2024), 2409.18464

    J. Albert et al., (2024), 2409.18464

    arXiv 2024

  6. [6]

    Athron et al., Eur

    GAMBIT, P. Athron et al., Eur. Phys. J. C 77, 879 (2017), 1705.07917

    arXiv 2017

  7. [7]

    Kvellestad, PoS EPS-HEP2017, 299 (2017), 1710.02503

    GAMBIT, A. Kvellestad, PoS EPS-HEP2017, 299 (2017), 1710.02503

    arXiv 2017

  8. [8]

    Foreman-Mackey, D

    D. Foreman-Mackey, D. W. Hogg, D. Lang, and J. Good- man, Publications of the Astronomical Society of the Pacific 125, 306 (2013)

    2013

Show all 137 references
  1. [9]

    De Blas et al., Eur

    J. De Blas et al., Eur. Phys. J. C 80, 456 (2020), 1910.14012

    2020 arXiv

  2. [10]

    G. O. R. Mary Kathryn. Cowles and J. S. Rosenthal, Journal of Statistical Computation and Simulation 64, 87 (1999), https://doi.org/10.1080/00949659908811968

    1999 doi

  3. [11]

    Caron, T

    S. Caron, T. Heskes, S. Otten, and B. Stienen, Eur. Phys. J. C 79, 944 (2019), 1905.08628

    2019 arXiv

  4. [12]

    Hollingsworth, M

    J. Hollingsworth, M. Ratz, P. Tanedo, and D. Whiteson, Eur. Phys. J. C 81, 1138 (2021), 2103.06957

    2021 arXiv

  5. [13]

    Gawlikowski et al., CoRR abs/2107.03342 (2021), 2107.03342

    J. Gawlikowski et al., CoRR abs/2107.03342 (2021), 2107.03342

    2021 arXiv

  6. [14]

    Hammad, M

    A. Hammad, M. Park, R. Ramos, and P. Saha, Comput. Phys. Commun. 293, 108902 (2023), 2207.09959

    2023 arXiv

  7. [15]

    M. D. Goodsell and A. Joury, Eur. Phys. J. C 83, 268 (2023), 2204.13950

    2023 arXiv

  8. [16]

    M. A. Diaz, S. Dasmahapatra, and S. Moretti, (2024), 2412.17675

    2024 arXiv

  9. [17]

    Binjonaid, (2024), 2409.05453

    M. Binjonaid, (2024), 2409.05453

    2024 arXiv

  10. [18]

    Hammad and R

    A. Hammad and R. Ramos, (2024), 2412.19675

    2024 arXiv

  11. [19]

    Hermans, V

    J. Hermans, V. Begy, and G. Louppe, (2019), 1903.04057

    2019 arXiv

  12. [20]

    Brehmer and K

    J. Brehmer and K. Cranmer, (2020), 2010.06439

    2020 arXiv

  13. [21]

    Morrison, S

    L. Morrison, S. Profumo, N. Smyth, and J. Tamanas, Phys. Rev. D 106, 115016 (2022), 2203.13403

    2022 arXiv

  14. [22]

    N. T. Hunt-Smith et al., Comput. Phys. Commun. 296, 109059 (2024), 2309.01454

    2024 arXiv

  15. [23]

    M. S. Albergo et al., (2021), 2101.08176

    2021 arXiv

  16. [24]

    Baruah, S

    R. Baruah, S. Mondal, S. K. Patra, and S. Roy, Eur. Phys. J. ST 233, 2597 (2024), 2404.02698

    2024 arXiv

  17. [25]

    Skilling, AIP Conference Proceedings 735, 395 (2004), https://pubs.aip.org/aip/acp/article- pdf/735/1/395/11702789/395 1 online.pdf

    J. Skilling, AIP Conference Proceedings 735, 395 (2004), https://pubs.aip.org/aip/acp/article- pdf/735/1/395/11702789/395 1 online.pdf

    2004

  18. [26]

    Skilling, Bayesian Analysis 1, 833 (2006)

    J. Skilling, Bayesian Analysis 1, 833 (2006)

    2006

  19. [27]

    Athron et al., Eur

    GAMBIT, P. Athron et al., Eur. Phys. J. C 77, 784 (2017), 1705.07908, [Addendum: Eur.Phys.J.C 78, 98 (2018)]

    2017 arXiv

  20. [28]

    Bloor et al., Eur

    S. Bloor et al., Eur. Phys. J. C 81, 1103 (2021), 2107.00030

    2021 arXiv

  21. [29]

    Feroz, M

    F. Feroz, M. P. Hobson, and M. Bridges, Mon. Not. Roy. Astron. Soc. 398, 1601 (2009), 0809.3437

    2009 arXiv

  22. [30]

    Feroz, M

    F. Feroz, M. P. Hobson, and M. Bridges, Monthly Notices of the Royal Astronomical Society 398, 1601 (2009), https://academic.oup.com/mnras/article- pdf/398/4/1601/3039078/mnras0398-1601.pdf

    2009

  23. [31]

    Konetschny and W

    W. Konetschny and W. Kummer, Phys. Lett. B 70, 433 (1977)

    1977

  24. [32]

    Magg and C

    M. Magg and C. Wetterich, Phys. Lett. B 94, 61 (1980)

    1980

  25. [33]

    Schechter and J

    J. Schechter and J. W. F. Valle, Phys. Rev. D 22, 2227 (1980)

    1980

  26. [34]

    Lazarides, Q

    G. Lazarides, Q. Shafi, and C. Wetterich, Nucl. Phys. B 181, 287 (1981)

    1981

  27. [35]

    R. N. Mohapatra and G. Senjanovic, Phys. Rev. D 23, 165 (1981)

    1981

  28. [36]

    T. P. Cheng and L.-F. Li, Phys. Rev. D 22, 2860 (1980)

    1980

  29. [37]

    Bak et al., Phys

    RENO, G. Bak et al., Phys. Rev. Lett. 121, 201801 (2018), 1806.00248

    2018 arXiv

  30. [38]

    Adey et al., Phys

    Daya Bay, D. Adey et al., Phys. Rev. Lett. 121, 241805 (2018), 1809.02261

    2018 arXiv

  31. [39]

    Jiang et al., PTEP 2019, 053F01 (2019), 1901.03230

    Super-Kamiokande, M. Jiang et al., PTEP 2019, 053F01 (2019), 1901.03230

    2019 arXiv

  32. [40]

    Abe et al., Phys

    T2K, K. Abe et al., Phys. Rev. Lett. 121, 171802 (2018), 1807.07891

    2018 arXiv

  33. [41]

    NOvA, M. A. Acero et al., Phys. Rev. Lett. 123, 151803 (2019), 1906.04907

    2019 arXiv

  34. [42]

    P. F. de Salas et al., JHEP 02, 071 (2021), 2006.11237

    2021 arXiv

  35. [43]

    Huitu, J

    K. Huitu, J. Maalampi, A. Pietila, and M. Raidal, Nucl. Phys. B487, 27 (1997), hep-ph/9606311

    1997 arXiv

  36. [44]

    Chakrabarti, D

    S. Chakrabarti, D. Choudhury, R. M. Godbole, and B. Mukhopadhyaya, Phys. Lett. B 434, 347 (1998), hep- ph/9804297

    1998

  37. [45]

    Muhlleitner and M

    M. Muhlleitner and M. Spira, Phys. Rev. D68, 117701 (2003), hep-ph/0305288

    2003 arXiv

  38. [46]

    A. G. Akeroyd and M. Aoki, Phys. Rev. D72, 035011 (2005), hep-ph/0506176

    2005 arXiv

  39. [47]

    Fileviez Perez, T

    P. Fileviez Perez, T. Han, G.-y. Huang, T. Li, and K. Wang, Phys. Rev. D78, 015018 (2008), 0805.3536

    2008 arXiv

  40. [48]

    del Aguila and J

    F. del Aguila and J. A. Aguilar-Saavedra, Nucl. Phys. B813, 22 (2009), 0808.2468

    2009 arXiv

  41. [49]

    A. G. Akeroyd and C.-W. Chiang, Phys. Rev. D80, 113010 (2009), 0909.4419

    2009 arXiv

  42. [50]

    Melfo, M

    A. Melfo, M. Nemevsek, F. Nesti, G. Senjanovic, and Y. Zhang, Phys. Rev. D 85, 055018 (2012), 1108.4416

    2012 arXiv

  43. [51]

    Arhrib, R

    A. Arhrib, R. Benbrik, M. Chabab, G. Moultaka, and L. Rahili, JHEP 1204, 136 (2012), 1112.5453

    2012 arXiv

  44. [52]

    A. G. Akeroyd, S. Moretti, and H. Sugiyama, Phys. Rev. D 85, 055026 (2012), 1201.5047

    2012 arXiv

  45. [53]

    E. J. Chun and P. Sharma, JHEP 08, 162 (2012), 1206.6278

    2012 arXiv

  46. [54]

    E. J. Chun, H. M. Lee, and P. Sharma, JHEP 11, 106 (2012), 1209.1303

    2012 arXiv

  47. [55]

    Englert, E

    C. Englert, E. Re, and M. Spannowsky, Phys. Rev. D88, 035024 (2013), 1306.6228. 16 Rajneil Baruah et al.: Normalizing Flow-Assisted Nested Sampling on Type-II Seesaw Model

    2013 arXiv

  48. [56]

    Kanemura, K

    S. Kanemura, K. Yagyu, and H. Yokoya, Phys.Lett. B726, 316 (2013), 1305.2383

    2013 arXiv

  49. [57]

    E. J. Chun and P. Sharma, Phys. Lett. B 728, 256 (2014), 1309.6888

    2014 arXiv

  50. [58]

    E. J. Chun, K. Y. Lee, and S. C. Park, Phys. Lett. B566, 142 (2003), hep-ph/0304069

    2003 arXiv

  51. [59]

    P. Dey, A. Kundu, and B. Mukhopadhyaya, J. Phys. G36, 025002 (2009), 0802.2510

    2009 arXiv

  52. [60]

    Arhrib et al., Phys

    A. Arhrib et al., Phys. Rev. D 84, 095005 (2011), 1105.1925

    2011 arXiv

  53. [61]

    Bhupal Dev, D

    P. Bhupal Dev, D. K. Ghosh, N. Okada, and I. Saha, JHEP 1303, 150 (2013), 1301.3453

    2013 arXiv

  54. [62]

    Arhrib, R

    A. Arhrib, R. Benbrik, G. Moultaka, and L. Rahili, (2014), 1411.5645

    2014 arXiv

  55. [63]

    Das and A

    D. Das and A. Santamaria, Phys. Rev. D 94, 015015 (2016), 1604.08099

    2016 arXiv

  56. [64]

    Banerjee, J

    Anisha, U. Banerjee, J. Chakrabortty, C. Englert, and M. Spannowsky, Phys. Rev. D 103, 096009 (2021), 2103.01810

    2021 arXiv

  57. [65]

    Y. Du, A. Dunbrack, M. J. Ramsey-Musolf, and J.-H. Yu, JHEP 01, 101 (2019), 1810.09450

    2019 arXiv

  58. [66]

    Li, JHEP 09, 079 (2018), 1802.00945

    T. Li, JHEP 09, 079 (2018), 1802.00945

    2018 arXiv

  59. [67]

    Kanemura, M

    S. Kanemura, M. Kikuchi, K. Yagyu, and H. Yokoya, Phys. Rev. D 90, 115018 (2014), 1407.6547

    2014 arXiv

  60. [68]

    Z.-L. Han, R. Ding, and Y. Liao, Phys. Rev. D 91, 093006 (2015), 1502.05242

    2015 arXiv

  61. [69]

    Z.-L. Han, R. Ding, and Y. Liao, Phys. Rev. D 92, 033014 (2015), 1506.08996

    2015 arXiv

  62. [70]

    Mitra, S

    M. Mitra, S. Niyogi, and M. Spannowsky, Phys. Rev. D95, 035042 (2017), 1611.09594

    2017 arXiv

  63. [71]

    K. S. Babu and S. Jana, Phys. Rev. D 95, 055020 (2017), 1612.09224

    2017 arXiv

  64. [72]

    P. S. Bhupal Dev and Y. Zhang, JHEP 10, 199 (2018), 1808.00943

    2018 arXiv

  65. [73]

    Antusch, O

    S. Antusch, O. Fischer, A. Hammad, and C. Scherb, JHEP 02, 157 (2019), 1811.03476

    2019 arXiv

  66. [74]

    M. M. Ferreira, T. B. de Melo, S. Kovalenko, P. R. D. Pinheiro, and F. S. Queiroz, Eur. Phys. J. C 79, 955 (2019), 1903.07634

    2019 arXiv

  67. [75]

    Banerjee, C

    U. Banerjee, C. Englert, and W. Naskar, Phys. Rev. D 110, 055010 (2024), 2403.17455

    2024 arXiv

  68. [76]

    Ashanujjaman and S

    S. Ashanujjaman and S. P. Maharathy, Phys. Rev. D 107, 115026 (2023), 2305.06889

    2023

  69. [77]

    Giarnetti, J

    A. Giarnetti, J. Herrero-Garcia, S. Marciano, D. Meloni, and D. Vatsyayan, JHEP 05, 055 (2024), 2312.13356

    2024 arXiv

  70. [78]

    Chiang, S

    C.-W. Chiang, S. Jana, and D. Sengupta, Phys. Rev. D 105, 055014 (2022), 2106.03888

    2022 arXiv

  71. [79]

    Ashanujjaman, K

    S. Ashanujjaman, K. Ghosh, and R. Sahu, Phys. Rev. D 107, 015018 (2023), 2211.00632

    2023

  72. [80]

    Ashanujjaman, K

    S. Ashanujjaman, K. Ghosh, and K. Huitu, Phys. Rev. D 106, 075028 (2022), 2205.14983

    2022 arXiv

  73. [81]

    Ashanujjaman and K

    S. Ashanujjaman and K. Ghosh, JHEP 03, 195 (2022), 2108.10952

    2022

  74. [82]

    O. A. Ducu et al., (2024), 2410.14830

    2024

  75. [83]

    Primulando, J

    R. Primulando, J. Julio, and P. Uttayarat, JHEP 08, 024 (2019), 1903.02493

    2019 arXiv

  76. [84]

    A. Das, S. Mandal, and S. Shil, Phys. Rev. D 108, 015022 (2023), 2304.06298

    2023 arXiv

  77. [85]

    A. Das, P. Das, and N. Okada, (2024), 2405.11820

    2024 arXiv

  78. [86]

    A. Das, J. Li, S. Mandal, T. Nomura, and R. Zhang, (2024), 2410.21956

    2024 arXiv

  79. [87]

    Amaldi et al., Phys

    U. Amaldi et al., Phys. Rev. D 36, 1385 (1987)

    1987

  80. [88]

    Costa, J

    G. Costa, J. R. Ellis, G. L. Fogli, D. V. Nanopoulos, and F. Zwirner, Nucl. Phys. B 297, 244 (1988)

    1988

  81. [89]

    Langacker and M.-x

    P. Langacker and M.-x. Luo, Phys. Rev. D44, 817 (1991)

    1991

  82. [90]

    Erler and P

    J. Erler and P. Langacker, Phys. Rev. D 52, 441 (1995), hep-ph/9411203

    1995 arXiv

  83. [91]

    Haller et al

    J. Haller et al. , Eur. Phys. J. C 78, 675 (2018), 1803.01853

    2018 arXiv

  84. [92]

    Particle Data Group, R. L. Workman et al., PTEP 2022, 083C01 (2022)

    2022

  85. [93]

    Navas et al., Phys

    Particle Data Group, S. Navas et al., Phys. Rev. D 110, 030001 (2024)

    2024

  86. [94]

    Chatrchyan et al., Phys

    CMS, S. Chatrchyan et al., Phys. Lett. B 716, 30 (2012), 1207.7235

    2012 arXiv

  87. [95]

    Aad et al., Phys

    ATLAS, G. Aad et al., Phys. Lett. B 716, 1 (2012), 1207.7214

    2012 arXiv

  88. [96]

    Aaboud et al., Phys

    ATLAS, M. Aaboud et al., Phys. Lett. B 784, 345 (2018), 1806.00242

    2018 arXiv

  89. [97]

    CMS, A. M. Sirunyan et al., Phys. Lett. B 805, 135425 (2020), 2002.06398

    2020 arXiv

  90. [98]

    ATLAS, ATLAS-CONF-2021-053 (2021)

    2021

  91. [99]

    Collaboration, Cms analysis physics summary, https://cds.cern.ch/record/2706103/files/HIG-19- 005-pas.pdf, 2019, CMS PAS HIG-2019-005

    C. Collaboration, Cms analysis physics summary, https://cds.cern.ch/record/2706103/files/HIG-19- 005-pas.pdf, 2019, CMS PAS HIG-2019-005

    2019

  92. [100]

    Bonilla, R

    C. Bonilla, R. M. Fonseca, and J. W. F. Valle, Phys. Rev. D 92, 075028 (2015), 1508.02323

    2015 arXiv

  93. [101]

    Staub, (2008), 0806.0538

    F. Staub, (2008), 0806.0538

    2008 arXiv

  94. [102]

    Staub, Comput

    F. Staub, Comput. Phys. Commun. 182, 808 (2011), 1002.0840

    2011 arXiv

  95. [103]

    Staub, Adv

    F. Staub, Adv. High Energy Phys. 2015, 840780 (2015), 1503.04200

    2015 arXiv

  96. [104]

    M. D. Goodsell, K. Nickel, and F. Staub, Eur. Phys. J. C 75, 32 (2015), 1411.0675

    2015 arXiv

  97. [105]

    Goodsell, K

    M. Goodsell, K. Nickel, and F. Staub, Eur. Phys. J. C 75, 290 (2015), 1503.03098

    2015 arXiv

  98. [106]

    Porod, Comput

    W. Porod, Comput. Phys. Commun. 153, 275 (2003), hep-ph/0301101

    2003 arXiv

  99. [107]

    Porod and F

    W. Porod and F. Staub, Comput. Phys. Commun. 183, 2458 (2012), 1104.1573

    2012 arXiv

  100. [108]

    Porod, F

    W. Porod, F. Staub, and A. Vicente, Eur. Phys. J. C 74, 2992 (2014), 1405.1434

    2014 arXiv

  101. [109]

    Bechtle, O

    P. Bechtle, O. Brein, S. Heinemeyer, G. Weiglein, and K. E. Williams, Comput. Phys. Commun. 181, 138 (2010), 0811.4169

    2010 arXiv

  102. [110]

    Bechtle, O

    P. Bechtle, O. Brein, S. Heinemeyer, G. Weiglein, and K. E. Williams, Comput. Phys. Commun. 182, 2605 (2011), 1102.1898

    2011 arXiv

  103. [111]

    Bechtle et al., PoS CHARGED2012, 024 (2012), 1301.2345

    P. Bechtle et al., PoS CHARGED2012, 024 (2012), 1301.2345

    2012 arXiv

  104. [112]

    Bechtle et al., Eur

    P. Bechtle et al., Eur. Phys. J. C 74, 2693 (2014), 1311.0055

    2014 arXiv

  105. [113]

    Bechtle, S

    P. Bechtle, S. Heinemeyer, O. Stal, T. Stefaniak, and G. Weiglein, Eur. Phys. J. C 75, 421 (2015), 1507.06706

    2015 arXiv

  106. [114]

    Bechtle et al., Eur

    P. Bechtle et al., Eur. Phys. J. C 80, 1211 (2020), 2006.06007

    2020 arXiv

  107. [115]

    Bechtle, S

    P. Bechtle, S. Heinemeyer, O. St ˚ al, T. Stefaniak, and G. Weiglein, Eur. Phys. J. C 74, 2711 (2014), 1305.1933

    2014 arXiv

  108. [116]

    St ˚ al and T

    O. St ˚ al and T. Stefaniak, PoS EPS-HEP2013, 314 (2013), 1310.4039

    2013 arXiv

  109. [117]

    Bechtle, S

    P. Bechtle, S. Heinemeyer, O. St ˚ al, T. Stefaniak, and G. Weiglein, JHEP 11, 039 (2014), 1403.1582

    2014 arXiv

  110. [118]

    Bechtle et al., Eur

    P. Bechtle et al., Eur. Phys. J. C 81, 145 (2021), 2012.09197. Rajneil Baruah et al.: Normalizing Flow-Assisted Nested Sampling on Type-II Seesaw Model 17

    2021 arXiv

  111. [119]

    Bahl et al., Comput

    H. Bahl et al., Comput. Phys. Commun. 291, 108803 (2023), 2210.09332

    2023 arXiv

  112. [120]

    Ashton et al., Nature Reviews Methods Primers 2, 39 (2022)

    G. Ashton et al., Nature Reviews Methods Primers 2, 39 (2022)

    2022

  113. [121]

    Amos et al., Foundations and Trends ® in Machine Learning 16, 592 (2023)

    B. Amos et al., Foundations and Trends ® in Machine Learning 16, 592 (2023)

    2023

  114. [122]

    Mitchell, Machine LearningMcGraw-Hill international editions - computer science series (McGraw-Hill Educa- tion, 1997)

    T. Mitchell, Machine LearningMcGraw-Hill international editions - computer science series (McGraw-Hill Educa- tion, 1997)

    1997

  115. [123]

    A. Y. Ng and M. I. Jordan, On discriminative vs. gen- erative classifiers: a comparison of logistic regression and naive bayes, in Proceedings of the 15th International Con- ference on Neural Information Processing Systems: Natu- ral and Synthetic, NIPS’01, p. 841–848, Cambrid...

    2001

  116. [124]

    Bhattacharya, S

    S. Bhattacharya, S. Nandi, S. K. Patra, and S. Sahoo, (2020), 2008.04316

    2020 arXiv

  117. [125]

    D. P. Kingma, arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  118. [126]

    Goodfellow et al., Advances in neural information pro- cessing systems 27 (2014)

    I. Goodfellow et al., Advances in neural information pro- cessing systems 27 (2014)

    2014

  119. [127]

    D. P. Kingma, arXiv preprint arXiv:1312.6114 (2013)

    2013 arXiv

  120. [128]

    Papamakarios, E

    G. Papamakarios, E. Nalisnick, D. J. Rezende, S. Mo- hamed, and B. Lakshminarayanan, Journal of Machine Learning Research 22, 1 (2021)

    2021

  121. [129]

    Kobyzev, S

    I. Kobyzev, S. J. Prince, and M. A. Brubaker, IEEE transactions on pattern analysis and machine intelligence 43, 3964 (2020)

    2020

  122. [130]

    L. Dinh, J. Sohl-Dickstein, and S. Bengio, arXiv preprint arXiv:1605.08803 (2016)

    2016 arXiv

  123. [131]

    Aad et al., JHEP 06, 146 (2021), 2101.11961

    ATLAS, G. Aad et al., JHEP 06, 146 (2021), 2101.11961

    2021 arXiv

  124. [132]

    Aad et al., Phys

    ATLAS, G. Aad et al., Phys. Lett. B 860, 139137 (2025), 2407.10798

    2025 arXiv

  125. [134]

    CMS, A. M. Sirunyan et al., Eur. Phys. J. C 81, 723 (2021), 2104.04762

    2021 arXiv

  126. [135]

    Alwall et al., JHEP 07, 079 (2014), 1405.0301

    J. Alwall et al., JHEP 07, 079 (2014), 1405.0301

    2014 arXiv

  127. [136]

    Frederix et al., JHEP 07, 185 (2018), 1804.10017, [Erratum: JHEP 11, 085 (2021)]

    R. Frederix et al., JHEP 07, 185 (2018), 1804.10017, [Erratum: JHEP 11, 085 (2021)]

    2018 arXiv

  128. [137]

    S. M. Lundberg and S.-I. Lee, A unified approach to interpreting model predictions, in Advances in Neural Information Processing Systems 30, edited by I. Guyon et al., pp. 4765–4774, Curran Associates, Inc., 2017

    2017

  129. [138]

    Papamakarios, T

    G. Papamakarios, T. Pavlakou, and I. Murray, Advances in neural information processing systems 30 (2017)

    2017

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.