Leptoquark searches at ATLAS and CMS

CMS Collaborations); Izaak Neutelings (for the ATLAS

Leptoquark searches at ATLAS and CMS

T0 review · 0 major / 8 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read No leptoquark signal in Run-2: pair production excluded up to ~2 TeV, nonresonant beyond 5 TeV.

desk verdict A competent, well-organized review of Run-2 leptoquark searches; no new physics, but a reliable map of current limits with a few minor typos. read the letter →

arxiv 2505.08738 v1 pith:CDLYVOKQ submitted 2025-05-13 hep-ex hep-ph

Izaak Neutelings (for the ATLAS , CMS Collaborations) This is my paper · ORCID

classification hep-exhep-ph

keywords leptoquarkATLASCMSLHCRun-2exclusionlimitsnonresonantproductionpairleptonflavoruniversality

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

Leptoquarks are hypothetical bosons that couple a quark to a lepton and could break lepton flavor universality, making them a plausible explanation for the B-meson anomalies. This review of ATLAS and CMS Run-2 searches reports that no leptoquark signal has been seen in any covered final state, so the collaborations set exclusion limits on leptoquark mass, coupling to fermions, and branching fractions. Under simplified benchmark models, pair production is excluded up to roughly 2 TeV, while nonresonant t-channel exchange is constrained beyond 5 TeV for large couplings. A newly highlighted ditau analysis is the first in that channel to include Drell-Yan interference and begins to probe the parameter region favored by the B anomalies.

What carries the argument

The engine of these results is the simplified benchmark parametrization: each search assumes a single leptoquark mass $m_{LQ}$, one coupling $\lambda$ (or $g$ for vector leptoquarks), a branching fraction $\beta$ to final states with charged leptons, and for vector leptoquarks a non-minimal gluon coupling $\kappa$, with $\kappa = 0$ minimal and $\kappa = 1$ Yang-Mills. On top of this, the searches exploit production-mode complementarity: pair production is QCD-driven and largely independent of $\lambda$, single production scales as $\lambda^2$, and nonresonant t-channel exchange scales as $\lambda^2/m_{LQ}^4$ and becomes a pure four-fermion interaction at high mass. The recent nonresonant analyses extract signals from the shape of mass and angular distributions—bins in $m_{\ell\ell}$, rapidity $|y|$, the Collins-Soper angle $\cos\theta^*$, or $b$-jet multiplicity—using templates that include the Drell-Yan interference.

What would settle it

A public reanalysis of the ATLAS and CMS likelihoods with a leptoquark model containing two non-zero couplings, or the observation of a clear excess in the Run-3 versions of these searches, would directly test whether the benchmark limits are robust and whether the no-signal conclusion holds.

Watch

Extended reading notes

Core claim

The central claim is that with the full Run-2 dataset at 13 TeV, ATLAS and CMS have collectively searched the main leptoquark production mechanisms—pair production, single production, and nonresonant dilepton exchange—and find no significant deviation from the Standard Model. For each benchmark model with a single mass $m_{LQ}$, a coupling $\lambda$, a branching fraction $\beta$, and for vector leptoquarks a non-minimal coupling $\kappa$, the collaborations set 95% confidence-level upper limits. Pair-produced scalar and vector leptoquarks are excluded up to approximately 2 TeV depending on the final state, and nonresonant dielectron, dimuon, and ditau searches exclude couplings above about 0.5 to 2.5 for masses up to and beyond 5 TeV. The ditau search also reports that it is now possible to begin probing the phase space favored by the B anomalies at 90% confidence.

Load-bearing premise

The reported limits hold only if a real leptoquark's production and decay are captured by the simplified benchmark models that assume a single leptoquark mass, one coupling to fermions, one branching fraction, and for vector leptoquarks a single non-minimal coupling; if the true particle couples across generations or interferes differently, the exclusions would not apply.

Editorial extensions

If this is right

If no leptoquark exists below the excluded range, explanations of the B anomalies that require light leptoquarks with large couplings are ruled out or pushed to the edge of the probed region.
Combining charged-lepton and neutrino final states lets the collaborations set limits as a function of the branching fraction beta, rather than only at beta equal to one or beta equal to zero.
At high mass, nonresonant searches behave as four-fermion interaction constraints, so their limits are largely independent of the leptoquark mass and coupling details.
The remaining uncovered final states—electrons or muons accompanied by light- or charm-flavored jets—leave room for a leptoquark that couples mainly to first- or second-generation fermions.

Reading between the lines

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

If a real leptoquark has several non-zero couplings across generations, the single-coupling benchmark limits could be misleading; a systematic reinterpretation of the ATLAS and CMS likelihoods in multi-coupling models is the natural next step.
The template-fitting technique used for nonresonant dilepton searches, using angular distributions in rapidity and Collins-Soper angle, could be applied to other beyond-Standard-Model four-fermion interactions such as Z-prime bosons or new contact interactions.
Because the new ditau analysis is the first nonresonant search in that channel to include Drell-Yan interference, earlier limits without interference may have been slightly different, and the size of this effect could be quantified by rerunning previous fits with the interference term included.
With Run-3 luminosity, the same search strategies will either reveal a signal in the B-anomaly-favored region or exclude most of it.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

Lena,

This is a conference proceedings review, not a research paper. It claims nothing new: it summarizes published ATLAS and CMS Run-2 leptoquark searches, highlights three recent analyses (one of which, the nonresonant ditau search, is presented for the first time at this conference), and packages the field into convenient tables. If you need a quick orientation to current LQ bounds, this is a fine place to start.

What the paper does well is the organizational work. Tables 1–4 map searches by production mode and final state in a way I haven't seen collected elsewhere, and the highlight sections are faithful to the original papers. The caveat that exclusions depend on simplified-model parameters (mLQ, lambda, beta, kappa) is stated up front, and the conclusion—no signal seen, pair limits around 2 TeV, nonresonant limits beyond 5 TeV at large couplings—matches the cited results. I checked the quoted numbers against the highlighted analyses as best I could from the text and they are consistent.

The soft spots are minor. There is a typo in Section 2 where the benchmark list says "beta = 1 (a mix)" instead of something like "0 < beta < 1", and the abstract has a subject-verb agreement slip. More substantively, the paper spends almost no space on single leptoquark production, giving no specific limits there, and it doesn't probe very deeply into how the simplified-model assumptions (e.g., a single LQ species, one coupling, negligible interference) affect the applicability of the limits. For a proceedings overview that is acceptable, but a reader should keep it in mind.

The stress-test note has it right: there is no load-bearing internal flaw. The review's value is as a compendium, not as a scientific advance. I would cite it as a convenient summary of the current LQ landscape, and I would send it to a referee if it were submitted to a journal—though the referee's job would be mainly to check faithfulness to the cited papers, which looks good. Bottom line: worth having on the shelf, not worth heavy scrutiny.

Referee Report

0 major / 8 minor

Summary. This paper is a conference proceedings overview of Run-2 leptoquark (LQ) searches by ATLAS and CMS at sqrt(s)=13 TeV with 137-140 fb^-1 of data. It introduces the simplified-model parameters used in LHC searches (mLQ, lambda, beta, kappa), catalogs pair-production, single-production, and nonresonant t-channel dilepton searches in Tables 1-4, and highlights three recent results: the ATLAS search for LQ pair production decaying to cc nu nu, excluding scalar (vector) LQ masses up to 0.9 (1.5) TeV; the CMS nonresonant dielectron/dimuon search setting exclusion limits up to about 5 TeV at large couplings; and the ATLAS nonresonant ditau search excluding mLQ > 3 TeV for large couplings. The paper reports no significant excess and concludes that no direct evidence for leptoquarks has been observed.

Significance. As a proceedings summary, the paper serves a useful archival and organizational purpose: it collects the current LHC leptoquark search landscape in compact tables and directs the reader to the relevant public analyses. Its strengths are the explicit statement of simplified-model assumptions in Section 2, the coverage of recent nonresonant searches, and the traceability of the quoted limits to Refs. 11-13. The paper makes no new derivation and does not re-analyze data, which is appropriate for this format. The stress-test concern that real leptoquarks might not be captured by the simplified benchmark models is explicitly acknowledged in Section 2, so it does not undermine the paper's claims; the reported exclusions are correctly framed as benchmark-dependent.

minor comments (8)

[Section 2 (bullet list)] In the benchmark bullet list, the entry 'beta = 1 (a mix)' should read '0 < beta < 1 (a mix)'; as written, the same value beta = 1 is assigned to both the no-neutrino and mixed scenarios.
[Section 3.1] The sentence 'Although that search could distinguish between jet flavors of a potential signal, it set more stringent exclusion limits' appears to be missing a negation: since the CMS search in Ref. 14 did not exploit charm tagging, the intended meaning is likely 'could not distinguish between jet flavors.'
[Section 3.1] The phrase 'The LQ signal is target by selecting events' should be 'The LQ signal is targeted by selecting events.'
[Section 3.2] The phrase 'contribute most at in the high-m_ll tail' contains an extra 'at'; it should read 'contribute most in the high-m_ll tail.'
[Section 2] The note 'Ref. 8 is currently only search the photon-induced single production' should be rephrased, for example 'Ref. 8 currently searches only for photon-induced single production.'
[Section 4] The conclusion's claim that pair-production searches reach 'up to about 2 TeV or beyond' is not explicitly tied to a specific cited result in the text; please add a reference to Refs. 9/10 or to a representative pair-production search.
[Abstract] The sentence 'An overview of searches for leptoquarks in proton-proton collisions at ATLAS and CMS are presented' has a subject-verb agreement error; it should be 'is presented.'
[Title/header] The title in the manuscript header contains a spacing artifact, 'A TLAS'; it should read 'ATLAS.'

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: review summarizes external ATLAS/CMS Run-2 searches without deriving new claims.

full rationale

This paper is a conference-proceedings review that compiles and reports ATLAS and CMS Run-2 leptoquark search results. It contains no derivation chain and no fitted parameters of its own; every quantitative claim is attributed to the cited experimental papers, and the figures are explicitly taken from those papers. The central conclusion, that no leptoquark signal has been observed and that mass/coupling limits reach roughly 2 TeV for pair production and beyond 5 TeV for nonresonant production at high coupling, is a summary of externally produced limits. The simplified-model parameters listed in Section 2 are described as assumptions of the searches rather than as outputs, so the benchmark-dependence of the limits is an explicit caveat rather than a hidden circular step. The author's affiliation with the collaborations is expected for a collaboration summary and does not turn reported experimental results into self-referential evidence. No self-definitional construction, fitted-input-called-prediction pattern, or load-bearing self-citation is present. The review is therefore self-contained as a summary and exhibits no significant circularity.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

This is a review, so the paper introduces no fitted parameters and no new entities. The central summary rests on assumptions about the adequacy of simplified LQ models, the correctness of the underlying ATLAS and CMS analyses, and the reliability of the experimental data. These assumptions are inherited from the cited papers.

free parameters (4)

mLQ (leptoquark mass) = scanned, e.g. 0.2-10 TeV
The mass is scanned in each search to produce exclusion limits; it is not fitted in this review but is an axis of the reported limits.
λ (or g) leptoquark coupling = scanned or fitted, e.g. up to 4
Coupling strength to quark-lepton pairs; used as the fitted parameter in the nonresonant searches (Refs. 12, 13). The review reports limits as a function of λ.
β (branching fraction to charged leptons) = benchmark values 0, 0.5, 1
Branching fraction to charged lepton final states; chosen as benchmarks in the simplified models.
κ (non-minimal coupling for vector LQs) = 0 (minimal) or 1 (Yang-Mills)
Parameter for vector leptoquark couplings to gluons; benchmark values used in the searches.

assumptions (3)

domain assumption The simplified leptoquark models (single mass mLQ, coupling λ, branching fraction β, and vector coupling κ) adequately describe the phenomenology needed to interpret search results.
Section 2 explicitly defines the model parameters and says LHC searches typically use these simplified models. If the real LQ has generation-dependent couplings or other production modes, the limits may not apply.
domain assumption The Standard Model background estimates and systematic uncertainties in the cited ATLAS and CMS analyses are correct.
The conclusion of 'no significant excess' is entirely dependent on these background predictions, which are not re-derived in this review.
domain assumption The integrated luminosity and detector performance reported by ATLAS and CMS are accurate.
The paper quotes 137-140 fb^-1 and relies on the collaborations' calibrations.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Leptoquark searches at ATLAS and CMS." pith.science (2026). https://pith.science/paper/CDLYVOKQ

@misc{pith2026250508738,
  author       = {Pith},
  title        = {Pith review of: Leptoquark searches at ATLAS and CMS},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CDLYVOKQ}},
  note         = {Machine review of arXiv:2505.08738}
}

abstract

An overview of searches for leptoquarks in proton-proton collisions at ATLAS and CMS are presented, with a focus on results from the Run-2 dataset collected between 2016 and 2018 at $\sqrt{s}=13\,\mathrm{TeV}$, and corresponding to an integrated luminosity of $137{-}140\,\mathrm{fb}^{-1}$. In particular, recent results are highlighted.

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

Exploring Fermionic Dark Matter in the Presence of Scalar Leptoquarks
hep-ph 2025-09 conditional novelty 6.0 of 10

A vector-like lepton dark matter model with two inert scalar leptoquarks produces a radiative pseudo-Dirac mass splitting that evades direct detection and opens up parameter space for the correct relic density.

Reference graph

Works this paper leans on

36 extracted references · 31 canonical work pages · cited by 1 Pith paper

[1]

C. Marin. These proceedings

work page
[2]

L. Eklund. These proceedings

work page
[3]

M. Reboud. These proceedings

work page
[4]

Martinov

T. Martinov. These proceedings

work page
[5]

Muon g 2 Collaboration. Phys. Rev. Lett. , 131(16):161802, 2023

work page 2023
[6]

JINST, 3:S08003, 2008

ATLAS Collaboration. JINST, 3:S08003, 2008

2008
[7]

JINST, 3:S08004, 2008

CMS Collaboration. JINST, 3:S08004, 2008

2008
[8]

CMS Collaboration. Phys. Rev. Lett. , 132(6):061801, 2024

work page 2024

Show all 36 references

[9]

ATL-PHYS-PUB-2025-013, 2025

ATLAS Collaboration. ATL-PHYS-PUB-2025-013, 2025

2025
[10]

https://twiki.cern.ch/twiki/bin/view/CMSPublic/ SummaryPlotsEXO13TeV#Leptoquark_summary_plot (Accessed May 7, 2025)

CMS Collaboration. https://twiki.cern.ch/twiki/bin/view/CMSPublic/ SummaryPlotsEXO13TeV#Leptoquark_summary_plot (Accessed May 7, 2025)

2025
[11]

JHEP, 02:193, 2025

ATLAS Collaboration. JHEP, 02:193, 2025

2025
[12]

CMS-EXO-22-013, arXiv:2503.20023, 2025

CMS Collaboration. CMS-EXO-22-013, arXiv:2503.20023, 2025

2025
[13]

EXOT-2022-42, arXiv:2503.19836, 2025

ATLAS Collaboration. EXOT-2022-42, arXiv:2503.19836, 2025

2022
[14]

CMS Collaboration. Eur. Phys. J. C , 80(1):3, 2020

2020
[15]

JHEP, 05:093, 2021

ATLAS Collaboration. JHEP, 05:093, 2021

2021
[16]

ATLAS Collaboration. Eur. Phys. J. C , 80(8):737, 2020

2020
[17]

JHEP, 10:112, 2020

ATLAS Collaboration. JHEP, 10:112, 2020

2020
[18]

ATLAS Collaboration. Eur. Phys. J. C , 81(4):313, 2021

2021
[19]

ATLAS Collaboration. Eur. Phys. J. C , 84(8):818, 2024

2024
[20]

CMS Collaboration. Phys. Rev. D , 105(11):112007, 2022

2022
[21]

CMS Collaboration. Phys. Rev. D , 109(11):112003, 2024

2024
[22]

JHEP, 06:199, 2023

ATLAS Collaboration. JHEP, 06:199, 2023

2023
[23]

ATLAS Collaboration. Phys. Rev. D , 104(11):112005, 2021

2021
[24]

ATLAS Collaboration. Eur. Phys. J. C , 83(11):1075, 2023

2023
[25]

JHEP, 05:311, 2024

CMS Collaboration. JHEP, 05:311, 2024

2024
[26]

JHEP, 06:179, 2021

ATLAS Collaboration. JHEP, 06:179, 2021

2021
[27]

JHEP, 2306:188, 2023

ATLAS Collaboration. JHEP, 2306:188, 2023

2023
[28]

CMS Collaboration. Phys. Lett. B , 819:136446, 2021

2021
[29]

JHEP, 11:153, 2021

CMS Collaboration. JHEP, 11:153, 2021

2021
[30]

ATLAS Collaboration. Phys. Lett. B , 830:137106, 2022

2022
[31]

ATLAS Collaboration. Phys. Rev. D , 110(1):012014, 2024

2024
[32]

CMS-TOP-22-011, arXiv:2504.08532, 2025

CMS Collaboration. CMS-TOP-22-011, arXiv:2504.08532, 2025

2025
[33]

JHEP, 10:001, 2023

ATLAS Collaboration. JHEP, 10:001, 2023

2023
[34]

JHEP, 07:067, 2022

CMS Collaboration. JHEP, 07:067, 2022

2022
[35]

JHEP, 07:073, 2023

CMS Collaboration. JHEP, 07:073, 2023

2023
[36]

JHEP, 2022(08):063, 2022

CMS Collaboration. JHEP, 2022(08):063, 2022

2022

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

C. Marin. These proceedings

work page

[2] [2]

L. Eklund. These proceedings

work page

[3] [3]

M. Reboud. These proceedings

work page

[4] [4]

Martinov

T. Martinov. These proceedings

work page

[5] [5]

Muon g 2 Collaboration. Phys. Rev. Lett. , 131(16):161802, 2023

work page 2023

[6] [6]

JINST, 3:S08003, 2008

ATLAS Collaboration. JINST, 3:S08003, 2008

2008

[7] [7]

JINST, 3:S08004, 2008

CMS Collaboration. JINST, 3:S08004, 2008

2008

[8] [8]

CMS Collaboration. Phys. Rev. Lett. , 132(6):061801, 2024

work page 2024