Pith. sign in

REVIEW 3 major objections 5 minor 60 references

A Comprehensive Search for Leptoquarks Decaying into Top-$\tau$ Final States at the Future LHC

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 A machine-learning search is projected to set 95% CL exclusion limits on top-tau leptoquarks up to 1.63 TeV at 200 fb^-1 and 1.77 TeV at 500 fb^-1 of 14 TeV LHC data.

desk verdict Useful ML-based search projection for LQd3 -> t tau at HL-LHC; the 1.63-1.77 TeV reach is plausible but rests on an unvalidated tagger, so it is a template rather than a prediction. read the letter →

arxiv 2501.07543 v1 pith:L6IA2ZS6 submitted 2025-01-13 hep-ph

Debabrata Sahoo , Rameswar Sahu , Kirtiman Ghosh This is my paper · ORCID
classification hep-ph
keywords scalarleptoquarksthird-generationtop-taufinalstatesboostedobjecttagginggraphneuralnetworkdecisiontreesLHCphenomenologyexpectedexclusionlimits
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper asks whether a machine-learning-driven search strategy can improve the LHC's ability to find third-generation scalar leptoquarks that decay exclusively into a top quark and a tau lepton. It claims that, at a 14 TeV proton-proton collider, combining a graph-neural-network tagger for boosted hadronic tops, W/Z, and Higgs bosons with signal-region-specific boosted-decision-tree classifiers yields expected 95% CL exclusion of leptoquark masses up to 1.63 TeV with 200 $fb^{-1}$ and 1.77 TeV with 500 $fb^{-1}$. That would extend current experimental bounds, which sit near 1.4 TeV, with only a modest luminosity extrapolation. A sympathetic reader should care because the analysis provides a concrete, simulation-based template for the coming high-luminosity LHC searches.

What carries the argument

The central machinery is the two-stage ML pipeline: (i) a Lorentz-equivariant graph neural network (the LorentzNet architecture with four equivariant blocks) trained to classify large-radius jets as coming from boosted top quarks, W/Z bosons, Higgs bosons, or QCD, producing continuous top/V/Higgs scores; and (ii) per-signal-region boosted decision trees that combine these scores with universal and signal-region-specific kinematic features. The signal regions themselves—fifteen mutually exclusive categories in lepton multiplicity, tagged boosted-object multiplicity, hadronic-tau multiplicity, and b-tag multiplicity—are the organizing device that lets the search cover fully hadronic, single-lepton, and multi-lepton decay topologies. The final observable is the BDT score distribution in the 0.7–1.0 range, binned into 12 bins and fed to a profile-likelihood limit calculation with log-normal nuisance parameters.

What would settle it

Re-run the analysis pipeline using full Geant4-based detector simulation instead of fast simulation, or calibrate the jet taggers on a control sample of ttbar events from real data; if the expected 95% CL mass reach falls below 1.63 TeV at 200 $fb^{-1}$, the central claim is falsified.

Watch

Extended reading notes

Core claim

The central claim is that the proposed search, built around fifteen mutually exclusive signal regions defined by lepton, boosted-tag, tau-tag, and b-tag multiplicities, can set expected 95% CL upper limits on the LQ^d_3 pair-production cross-section down to masses of 1.63 TeV and 1.77 TeV at integrated luminosities of 200 and 500 $fb^{-1}$ at the 14 TeV LHC, assuming a 100% branching ratio to t-tau. The improvement over existing searches comes from replacing a handful of kinematic cuts with a two-stage machine-learning pipeline: a Lorentz-equivariant graph neural network assigns each large-radius jet a top/V/Higgs score, and boosted decision trees per signal region use those scores plus kinematic features such as effective mass and M_{J tau} to suppress Standard Model backgrounds. The expected limits are derived from a combined profile-likelihood fit over the BDT-score distributions, with flat 20% systematic uncertainties on signal and background in each bin.

Load-bearing premise

The reach projection assumes the graph-neural-network and boosted-decision-tree classifiers perform as well on real detector data as on the simulated events used to train and test them, and a flat 20% systematic uncertainty may not cover a tagger-performance shift.

Editorial extensions

If this is right

  • The expected 95% CL exclusion reach extends to 1.63 TeV at 200 fb^-1 and 1.77 TeV at 500 fb^-1, above the current experimental limit near 1.4 TeV.
  • The use of hadronically decaying boosted tops as fat jets increases signal acceptance compared to lepton-only final states, making the search sensitive to fully hadronic topologies.
  • The fifteen orthogonal signal regions cover fully hadronic, single-lepton, and multi-lepton decay cascades, so the analysis does not rely on one topology.
  • Feature ranking shows that effective mass and the M_{J tau} invariant mass are the strongest discriminants, guiding future search design.
  • If a scalar leptoquark with mass below 1.77 TeV decaying to top-tau with 100% branching ratio exists, the combined search would exclude it at 95% CL with 500 fb^-1 of 14 TeV data.

Reading between the lines

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

  • The same ML-tagging-plus-BDT template could be adapted to other signals featuring boosted tops, such as vector-like T quarks or charged Higgs production, where the continuous tagger scores would likely improve discrimination over cut-based searches.
  • A key risk not addressed by the paper is tagger calibration: if the graph neural network's top-tagging efficiency on real data is lower than in fast simulation, the reach could degrade by more than the flat 20% systematic assumes.
  • The 100% branching-ratio assumption is a strong idealization; if LQ^d_3 decays to b nu or other modes with nonnegligible rates, the t-tau reach would shrink, and the limits should be rescaled as a function of the t-tau branching fraction.
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 studies pair production of the third-generation scalar leptoquark LQd3 at the 14 TeV LHC, assuming a 100% branching ratio into top-τ final states. It proposes a search strategy that uses a LorentzNet-based GNN multi-class tagger to identify boosted hadronically decaying top, W/Z, and Higgs jets, followed by xgboost BDT classifiers trained separately for 15 mutually exclusive signal regions that cover 0-, 1-, and 2-lepton topologies. Expected 95% CL upper limits on the LQd3 pair-production cross-section are computed using the HistFactory/pyhf framework with flat 20% per-bin systematic uncertainties. The central result is an expected exclusion reach of 1.63 TeV at 200 fb^-1 and 1.77 TeV at 500 fb^-1 of 14 TeV LHC data, extending beyond current experimental bounds near 1.4 TeV.

Significance. If the quoted reach is robust, the paper provides a concrete, modern search template for third-generation leptoquarks at the HL-LHC and demonstrates the value of ML-based boosted-object tagging in a realistic analysis workflow. The strengths of the paper are its internal coherence: the Monte Carlo chain (MadGraph+PYTHIA+DELPHES) is standard, the signal-region definitions are disjoint and physically motivated, the feature-selection procedure is documented, the statistical treatment follows the widely used pyhf/HistFactory approach, and the signal normalization uses external NLO/NNLO+NNLL cross-sections. The main weakness is that the entire expected reach is driven by GNN tagger scores that are validated only against the same fast-simulation samples used for the rest of the analysis, with no closure test against full simulation or data and no correlated systematic assigned to tagger performance. The paper would be considerably strengthened by a robustness scan that degrades the tagger confusion matrix and shows the resulting change in the mass reach.

major comments (3)
  1. [Sec. III A 1, Table I; Sec. III C, Table V; Sec. III D, Fig. 3] The central mass-reach claim (1.63 TeV at 200 fb^-1 and 1.77 TeV at 500 fb^-1) depends on the GNN tagger scores SJ_T, SJ_V, and SJ_H being effective BDT inputs, especially in the 0-lepton signal regions. The only validation of these scores is the confusion matrix in Table I, computed for 500-600 GeV fat jets generated with the same MadGraph+PYTHIA+DELPHES chain used for the signal and background samples. No comparison with a full Geant4-based detector simulation, a different pileup scenario, or data is provided. The flat 20% per-bin nuisance parameters introduced in Sec. III D modify the overall signal and background normalization in each BDT-score bin but do not model a correlated shift in tagger efficiency or QCD mistag rate. Because QCD multijet is the dominant background in the 0-lepton regions (e.g., 3.5e4 events for 0l1J1tau_h at 500 fb^-1 versus 8.4 signal events in Table III), a modest degradation of the tagger — say, top-tagging efficiency dropping from 0.902 toward 0.7 or QCD rejection dropping from 20 toward 10 — would substantially alter the expected limit and could remove the claimed improvement over the existing ~1.4 TeV bounds. I ask the authors to add a sensitivity scan where the confusion matrix is degraded by a plausible amount, or to include a correlated nuisance parameter that shifts the tagger scores, and to report how the final mass reach changes.
  2. [Sec. III B (first paragraph) versus Sec. III A (Object Reconstruction)] There is a direct contradiction in the definition of the tagged boosted object J. Section III A states that fat jets are reconstructed with anti-kT and radius parameter R=1.2, and the GNN tagger is described as acting on these R=1.2 jets. Section III B, however, defines the tagged SM boosted heavy particles J as "R = 1.0 radius jets." Since the BDT features for signal regions containing J use the GNN tagger scores (Table V), the actual radius used in the analysis must be specified unambiguously, and the tagger performance in Table I should correspond to that same radius. As written, the reader cannot determine which jet radius enters the signal regions, the BDT training, and the final limits.
  3. [Sec. III D, likelihood expression] The text says that the nuisance constraints g(theta, Delta) are log-normal, but the displayed formula is a Gaussian in theta: g(theta, Delta) = (1/sqrt(2pi)Delta) exp(-theta^2/(2 Delta^2)). These are different parameterizations, and pyhf's standard log-normal constraint is not the same as a Gaussian. The authors should either show the correct log-normal form or explicitly state that Gaussian constraints were used and confirm that the actual pyhf implementation matches the published formula. This matters because the per-bin 20% systematics are the only uncertainties included in the limit calculation, and the shape of the constraint affects the resulting CLs values, especially in bins with small background counts.
minor comments (5)
  1. [Sec. III A 1] The phrase "bosted parent particle" contains a typo; it should be "boosted parent particle."
  2. [Sec. III A 1] The Higgs jet sample is said to use the "default heft model [] of MadGraph," but the citation is missing; please add the appropriate reference for the Higgs effective field theory model.
  3. [Table III] The table header is garbled and does not clearly separate the background process columns (t-tbar+tW, W/Z+jets, t-tbar V, 2-4 V, multijet, 4-top) from the signal column. This makes the table difficult to read and should be reformatted with clear column labels.
  4. [Fig. 1 and Fig. 2 captions] The y-axis label in Fig. 1, "1/N dN/0.5", is nonstandard; please use a proper density notation with the bin width. In addition, the Fig. 2 inset is described as showing events at 1000 fb^-1, while the analysis uses 200 and 500 fb^-1; please clarify whether 1000 fb^-1 is used only for illustration or is part of the statistical treatment.
  5. [Sec. III C, training sample] The BDT training uses an equal mixture of leptoquark signals with masses from 1.2 to 2.0 TeV. The paper does not discuss how the limit scan handles masses at the edges of or outside this range; a brief comment on the expected performance for masses beyond 2.0 TeV would be useful, though this does not affect the quoted reach within the trained range.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the search projection is a standard simulation-based limit-setting exercise; no fitted parameter is repackaged as a prediction, and the self-citations are not load-bearing.

full rationale

The derivation chain is self-contained in the relevant sense. Signal and background event counts in Table III come from MadGraph+PYTHIA+DELPHES simulations normalized to external NLO/NNLO cross sections, not from any parameter fitted to the final observable. The GNN tagger is trained and evaluated by the authors on their own simulated samples, but the BDTs are trained on one simulated sample and tested on statistically independent simulated samples, which is standard generalization testing rather than a circular reduction. The expected 95% CL limits in Sec. III D are computed with the HistFactory/pyhf likelihood using Asimov counts, so no observed data are used to define the prediction. The citations to the authors' own tagger papers are methodological references for dataset generation and LorentzNet usage, but the tagger is reimplemented and revalidated here with its own confusion matrix, and the architecture is externally sourced from LorentzNet; hence the self-citations are not load-bearing. The only textual anomaly is an empty citation for the MadGraph heft model, which is an editorial omission and has no bearing on circularity. The internal fast-simulation validation of tagger and classifier performance is a physics-validity limitation (real-detector closure is untested), not a circularity: no equation in the paper reduces the claimed reach to the tagger's own training objective by construction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central projection rests on the existence and assumed couplings of a known scalar leptoquark, external cross-section calculations, a fast-simulation detector model, and the generalization of ML taggers trained on the same simulation chain. No new physical entity is introduced. The only hand-set numerical inputs with direct impact on the limit are the flat 20% systematic widths and classifier training choices.

free parameters (3)
  • Flat systematic uncertainty Delta_s = Delta_b = 0.2
    Chosen by hand as log-normal nuisance widths for signal and background in each BDT score bin; directly sets the CLs exclusion reach (Sec. III D).
  • QCD threshold score for tagger confusion matrix = 0.4
    Used to define discrete fat-jet classes and reported tagger efficiencies in Table I; the final analysis uses continuous scores, so the impact on the limit is indirect.
  • Signal weighting fractions in BDT training = 0.2, 0.1, 0.05 for 0, 1, and 2 lepton signal regions
    Hand-chosen to balance signal and background samples during BDT training; affects classifier performance but is a training choice rather than a physics input.
assumptions (5)
  • ad hoc to paper A scalar leptoquark LQd3 exists with gauge quantum numbers allowing top-tau coupling and decays 100% to top-tau.
    Model setup in Sec. II; this is a proof-of-concept benchmark branching ratio, not derived from data.
  • domain assumption Leptoquark pair production cross-sections from Ref. [42,43] at NLO and the theory line at NNLO+NNLL are correct.
    Used as the theory prediction for signal rates and for comparing the exclusion limit without recalculating the cross-section.
  • domain assumption DELPHES with the default ATLAS card adequately models detector response for R=0.4 and R=1.2 jets, b-tagging, tau-tagging, and missing transverse energy.
    Object reconstruction in Sec. III A relies entirely on fast simulation with no validation against full ATLAS or CMS simulation.
  • domain assumption The GNN tagger trained on generated samples generalizes to actual LHC events.
    Sec. III A 1 and Table I report tagger performance only within the same simulation chain; no data closure test is shown.
  • domain assumption The signal regions are mutually exclusive and statistically independent, justifying the product of per-region likelihoods.
    Sec. III B and III D; regions are disjoint by object-count categories, but independent nuisance parameters per bin are assumed.

how reviews work

0 comments
Cite this review

Pith. "Pith review of A Comprehensive Search for Leptoquarks Decaying into Top-$\tau$ Final States at the Future LHC." pith.science (2026). https://pith.science/paper/L6IA2ZS6

@misc{pith2026250107543,
  author       = {Pith},
  title        = {Pith review of: A Comprehensive Search for Leptoquarks Decaying into Top-$\tau$ Final States at the Future LHC},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L6IA2ZS6}},
  note         = {Machine review of arXiv:2501.07543}
}
read the original abstract

We studied the collider phenomenology of third-generation scalar leptoquarks at the Large Hadron Collider (LHC) with a 14 TeV center-of-mass energy. The analysis focuses on leptoquarks decaying exclusively into top quarks and tau leptons, employing machine learning-based tagging techniques for identifying hadronically decaying boosted top quarks, W/Z, and Higgs bosons, as well as a multivariate classifier to distinguish signal events from Standard Model (SM) backgrounds. The expected 95% confidence level (CL) upper limits on the leptoquark production cross-section are computed assuming integrated luminosities of 200 and 500 inverse femtobarns at the 14 TeV LHC. The results demonstrate significant sensitivity improvements for detecting leptoquarks at masses beyond the current experimental limits.

Figures

Figures reproduced from arXiv: 2501.07543 by the authors.

Figure 1
Figure 1. FIG. 1. Normalized distribution of the top score (left), [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. The main figure shows the normalized BDT score distribution for the signal with a leptoquark mass of 1.5 TeV and [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. Expected (dashed line) 95% CL upper limits on the [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

60 extracted references · 23 canonical work pages

  1. [1]

    With this motivation, we have reconstructed large-radius (R = 1.2) jets, referred to as ”fat jets”, in the previous section

    Boosted object tagging In this analysis, our goal is to exploit the presence of boosted top quarks from the decay of TeV-scale lepto- quarks to suppress SM background contributions. With this motivation, we have reconstructed large-radius (R = 1.2) jets, referred to as ”fat jets”, in the previous section. Such jets can also originate from the production o...

    work page

  2. [2]

    Georgi and S

    H. Georgi and S. L. Glashow, Unity of All Elementary Particle Forces, Phys. Rev. Lett. 32, 438 (1974)

    1974

  3. [3]

    For training the classi- fiers, simulated signal and background samples are used for each signal region

    The Multivariate Classifiers To discriminate between signal events and SM back- ground, we developed multivariate classifiers for each sig- nal region using Boosted Decision Trees (BDTs) imple- mented with the xgboost library. For training the classi- fiers, simulated signal and background samples are used for each signal region. In the absence of specifi...

    work page

  4. [4]

    J. C. Pati and A. Salam, Lepton Number as the Fourth Color, Phys. Rev. D 10, 275 (1974), [Erratum: Phys.Rev.D 11, 703–703 (1975)]

    1974

  5. [5]

    A Comprehensive Search for Leptoquarks Decaying into Top-$\tau$ Final States at the Future LHC

    and technicolor model [6]. Leptoquarks can couple to both leptons and quarks and have color charge. Recently, there has been a boom in studies related to leptoquarks because of the interest in the flavor sector. Leptoquarks are one of the promising candidates for explaining the observed B-meson anomaly known as RK(∗) and RD(∗) [7–16], although statistical...

    work page Pith review arXiv 2025

  6. [6]

    Georgi, The State of the Art—Gauge Theories, AIP Conf

    H. Georgi, The State of the Art—Gauge Theories, AIP Conf. Proc. 23, 575 (1975)

    work page 1975

  7. [7]

    Fritzsch and P

    H. Fritzsch and P. Minkowski, Unified Interactions of Leptons and Hadrons, Annals Phys. 93, 193 (1975)

    work page 1975

  8. [8]

    Dimopoulos and L

    S. Dimopoulos and L. Susskind, Mass Without Scalars, Nucl. Phys. B 155, 237 (1979)

    work page 1979

Show all 60 references
  1. [9]

    Farhi and L

    E. Farhi and L. Susskind, Technicolor, Phys. Rept. 74, 277 (1981)

    1981

  2. [10]

    Aaij et al

    R. Aaij et al. (LHCb), Measurement of Form-Factor- Independent Observables in the Decay B0 → K ∗0µ+µ−, Phys. Rev. Lett. 111, 191801 (2013), arXiv:1308.1707 [hep-ex]

    2013 arXiv

  3. [11]

    J. P. Lees et al. (BaBar), Evidence for an excess of ¯B → D(∗)τ − ¯ντ decays, Phys. Rev. Lett. 109, 101802 (2012), arXiv:1205.5442 [hep-ex]

    2012 arXiv

  4. [12]

    J. P. Lees et al. (BaBar), Measurement of an Ex- cess of ¯B → D(∗)τ − ¯ντ Decays and Implications for Charged Higgs Bosons, Phys. Rev. D 88, 072012 (2013), arXiv:1303.0571 [hep-ex]

    2013 arXiv

  5. [13]

    Fajfer and N

    S. Fajfer and N. Koˇ snik, Vector leptoquark resolution of RK and RD(∗) puzzles, Phys. Lett. B 755, 270 (2016), arXiv:1511.06024 [hep-ph]

    2016 arXiv

  6. [14]

    Barbieri, C

    R. Barbieri, C. W. Murphy, and F. Senia, B-decay Anomalies in a Composite Leptoquark Model, Eur. Phys. J. C 77, 8 (2017), arXiv:1611.04930 [hep-ph]

    2017 arXiv

  7. [15]

    Beˇ cirevi´ c, S

    D. Beˇ cirevi´ c, S. Fajfer, and N. Koˇ snik, Lepton flavor nonuniversality in b→sℓ+ℓ− processes, Phys. Rev. D 92, 014016 (2015), arXiv:1503.09024 [hep-ph]

    2015 arXiv

  8. [16]

    Dorˇ sner, S

    I. Dorˇ sner, S. Fajfer, N. Koˇ snik, and I. Niˇ sandˇ zi´ c, Mini- mally flavored colored scalar in ¯B → D(∗)τ ¯ν and the mass matrices constraints, JHEP 11, 084, arXiv:1306.6493 [hep-ph]

    arXiv

  9. [17]

    Bauer and M

    M. Bauer and M. Neubert, Minimal Leptoquark Expla- nation for the RD(∗) , RK , and ( g − 2)µ Anomalies, Phys. Rev. Lett. 116, 141802 (2016), arXiv:1511.01900 [hep-ph]

    2016 arXiv

  10. [18]

    Beˇ cirevi´ c, N

    D. Beˇ cirevi´ c, N. Koˇ snik, O. Sumensari, and R. Zukanovich Funchal, Palatable Leptoquark Scenarios for Lepton Flavor Violation in Exclusive b → sℓ1ℓ2 modes, JHEP 11, 035, arXiv:1608.07583 [hep-ph]

    arXiv

  11. [19]

    Beˇ cirevi´ c, S

    D. Beˇ cirevi´ c, S. Fajfer, N. Koˇ snik, and O. Sumen- sari, Leptoquark model to explain the B-physics anoma- lies, RK and RD, Phys. Rev. D 94, 115021 (2016), arXiv:1608.08501 [hep-ph]

    2016 arXiv

  12. [20]

    Abdesselam et al

    A. Abdesselam et al. (Belle), Test of Lepton-Flavor Uni- versality in B → K ∗ℓ+ℓ− Decays at Belle, Phys. Rev. Lett. 126, 161801 (2021), arXiv:1904.02440 [hep-ex]

    2021 arXiv

  13. [21]

    Aaij et al

    R. Aaij et al. (LHCb), Measurement of lepton uni- versality parameters in B+ → K +ℓ+ℓ− and B0 → K ∗0ℓ+ℓ− decays, Phys. Rev. D 108, 032002 (2023), arXiv:2212.09153 [hep-ex]

    2023

  14. [22]

    Parashar, A

    S. Parashar, A. Karan, Avnish, P. Bandyopadhyay, and K. Ghosh, Phenomenology of scalar leptoquarks at the LHC in explaining the radiative neutrino masses, muon g-2, and lepton flavor violating observables, Phys. Rev. D 106, 095040 (2022), arXiv:2209.05890 [hep-ph]

    2022 arXiv

  15. [23]

    Choi, Y.-J

    S.-M. Choi, Y.-J. Kang, H. M. Lee, and T.-G. Ro, Lepto-Quark Portal Dark Matter, JHEP 10, 104, arXiv:1807.06547 [hep-ph]

    arXiv

  16. [24]

    Mandal, Fermionic dark matter in leptoquark portal, Eur

    R. Mandal, Fermionic dark matter in leptoquark portal, Eur. Phys. J. C 78, 726 (2018), arXiv:1808.07844 [hep- ph]

    2018 arXiv

  17. [25]

    Aad et al

    G. Aad et al. (ATLAS), Exploration at the high- energy frontier: ATLAS Run 2 searches investigating the exotic jungle beyond the Standard Model, (2024), arXiv:2403.09292 [hep-ex]

    2024 arXiv

  18. [26]

    Aad et al

    G. Aad et al. (ATLAS), Search for pairs of scalar lep- toquarks decaying into quarks and electrons or muons in √s = 13 TeV pp collisions with the ATLAS detector, JHEP 10, 112, arXiv:2006.05872 [hep-ex]

    2006 arXiv

  19. [27]

    A. M. Sirunyan et al. (CMS), Search for pair production of second-generation leptoquarks at √s = 13 TeV, Phys. Rev. D 99, 032014 (2019), arXiv:1808.05082 [hep-ex]

    2019 arXiv

  20. [28]

    A. M. Sirunyan et al. (CMS), Search for pair production of first-generation scalar leptoquarks at √s = 13 TeV, Phys. Rev. D 99, 052002 (2019), arXiv:1811.01197 [hep- ex]

    2019 arXiv

  21. [29]

    Aad et al

    G. Aad et al. (ATLAS), Combination of searches for pair-produced leptoquarks at s=13 TeV with the AT- LAS detector, Phys. Lett. B 854, 138736 (2024), arXiv:2401.11928 [hep-ex]

    2024 arXiv

  22. [30]

    Aad et al

    G. Aad et al. (ATLAS), Search for new phenomena in pp collisions in final states with tau leptons, b-jets, and missing transverse momentum with the ATLAS detector, Phys. Rev. D 104, 112005 (2021), arXiv:2108.07665 [hep- ex]

    2021 arXiv

  23. [31]

    Aad et al

    G. Aad et al. (ATLAS), Search for pair production of third-generation leptoquarks decaying into a bottom quark and a τ -lepton with the ATLAS detector, Eur. Phys. J. C 83, 1075 (2023), arXiv:2303.01294 [hep-ex]

    2023 arXiv

  24. [32]

    Aad et al

    G. Aad et al. (ATLAS), Search for pair production of third-generation scalar leptoquarks decaying into a top quark and a τ -lepton in pp collisions at √s = 13 TeV with the ATLAS detector, JHEP 06, 179, arXiv:2101.11582 [hep-ex]

    arXiv

  25. [33]

    Aad et al

    G. Aad et al. (ATLAS), Search for a scalar partner of the top quark in the all-hadronic t¯t plus missing trans- verse momentum final state at √s = 13 TeV with the ATLAS detector, Eur. Phys. J. C 80, 737 (2020), arXiv:2004.14060 [hep-ex]

    2020 arXiv

  26. [34]

    Aad et al

    G. Aad et al. (ATLAS), Search for new phenomena in final states with b-jets and missing transverse momentum in √s = 13 TeV pp collisions with the ATLAS detector, JHEP 05, 093, arXiv:2101.12527 [hep-ex]

    arXiv

  27. [35]

    R. Sahu, S. Ashanujjaman, and K. Ghosh, Unveiling the secrets of new physics through top quark tagging, Eur. Phys. J. ST 233, 2465 (2024), arXiv:2409.12085 [hep-ph]. 12

    2024

  28. [36]

    Sahu and K

    R. Sahu and K. Ghosh, ML-Based Top Taggers: Performance, Uncertainty and Impact of Tower & Tracker Data Integration, SciPost Phys. 17, 166 (2024), arXiv:2309.01568 [hep-ph]

    2024 arXiv

  29. [37]

    Butter, G

    A. Butter, G. Kasieczka, T. Plehn, and M. Russell, Deep- learned Top Tagging with a Lorentz Layer, SciPost Phys. 5, 028 (2018), arXiv:1707.08966 [hep-ph]

    2018 arXiv

  30. [38]

    Buchmuller, R

    W. Buchmuller, R. Ruckl, and D. Wyler, Leptoquarks in Lepton - Quark Collisions, Phys. Lett. B191, 442 (1987), [Erratum: Phys.Lett.B 448, 320–320 (1999)]

    1987

  31. [39]

    Pontecorvo, Mesonium and anti-mesonium, Sov

    B. Pontecorvo, Mesonium and anti-mesonium, Sov. Phys. JETP 6, 429 (1957)

    1957

  32. [40]

    Pontecorvo, Inverse beta processes and nonconserva- tion of lepton charge, Zh

    B. Pontecorvo, Inverse beta processes and nonconserva- tion of lepton charge, Zh. Eksp. Teor. Fiz.34, 247 (1957)

    1957

  33. [41]

    Pontecorvo, Neutrino Experiments and the Problem of Conservation of Leptonic Charge, Zh

    B. Pontecorvo, Neutrino Experiments and the Problem of Conservation of Leptonic Charge, Zh. Eksp. Teor. Fiz. 53, 1717 (1967)

    1967

  34. [42]

    Z. Maki, M. Nakagawa, and S. Sakata, Remarks on the unified model of elementary particles, Prog. Theor. Phys. 28, 870 (1962)

    1962

  35. [43]

    Cabibbo, Unitary Symmetry and Leptonic Decays, Phys

    N. Cabibbo, Unitary Symmetry and Leptonic Decays, Phys. Rev. Lett. 10, 531 (1963)

    1963

  36. [44]

    Kobayashi and T

    M. Kobayashi and T. Maskawa, CP Violation in the Renormalizable Theory of Weak Interaction, Prog. Theor. Phys. 49, 652 (1973)

    1973

  37. [45]

    Mandal, S

    T. Mandal, S. Mitra, and S. Seth, Pair Production of Scalar Leptoquarks at the LHC to NLO Parton Shower Accuracy, Phys. Rev. D 93, 035018 (2016), arXiv:1506.07369 [hep-ph]

    2016 arXiv

  38. [46]

    Borschensky, B

    C. Borschensky, B. Fuks, A. Kulesza, and D. Schwartl¨ ander, Scalar leptoquark pair produc- tion at hadron colliders, Phys. Rev. D 101, 115017 (2020), arXiv:2002.08971 [hep-ph]

    2020 arXiv

  39. [47]

    Alwall, R

    J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Maltoni, O. Mattelaer, H. S. Shao, T. Stelzer, P. Torrielli, and M. Zaro, The automated computation of tree-level and next-to-leading order differential cross sections, and their matching to parton shower simulations, JHEP 07...

    arXiv

  40. [48]

    Carrazza, S

    S. Carrazza, S. Forte, and J. Rojo, Parton Distribu- tions and Event Generators, in 43rd International Sym- posium on Multiparticle Dynamics (2013) pp. 89–96, arXiv:1311.5887 [hep-ph]

    2013 arXiv

  41. [49]

    Bierlich et al

    C. Bierlich et al. , A comprehensive guide to the physics and usage of PYTHIA 8.3, SciPost Phys. Codeb. 2022, 8 (2022), arXiv:2203.11601 [hep-ph]

    2022 arXiv

  42. [50]

    de Favereau, C

    J. de Favereau, C. Delaere, P. Demin, A. Giammanco, V. Lema ˆ ıtre, A. Mertens, and M. Selvaggi (DELPHES 3), DELPHES 3, A modular framework for fast simu- lation of a generic collider experiment, JHEP 02, 057, arXiv:1307.6346 [hep-ex]

    arXiv

  43. [51]

    Cacciari, G

    M. Cacciari, G. P. Salam, and G. Soyez, The anti- kt jet clustering algorithm, JHEP 04, 063, arXiv:0802.1189 [hep-ph]

    arXiv

  44. [52]

    Cacciari, G

    M. Cacciari, G. P. Salam, and G. Soyez, FastJet User Manual, Eur. Phys. J. C 72, 1896 (2012), arXiv:1111.6097 [hep-ph]

    2012 arXiv

  45. [53]

    S. Gong, Q. Meng, J. Zhang, H. Qu, C. Li, S. Qian, W. Du, Z.-M. Ma, and T.-Y. Liu, An efficient Lorentz equivariant graph neural network for jet tagging, JHEP 07, 030, arXiv:2201.08187 [hep-ph]

    arXiv

  46. [54]

    M. L. Mangano, M. Moretti, F. Piccinini, and M. Trec- cani, Matching matrix elements and shower evolution for top-quark production in hadronic collisions, JHEP 01, 013, arXiv:hep-ph/0611129

    arXiv

  47. [55]

    Hoeche, F

    S. Hoeche, F. Krauss, N. Lavesson, L. Lonnblad, M. Mangano, A. Schalicke, and S. Schumann, Match- ing parton showers and matrix elements, in HERA and the LHC: A Workshop on the Implications of HERA for LHC Physics: CERN - DESY Workshop 2004/2005 (Midterm Meeting, CERN, 11-13 O...

    2005 arXiv

  48. [56]

    Muselli, M

    C. Muselli, M. Bonvini, S. Forte, S. Marzani, and G. Ri- dolfi, Top Quark Pair Production beyond NNLO, JHEP 08, 076, arXiv:1505.02006 [hep-ph]

    arXiv

  49. [57]

    Catani, L

    S. Catani, L. Cieri, G. Ferrera, D. de Florian, and M. Grazzini, Vector boson production at hadron collid- ers: a fully exclusive QCD calculation at NNLO, Phys. Rev. Lett. 103, 082001 (2009), arXiv:0903.2120 [hep-ph]

    2009 arXiv

  50. [58]

    Balossini, G

    G. Balossini, G. Montagna, C. M. Carloni Calame, M. Moretti, O. Nicrosini, F. Piccinini, M. Treccani, and A. Vicini, Combination of electroweak and QCD correc- tions to single W production at the Fermilab Tevatron and the CERN LHC, JHEP 01, 013, arXiv:0907.0276 [hep-ph]

    arXiv

  51. [59]

    J. M. Campbell, R. K. Ellis, and C. Williams, Vec- tor Boson Pair Production at the LHC, JHEP 07, 018, arXiv:1105.0020 [hep-ph]

    arXiv

  52. [60]

    Cranmer, G

    K. Cranmer, G. Lewis, L. Moneta, A. Shibata, and W. Verkerke (ROOT), HistFactory: A tool for creat- ing statistical models for use with RooFit and RooStats, (2012)

    2012

Pith tools

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