Search for heavy top quark partners with charge 5/3 in the single-lepton final state in proton-proton collisions at sqrt s = 13 TeV

arXiv:2611.14335 · hep-ex · Submitted 2026-11-10 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: High Energy Physics: Experiment. The day in high-energy experiment: detectors, searches and limits.

Kai: We're looking at the paper: "Search for heavy top quark partners with charge 5/3 in the single-lepton final state in proton-proton collisions at sqrt s = 13 TeV".

Mira: Here's the gist of the paper. A search for heavy top quark partners with charge 5/3 in the single-lepton final state in proton-proton collisions at sqrt s = 13 TeV is presented The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV Search Scope and Decay Modes The search presents a pair production of a vector-like quark, X5/3, which is a heavy fermionic partner of the top quark with an electric charge of 5/3.

Kai: Two decay modes are considered: the conventional decay mode where X5/3 is assumed to decay exclusively into a top quark and a W boson, and an alternative exotic scenario where it decays exclusively into a top quark and a charged Higgs boson, which subsequently decays exclusively into top and bottom quarks.

Mira: Data Selection Criteria The analysis targets events with one electron or muon in the final state, in which one W boson decays leptonically and the remaining W bosons decay hadronically.

Kai: The final state is characterized by one isolated charged lepton, l, either an electron or a muon, missing transverse momentum (⃗p T) from the neutrino, and multiple hadronically decaying top quarks and W bosons.

Paper summary: Kai: So, the title of this paper is "Search for heavy top quark partners with charge five over three in the single-lepton final state in proton-proton collisions at sqrt s equals thirteen tera electron volt" <ref:2611.14335#pg1>. It’s a very direct statement about what they were trying to achieve.

Mira: It essentially summarizes the whole effort: searching for this Xfive/three particle, which is a heavy fermion partner of the top quark with that specific charge of five over three <ref:2611.14335#pg2>.

Lev: The main implication is that they’ve set upper limits on the pair production cross section of this Xfive/three at ninety five percent confidence level <ref:2611.14335#pg2>. They found no excess of data events over the standard model expectation <ref:2611.14335#pg2>.

Kai: For someone listening, it means that as of this search, we haven't seen any evidence for these heavy top quark partners in either decay mode within the mass ranges they tested <ref:2611.14335#pg2>.

Mira: The paper is setting constraints on the theory predicting these vector-like quarks and their potential role in stabilizing the Higgs mass via little Higgs models <ref:2611.14335#pg3>. They are constraining those theoretical models based on what's seen experimentally.

Lev: It tells us what experimental sensitivity looks like for finding something this heavy, especially when you have to deal with the backgrounds from top top production <ref:2611.14335#pg2>. It sets a benchmark for future searches at higher energies or with better luminosity <ref:2611.14335#pg2>.

Conclusion: Kai: So, we’re wrapping up this search for those heavy top partners with charge five over three <ref:2611.14335#pg2>.

Mira: That paper is focused on looking for these particles in proton-proton collisions at thirteen tera electron volt energy levels.

Lev: It sets limits on the pair production cross section, which tells us how often we expect to see them in those collisions.

Kai: The authors are testing two main decay scenarios, one where the partner decays into a top quark and a W boson, and another more exotic way involving a charged Higgs boson.

Mira: They set upper limits on these Xfive/three masses up to one thousand six hundred thirty giga electron volt for the first decay mode.

Lev: And for the second decay mode, they pushed that limit up to one thousand seven hundred forty giga electron volt, especially when the charged Higgs boson mass is around eight hundred giga electron volt.

Kai: The overall conclusion is that they didn't find any significant excess of data events over what we expect from the standard model.

Mira: They found good agreement between their data and their background predictions, which is important because it means their method for separating the signal from the noise seems solid.

Lev: But remember, this entire search is heavily limited by systematic uncertainties in how they model things like top top background and initial state radiation.

Kai: So, while no new particle was seen yet, these limits are still really useful for constraining theoretical models that predict these heavy partners.

Mira: And understanding these limits helps us know what kind of physics we need to see at future colliders to find something like this.

CMS Collaboration

European Organization for Nuclear Research (CERN)

hep-ex

Submitted: 2026-11-10

Updated: 2026-11-10

Comments: Submitted to Physical Review D. All figures and tables can be found at http://cms-results.web.cern.ch/cms-results/public-results/publications/B2G-24-005 (CMS Public Pages)

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 71/100

The gist: A search for heavy top quark partners with charge 5/3 in the single-lepton final state in proton-proton collisions at sqrt s = 13 TeV is presented The observed exclusion ranges of the X5/3 mass…

Key concepts

X5/3
This is a hypothetical heavy fermionic partner of the top quark with an electric charge of 5/3. The search looks for evidence of this particle decaying into known particles like top quarks and W bosons or a top quark and a charged Higgs boson.
Conventional Decay Mode
This scenario assumes the X5/3 decays exclusively into a standard top quark and a W boson. The analysis focuses on events with one isolated lepton, one neutrino (from the W), and multiple jets from the hadronic decay of the other W boson.
Exotic Decay Mode
This alternative scenario posits that X5/3 decays into a top quark and a charged Higgs boson. This charged Higgs then decays into top and bottom quarks, leading to a different final state signature that is also being searched for in the data.

Terminology

Summary

A search for heavy top quark partners with charge 5/3 in the single-lepton final state in proton-proton collisions at sqrt s = 13 TeV is presented The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV

Search Scope and Decay Modes

The search presents a pair production of a vector-like quark, X5/3, which is a heavy fermionic partner of the top quark with an electric charge of 5/3. Two decay modes are considered: the conventional decay mode where X5/3 is assumed to decay exclusively into a top quark and a W boson, and an alternative exotic scenario where it decays exclusively into a top quark and a charged Higgs boson, which subsequently decays exclusively into top and bottom quarks. p. 0

Where the paper says it

In the conventional decay mode, the X5/3 is assumed to decay exclusively into a top quark and a W boson. In an alternative, exotic scenario, it is assumed to decay exclusively into a top quark and a charged Higgs boson, assumed to subsequently decay exclusively into top and bottom quarks.

Data Selection Criteria

The analysis targets events with one electron or muon in the final state, in which one W boson decays leptonically and the remaining W bosons decay hadronically. The final state is characterized by one isolated charged lepton, l, either an electron or a muon, missing transverse momentum (⃗p T) from the neutrino, and multiple hadronically decaying top quarks and W bosons. p. 3

Where the paper says it

This channel provides high signal acceptance and background rejection. The final state is characterized by one isolated charged lepton, ℓ, either an electron or a muon, missing transverse momentum (⃗p miss T ) from the neutrino, and multiple hadronically decaying top quarks and W bosons, including several jets originating from bottom quarks.

Object Reconstruction

The particle-flow (PF) algorithm is used to reconstruct and identify each individual particle in an event using information from the various elements of the CMS detector. Events are required to have at least one PV within the luminous region, defined by a position along the beamline axis of less than 24. p. 5

Where the paper says it

The PV is reconstructed from all tracks in the event that are compatible with the LHC beam spot in the transverse plane. Events are required to have at least one PV within the luminous region, defined by a position along the beamline axis of less than 24.

0 cm and a radial distance from the beamline in the transverse plane of less than 2. p. 5
Where the paper says it

Events are required to have at least one PV within the luminous region, defined by a position along the beamline axis of less than 24. 0 cm and a radial distance from the beamline in the transverse plane of less than 2.

0 cm, and with the number of degrees of freedom greater than 4. p. 5
Where the paper says it

0 cm and a radial distance from the beamline in the transverse plane of less than 2. 0 cm, and with the number of degrees of freedom greater than 4.

Signal Discrimination

A multivariate discriminant based on XGBOOST is employed to enhance signal-to-background separation. The BDT is trained separately for the conventional and exotic decay modes, and independently for each data set corresponding to a different data-taking period. p. 8

Where the paper says it

A boosted decision tree (BDT) is trained using simulated events for X5/3 signals with masses of 1200 GeV in the exotic decay mode and 1300 GeV in the conventional decay mode, together with tt background events, after applying the final selection criteria. The BDT is trained separately for the conventional and exotic decay modes, and independently for each data set corresponding to a different data-taking period, to account for variations in detector conditions and data quality that may affect the performance of the discriminant.

p. 8
Where the paper says it

A boosted decision tree (BDT) is trained using simulated events for X5/3 signals with masses of 1200 GeV in the exotic decay mode and 1300 GeV in the conventional decay mode, together with tt background events, after applying the final selection criteria. The BDT is trained separately for the conventional and exotic decay modes, and independently for each data set corresponding to a different data-taking period, to account for variations in detector conditions and data quality that may affect the performance of the discriminant.

The BDT provides effective discrimination for other X5/3 signal mass points not used during training, with the area under the receiver operating characteristic curve reaching at least 0. p. 8
Where the paper says it

The BDT is trained separately for the conventional and exotic decay modes, and independently for each data set corresponding to a different data-taking period, to account for variations in detector conditions and data quality that may affect the performance of the discriminant. The BDT provides effective discrimination for other X5/3 signal mass points not used during training, with the area under the receiver operating characteristic curve reaching at least 0.

9 for masses above 1000 GeV. p. 8
Where the paper says it

The BDT provides effective discrimination for other X5/3 signal mass points not used during training, with the area under the receiver operating characteristic curve reaching at least 0. 9 for masses above 1000 GeV.

Results and Limits

Upper limits on the X5/3 pair production cross section are set using a binned likelihood fit to the distributions of the XGBOOST output in the signal region. The X5/3 masses up to 1630 GeV have been excluded at 95% CL in the decay mode to a top quark and a W boson. p. 0

Where the paper says it

The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV. These are the most stringent limits on the X5/3 decaying to a top quark and a W boson, and the first limits on the X5/3 decay into a top quark and a charged Higgs boson.

In the decay mode to a top quark and a charged Higgs boson, X5/3 masses up to 1740 GeV are excluded at 95% CL, with the strongest exclusion obtained for an H± mass of 800 GeV. p. 0
Where the paper says it

The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV. These are the most stringent limits on the X5/3 decaying to a top quark and a W boson, and the first limits on the X5/3 decay into a top quark and a charged Higgs boson.

Systematic Uncertainties

The systematic uncertainties listed in Table 4 are considered from several sources that affect the normalization and shape of the background or signal distributions. The dominant systematic uncertainties in the conventional decay mode are associated with tt background normalization, FSR and ISR modeling, and the hdamp parameter choice. p. 9

Where the paper says it

The input variables are validated through studies of data-to-simulation agreement and checks of intervariable correlations. In the conventional decay mode, variables related to W bosons and W-tagged jets are ranked most highly, while in the exotic decay mode, variables associated with bottom quarks and b-tagged jets are dominant.

The analysis is primarily limited by systematic uncertainties, while statistical uncertainties have a negligible impact (<1%) on the expected limits. p. 3
Where the paper says it

Section 4 describes the reconstruction of physics objects, while Section 5 details the event selection. Section 6 presents the systematic uncertainties, followed by Section 7 that describes the results.

p. 3
Where the paper says it

Section 4 describes the reconstruction of physics objects, while Section 5 details the event selection. Section 6 presents the systematic uncertainties, followed by Section 7 that describes the results.

p. 3
Where the paper says it

Section 4 describes the reconstruction of physics objects, while Section 5 details the event selection. Section 6 presents the systematic uncertainties, followed by Section 7 that describes the results.

Conclusion

No significant excess of data events over the SM expectation is observed. The results are obtained using the CMS statistical analysis tool COMBINE. p. 3

Where the paper says it

Tabulated results are available in the HEPData record [12] for this analysis. 2 The CMS detector The CMS apparatus [13, 14] is a multipurpose, nearly hermetic detector designed to trigger on [15–17] and identify electrons, photons, muons, and charged and neutral hadrons [18–20].

Good agreement between the data and the background prediction is also observed. p. 4
Where the paper says it

Simulated events are reweighted to reproduce the mean number of interactions per bunch crossing observed in data. Background processes are grouped into several categories.

The sensitivity of the analysis is driven by the high-score BDT region. p. 8
Where the paper says it

The requirement was determined through optimization studies and was found to improve the expected sensitivity of the analysis by preserving a high signal efficiency. A signal-depleted control region is defined by inverting this requirement, ∆R(ℓ, j2 ) < 1.

The analysis is primarily limited by systematic uncertainties, while statistical uncertainties have a negligible impact (<1%) on the expected limits. The residual mismodeling in the tt background, including the known overprediction at high ST, is constrained by the simultaneous fit to data through the relevant nuisance parameters. p. 7
Where the paper says it

2 for the combined 2016–2018 data set. An overprediction of the yield is observed at high ST , primarily due to mismodeling of tt

The goodness-of-fit tests yield p-values in the range 0. p. 2
Where the paper says it

However, exotic decays of the X5/3 are motivated by extensions of the SM. These decays can proceed via X5/3 → H+t, where H+ denotes a charged Higgs boson whose mass is treated as a free parameter in the range 200–1000 GeV [7].

32–0.78, indicating good agreement. p. 9
Where the paper says it

Restricting the input to the 20 highest-ranked variables balances discrimination power with the ability to validate the input variable modeling, while avoiding the inclusion of variables with little additional discrimination power. The input variables are validated through studies of data-to-simulation agreement and checks of intervariable correlations.

The upper limits are compared with the theoretical X5/3 pair production cross section at NNLO, and the PDF and µR and µF scale uncertainties in the theoretical cross section are shown as a band around the theoretical prediction. p. 4
Where the paper says it

The X5/3 mass spacing of 100 GeV is consistent with that used in previous searches and provides sufficient granularity to capture variations in signal acceptance across the mass spectrum. The theoretical prediction for the X5/3 pair production cross section depends only on the X5/3 mass and is calculated at next-to-next-to-leading order (NNLO) using the TOP++ 2.

The observed limit lies below the expected limit and remains within the two standard deviation uncertainty band at low mass. p. 0
Where the paper says it

No excess of data events over the standard model (SM) expectation is observed. Expected and observed upper limits on the pair production cross section of the X5/3 are set at 95% confidence level as functions of the X5/3 mass.

The expected and observed limits on the X5/3 cross section exclude X5/3 masses in the exotic decay mode below 1620–1690 and 1620–1740 GeV, respectively, for H± masses between 200 and 1000 GeV. p. 0
Where the paper says it

Expected and observed upper limits on the pair production cross section of the X5/3 are set at 95% confidence level as functions of the X5/3 mass. The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV.

The interpretation in the X5/3 → tH+ decay chain is restricted to simulated mass points satisfying the kinematic condition that the X5/3 mass exceeds the sum of the charged Higgs boson and top quark masses, and no interpolation of signal efficiencies or templates between mass hypotheses is performed. p. 0
Where the paper says it

The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV. These are the most stringent limits on the X5/3 decaying to a top quark and a W boson, and the first limits on the X5/3 decay into a top quark and a charged Higgs boson.

The first limits on the X5/3 decaying into a top quark and a charged Higgs boson are set. p. 0
Where the paper says it

The observed exclusion ranges of the X5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV. These are the most stringent limits on the X5/3 decaying to a top quark and a W boson, and the first limits on the X5/3 decay into a top quark and a charged Higgs boson.

How the training signal is built

The Monte Carlo (MC) simulation of signal and background processes is performed using several matrix element generators. Signal events are generated for the right-handed coupling scenario, with a decay width below 1% of the X5/3 mass to ensure the validity of the NWA. p. 4

Where the paper says it

5 [21] generator. Signal events are generated for the right-handed coupling scenario, with a decay width below 1% of the X5/3 mass to ensure the validity of the NWA.

Two decay modes of the X5/3 are considered separately: conventional decay mode, where each X5/3 is pair produced and decays into a top quark and a W boson with a 100% branching fraction, and exotic decay mode, where an additional signal sample is generated at 1900 GeV to extend the high-mass coverage. p. 4
Where the paper says it

An additional signal sample is generated at 1900 GeV to extend the high-mass coverage of the analysis, particularly for the unexplored exotic decay mode. Each X5/3 is pair produced and decays into a top quark and an H+ with a 100% branching fraction [7].

Where the gains come from

The requirement ∆R(l, j2) > 1.0 retains approximately 90% of the signal in the conventional decay mode and 84–89% in the exotic decay mode, depending on the H± mass. p. 8

Where the paper says it

The requirement ∆R(ℓ, j2 ) > 1. 0 retains approximately 90% of the signal in the conventional decay mode and 84–89% in the exotic decay mode, depending on the H± mass.

The BDT is trained using simulated events for X5/3 signals with masses of 1200 GeV in the exotic decay mode and 1300 GeV in the conventional decay mode, together with tt background events after applying the final selection criteria. p. 8
Where the paper says it

To enhance signal-to-background separation, a multivariate discriminant based on XGBOOST, a gradient boosting library [52], is employed. A boosted decision tree (BDT) is trained using simulated events for X5/3 signals with masses of 1200 GeV in the exotic decay mode and 1300 GeV in the conventional decay mode, together with tt background events, after applying the final selection criteria.

The BDT is trained separately for the conventional and exotic decay modes, and independently for each data set corresponding to a different data-taking period. The input variables are validated through studies of data-to-simulation agreement and checks of intervariable correlations. p. 9
Where the paper says it

Restricting the input to the 20 highest-ranked variables balances discrimination power with the ability to validate the input variable modeling, while avoiding the inclusion of variables with little additional discrimination power. The input variables are validated through studies of data-to-simulation agreement and checks of intervariable correlations.

The conventional decay mode variables related to W bosons and W-tagged jets are ranked most highly, while in the exotic decay mode, variables associated with bottom quarks and b-tagged jets are dominant. p. 9
Where the paper says it

The input variables are validated through studies of data-to-simulation agreement and checks of intervariable correlations. In the conventional decay mode, variables related to W bosons and W-tagged jets are ranked most highly, while in the exotic decay mode, variables associated with bottom quarks and b-tagged jets are dominant.

The low-BDT-score region is excluded by requiring a BDT score greater than 0. p. 7
Where the paper says it

6 jet energy measurements in that region and can lead to spurious p miss T in the signal region. To mitigate this effect, jets within the affected region are excluded from the analysis.

1 to reduce potential biases from regions with limited discriminating power. p. 8
Where the paper says it

The signal is characterized by larger ST values than the background processes. This observable provides strong discriminating power and is a key component of the analysis strategy.

The signal is extracted from these XGBOOST distributions. p. 8
Where the paper says it

This control region is used for validation purposes only and is not included in the signal extraction. To enhance signal-to-background separation, a multivariate discriminant based on XGBOOST, a gradient boosting library [52], is employed.

The analysis is primarily limited by systematic uncertainties, while statistical uncertainties have a negligible impact (<1%) on the expected limits.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Precision limit setting for heavy top quark partners: The analysis can set 95% confidence level upper limits on X5/3 pair production cross section as functions of mass using a multivariate discriminant. This allows exclusion ranges up to 1630 GeV in the conventional decay mode and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV.

  2. Exotic decay channel discovery: The system can provide the first limits on X5/3 decaying into a top quark and an H+, with observed exclusion ranges up to 1740 GeV for an H+ mass of 800 GeV, which is the first limits on the X5/3 decaying into a top quark and an H+.

  3. Signal discrimination through angular separation: The system can utilize the angular separation variable Delta R (Delta eta squared plus Delta phi squared) between the selected lepton and subleading AK4 jet to probe event topology, showing a bimodal structure with signal events preferentially populating larger values.

  4. Multivariate analysis for mass point discrimination: A boosted decision tree (BDT) trained on simulated events can provide effective discrimination for other X5/3 signal mass points, with an area under the receiver operating characteristic curve reaching at least 0.9 for masses above 1000 GeV.

  5. Systematic uncertainty quantification: The system incorporates detailed systematic uncertainties, including integrated luminosity variations (up to 1.6%), jet energy scale and resolution uncertainties, and modeling of hard radiation in tt events, which are treated as nuisance parameters constrained by log-normal or Gaussian distributions.

Abstract

A search for the pair production of a vector-like quark, X 5/3, a heavy fermionic partner of the top quark with an electric charge of 5/3, is presented. The search is performed in the final state with one lepton, an electron or a muon. Proton-proton collision data collected by the CMS experiment at the LHC from 2016 to 2018, at a center-of-mass energy of 13 TeV, corresponding to an integrated luminosity of 138 fb-1, are used. Two decay modes are considered. In the conventional decay mode, the X 5/3 is assumed to decay exclusively into a top quark and a W boson. In an alternative, exotic scenario, it is assumed to decay exclusively into a top quark and a charged Higgs boson, assumed to subsequently decay exclusively into top and bottom quarks. No excess of data events over the standard model (SM) expectation is observed. Expected and observed upper limits on the pair production cross section of the X 5/3 are set at 95% confidence level as functions of the X 5/3 mass. The observed exclusion ranges of the X 5/3 mass extend up to 1630 GeV in the decay to SM particles, and up to 1740 GeV in the exotic decay mode for a charged Higgs boson mass of 800 GeV. These are the most stringent limits on the X 5/3 decaying to a top quark and a W boson, and the first limits on the X 5/3 decay into a top quark and a charged Higgs boson.

Sources

Related papers