Search for ttbar resonances in final states with exactly one or two leptons using 140 fb-1 of pp collision data at sqrt s =13 TeV with the ATLAS experiment

arXiv:2512.17856 · hep-ex · Submitted 2025-12-19 · 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: I'm Kai.

Mira: And I'm Mira, here with Lev.

Kai: We're looking at the paper: "Search for ttbar resonances in final states with exactly one or two leptons using 140 fb-1 of pp collision data at sqrt s =13 TeV with the ATLAS experiment".

Mira: Here's the gist of the paper. My task is to synthesize these fragments into a long, detailed summary that accurately reflects the scope, methodology, and conclusions of the research.

Kai: Here is the comprehensive summary: Detailed Research Summary: Searches for Heavy Spin-1 and Spin-2 Resonances Decaying into Top-Antitop Quark Pairs (t) at sqrt s = 13 TeV using ATLAS Data This research presents a comprehensive search for heavy resonances, specifically those with spin-1 and spin-2 quantum numbers, that decay into a top quark and an antitop quark pair (t).

Mira: The investigation utilizes high-luminosity data collected by the ATLAS experiment at the Large Hadron Collider (LHC), specifically 140 fb-1 of proton–proton collision data recorded in 2015–2018 at a center-of-mass energy (sqrt s) of 13 TeV.

Kai: Search Methodology and Final States The analysis focuses on searching across a broad invariant mass spectrum (m t) ranging from 0.4 TeV to 5.0 TeV, encompassing both narrow and broad resonance widths relative to detector resolution.

Paper summary: Mira: So what does this all mean for the particle physics community? The paper is titled "Search for ttbar resonances in final states with exactly one or two leptons using one hundred forty fb-one of pp collision data at sqrt s =thirteen TeV with the ATLAS experiment" <ref:2512.17856#pg1,resonances in final states with exactly one or two leptons using 140>.

Lev: It means that based on this data, we have set some of the most powerful exclusion limits currently available for heavy spin-one and spin-two particles decaying into top quark pairs <ref:2512.17856#pg3>.

Kai: Essentially, they are telling us where these specific types of heavy resonances are not. They are constraining the parameter space for Z' bosons, Kaluza–Klein gravitons, and KK gluons that couple strongly to the top quark <ref:2512.17856#pg1>.

Mira: The implication is that if these particles exist within certain mass ranges, they must have couplings smaller than what was excluded here. It tells us what kind of new physics we should expect to see in future experiments at the LHC or even at other colliders.

Lev: For someone looking at how this translates to building real hardware, it shows the sensitivity of these searches when applied to current data sets and the necessary background modeling needed for those kinds of analyses <ref:2512.17856#pg2>.

Kai: The authors explicitly state that they are targeting resonances with both narrow and broad widths relative to the detector resolution, covering a mass range from zero point four TeV to five point zero TeV <ref:2512.17856#pg1>.

Mira: They also noted that they used data-driven corrections for the W + jets and Z + jets backgrounds, which is a crucial part of how these limits are derived <ref:2512.17856#pg2>.

Lev: The limitation mentioned in the paper is that they focused on specific spin-one and spin-two models, so they aren't constraining other types of heavy particles, like scalar resonances, which is something that needs further investigation <ref:2512.17856#pg3>.

Kai: So to wrap up the paper "Search for ttbar resonances in final states with exactly one or two leptons using one hundred forty fb-one of pp collision data at sqrt s =thirteen TeV with the ATLAS experiment", it’s a very thorough check on these specific new physics signatures <ref:2512.17856#pg1,resonances in final states with exactly one or two leptons using 140>.

Mira: It confirms that within the mass range they looked at, no strong signal was found, which tightens the constraints on those models significantly <ref:2512.17856#pg3>.

Lev: The work provides a solid baseline for what to expect in future searches and helps guide where we should focus our experimental resources when we look for evidence of physics beyond the Standard Model <ref:2512.17856#pg3>.

Conclusion: Kai: So, we're wrapping up this look at the paper titled "Search for ttbar resonances in final states with exactly one or two leptons using one hundred forty fb-one of pp collision data at sqrt s =thirteen TeV with the ATLAS experiment".

Mira: This study basically checks if there are any heavy particles, specifically spin-one and spin-two stuff, that might be decaying into a top quark pair.

Lev: The authors used one hundred forty femtobarns of data from the ATLAS experiment at fifteen teravolt collision energy.

Kai: They focused on specific final states where you see exactly one or two leptons.

Mira: They were looking for these resonances across a wide invariant mass spectrum, from about four hundred gigaelectron volts up to five thousand gigaelectron volts.

Lev: The selection criteria were pretty strict, requiring at least one jet that looked like a bottom quark decay and some missing transverse momentum.

Kai: The main takeaway is that they found no significant deviations from what the Standard Model predicts across the whole mass range they checked.

Mira: That means for spin-one and spin-two models in this mass window, the constraints on their production cross section times branching ratio are quite strong.

Lev: Specifically, it excludes certain Z' bosons with a relative width of about three percent up to a mass near four point two teravolt.

Kai: And they also set very tight limits on broad resonances like the KK gluon with a thirty percent width, ruling out masses below four point one teravolt.

Mira: The strongest exclusion they found was actually for Kaluza–Klein graviton signals, setting limits below one point three teravolt.

Lev: It shows how sensitive these searches are when you use data-driven corrections to model those backgrounds, like the W plus jets stuff in the one-lepton channel.

Kai: So, if these particles exist in that mass range and with those specific spins, they have to be very weakly coupled.

Mira: It’s a really solid set of constraints that helps narrow down where we might expect to see new physics at future runs or at other colliders.

Lev: We can use this baseline data to figure out exactly what kind of signal would need to be seen in the next generation of experiments to start seeing something.

Kai: It’s all about pushing those limits and seeing if we find anything that breaks the Standard Model assumptions we're currently working with.

Mira: Next time, we'll look at how this translates into real-world physics constraints on other types of particles that might couple to the top quark.

EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH (CERN)

hep-ex

Submitted: 2025-12-19

Updated: 2025-12-19

Comments: 61 pages in total, author list starting page 46, 10 figures, 6 tables, submitted to JHEP

Journal ref: JHEP 10 (2026) 219

DOI: 10.1007/JHEP10(2026)219

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

Importance score: 88/100

The gist: My task is to synthesize these fragments into a long, detailed summary that accurately reflects the scope, methodology, and conclusions of the research.

Key concepts

Spin-1 and Spin-2 Resonances
These are hypothetical heavy particles predicted by some theories that have a spin of one or two. The study looks for evidence of these particles decaying into a top quark and an antitop quark pair, which is the specific signal being sought in this experiment.
Invariant Mass ($m_{tar{t}}$)
This is a calculated mass value derived from the energy and momentum of the detected top quark pair. By plotting these values, researchers can look for 'bumps' or peaks that might indicate a new, heavy resonance particle decaying into those two quarks.
Data-Driven Corrections
When modeling background noise from known Standard Model processes, researchers use Monte Carlo simulations. However, they apply data-driven corrections to these simulations to ensure the predicted background accurately reflects what is actually happening in the real experimental data.

Terminology

Summary

My task is to synthesize these fragments into a long, detailed summary that accurately reflects the scope, methodology, and conclusions of the research. p. 0

Where the paper says it

EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH (CERN) Submitted to: JHEP CERN-EP-2025-281 22nd December 2025 Search for 𝒕𝒕¯ resonances in final states with exactly one or two leptons using 140 fb−1 of 𝒑 𝒑 collision data at √ 𝒔 = 13 TeV with the ATLAS experiment The ATLAS Collaboration A search for heavy spin-1 and spin-2 resonances decaying into a top-antitop-quark pair has been performed with 140 fb−1 of proton–proton collision data collected by the ATLAS experiment at the Large Hadron Collider at a centre-of-mass energy of √ 𝑠 = 13 TeV. Final states with either exactly one electron or muon, or exactly two leptons (𝑒𝑒, 𝜇𝜇 or 𝑒𝜇), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered.

Here is the comprehensive summary:


Detailed Research Summary: Searches for Heavy Spin-1 and Spin-2 Resonances Decaying into Top-Antitop Quark Pairs (t t bar) at sqrt s = 13 TeV using ATLAS Data

This research presents a comprehensive search for heavy resonances, specifically those with spin-1 and spin-2 quantum numbers, that decay into a top quark and an antitop quark pair (t t bar). The investigation utilizes high-luminosity data collected by the ATLAS experiment at the Large Hadron Collider (LHC), specifically 140 fb-1 of proton–proton collision data recorded in 2015–2018 at a center-of-mass energy (sqrt s) of 13 TeV. p. 0

Where the paper says it

EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH (CERN) Submitted to: JHEP CERN-EP-2025-281 22nd December 2025 Search for 𝒕𝒕¯ resonances in final states with exactly one or two leptons using 140 fb−1 of 𝒑 𝒑 collision data at √ 𝒔 = 13 TeV with the ATLAS experiment The ATLAS Collaboration A search for heavy spin-1 and spin-2 resonances decaying into a top-antitop-quark pair has been performed with 140 fb−1 of proton–proton collision data collected by the ATLAS experiment at the Large Hadron Collider at a centre-of-mass energy of √ 𝑠 = 13 TeV. Final states with either exactly one electron or muon, or exactly two leptons (𝑒𝑒, 𝜇𝜇 or 𝑒𝜇), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered.

Search Methodology and Final States

The analysis focuses on searching across a broad invariant mass spectrum (m t t bar) ranging from 0.4 TeV to 5. p. 0

Where the paper says it

EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH (CERN) Submitted to: JHEP CERN-EP-2025-281 22nd December 2025 Search for 𝒕𝒕¯ resonances in final states with exactly one or two leptons using 140 fb−1 of 𝒑 𝒑 collision data at √ 𝒔 = 13 TeV with the ATLAS experiment The ATLAS Collaboration A search for heavy spin-1 and spin-2 resonances decaying into a top-antitop-quark pair has been performed with 140 fb−1 of proton–proton collision data collected by the ATLAS experiment at the Large Hadron Collider at a centre-of-mass energy of √ 𝑠 = 13 TeV. Final states with either exactly one electron or muon, or exactly two leptons (𝑒𝑒, 𝜇𝜇 or 𝑒𝜇), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered.

p. 0
Where the paper says it

EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH (CERN) Submitted to: JHEP CERN-EP-2025-281 22nd December 2025 Search for 𝒕𝒕¯ resonances in final states with exactly one or two leptons using 140 fb−1 of 𝒑 𝒑 collision data at √ 𝒔 = 13 TeV with the ATLAS experiment The ATLAS Collaboration A search for heavy spin-1 and spin-2 resonances decaying into a top-antitop-quark pair has been performed with 140 fb−1 of proton–proton collision data collected by the ATLAS experiment at the Large Hadron Collider at a centre-of-mass energy of √ 𝑠 = 13 TeV. Final states with either exactly one electron or muon, or exactly two leptons (𝑒𝑒, 𝜇𝜇 or 𝑒𝜇), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered.

0 TeV, encompassing both narrow and broad resonance widths relative to detector resolution. p. 0
Where the paper says it

Final states with either exactly one electron or muon, or exactly two leptons (𝑒𝑒, 𝜇𝜇 or 𝑒𝜇), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered. The search targets resonances with both narrow and broad widths relative to the detector resolution, and with masses in the range of 0.

The search employs two primary final state channels:

  1. Single-Lepton Channel: Events where exactly one top quark decays semi-leptonically and the other decays hadronically (1-lepton channel).

  2. Dilepton Channel: Events where both top quarks decay leptonically (ee, mu mu, or e mu final states, 2-lepton channel).

The selection criteria for these channels are stringent: they require the presence of exactly one or two leptons, large missing transverse momentum (p T miss), and at least one jet identified as likely originating from a b-hadron decay. p. 0

Where the paper says it

EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH (CERN) Submitted to: JHEP CERN-EP-2025-281 22nd December 2025 Search for 𝒕𝒕¯ resonances in final states with exactly one or two leptons using 140 fb−1 of 𝒑 𝒑 collision data at √ 𝒔 = 13 TeV with the ATLAS experiment The ATLAS Collaboration A search for heavy spin-1 and spin-2 resonances decaying into a top-antitop-quark pair has been performed with 140 fb−1 of proton–proton collision data collected by the ATLAS experiment at the Large Hadron Collider at a centre-of-mass energy of √ 𝑠 = 13 TeV. Final states with either exactly one electron or muon, or exactly two leptons (𝑒𝑒, 𝜇𝜇 or 𝑒𝜇), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered.

The search is designed to be sensitive to resonances where interference effects between the signal and the Standard Model (SM) background are negligible. p. 2
Where the paper says it

The results in this paper consider spin-1 and spin-2 signal models and are valid for the case when interference is negligible. It is the first time in Run 2 that the dilepton channel has been included in an ATLAS search for these types of resonances, and both the single-lepton and dilepton channels use the full Run-2 data for the first time.

A key methodological advancement in this study is the inclusion of the dilepton channel for these types of resonances, marking its first use in an ATLAS search for spin-1 and spin-2 models. Furthermore, the single-lepton channel analysis has been significantly enhanced compared to previous analyses (which utilized only 36 fb-1), incorporating improved t t bar reconstruction techniques and the addition of dedicated validation regions to refine background modeling. p. 2

Where the paper says it

It is the first time in Run 2 that the dilepton channel has been included in an ATLAS search for these types of resonances, and both the single-lepton and dilepton channels use the full Run-2 data for the first time. Additionally, the single-lepton channel improves on the previous 36 fb−1 analysis through improved 𝑡𝑡¯ reconstruction and the addition of validation regions to better evaluate the background modelling.

Signal Modeling and Background Processes

The primary signal models considered are those corresponding to hypothetical particles such as Z' bosons, Kaluza–Klein (KK) gravitons, and KK gluons, which are predicted to decay into top-quark pairs. The dominant background process in both the 1-lepton and 2-lepton channels is the Standard Model t t bar production. p. 7

Where the paper says it

An overview of the generator choices for the signal and background processes can be found in Table 1, with further details given in the following. The dominant background process is SM 𝑡𝑡¯ production.

To rigorously model this background, the analysis employs sophisticated techniques:

  • Monte Carlo (MC) Simulation: Most background components are modeled using MC simulation.

  • Data-Driven Corrections: Crucially, data-driven corrections are applied to the MC-based estimates for specific backgrounds: W + jets in the 1-lepton channel and Z + jets in the 2-lepton channel, to accurately normalize and shape these distributions.

  • Statistical Uncertainty: The statistical uncertainty arising from the estimation of fake backgrounds is also accounted for in a bin-by-bin manner.

The primary observable used to distinguish signal from background is the invariant mass of the t t bar system (m t t bar). Additionally, angular variables sensitive to t t bar spin correlations are utilized in both channels. p. 2

Where the paper says it

It is the first time in Run 2 that the dilepton channel has been included in an ATLAS search for these types of resonances, and both the single-lepton and dilepton channels use the full Run-2 data for the first time. Additionally, the single-lepton channel improves on the previous 36 fb−1 analysis through improved 𝑡𝑡¯ reconstruction and the addition of validation regions to better evaluate the background modelling.

These angular variables are employed not only to define validation regions for the SM t t bar background but also to enhance the sensitivity of the signal regions, particularly in the 2-lepton channel. p. 2
Where the paper says it

It is the first time in Run 2 that the dilepton channel has been included in an ATLAS search for these types of resonances, and both the single-lepton and dilepton channels use the full Run-2 data for the first time. Additionally, the single-lepton channel improves on the previous 36 fb−1 analysis through improved 𝑡𝑡¯ reconstruction and the addition of validation regions to better evaluate the background modelling.

Results and Exclusion Limits

The analysis found no significant deviations from Standard Model background expectations across any probed invariant mass spectrum (m t t bar). Consequently, upper limits are set at the 95% confidence level (CL) on the production cross-section times branching ratio (sigma times BR) for the benchmark scenarios investigated. p. 3

Where the paper says it

The description of the statistical analysis is provided in Section 9 and the systematic uncertainties are detailed in Section 10. Finally, results are presented and discussed in Section 11, where, in the absence of any significant excess, exclusion limits on the production cross-section times branching ratio for the hypothetical new particles are presented at the 95% confidence level.

The constraints derived are highly impactful:

  • Narrow Resonances (m t t bar < 1.5 TeV): The search establishes the most stringent constraints to date on narrow t t bar resonances in this lower mass range. For instance, it excludes a Z'TC2 signal with a relative width of 3% (or 1.2%) up to a mass of approximately 4. p. 2

    Where the paper says it

    These signals are hence not considered in this paper. In the search performed by CMS, a significant deviation from the SM expected limits for a 𝑡𝑡¯ mass less than approximately 400 GeV was observed.

    2 TeV (3.65 TeV).

  • Broad Resonances: The analysis yields the strongest exclusions to date for the KK gluon (gKK) signal with a relative width of 30%, ruling out masses below 4.1 TeV. The constraints on Kaluza–Klein graviton (GKK) signals are reported as being the strongest in this study, excluding masses below 1. p. 2

    Where the paper says it

    (TC2) [3–5] and Randall–Sundrum (RS) models of warped extra dimensions [6, 7], predicting one or several additional heavy bosons, such as a spin-1 𝑍 ′ particle, a spin-2 Kaluza–Klein (KK) graviton (𝐺KK), or a spin-1 KK gluon (𝑔KK). The new states can have strong or preferential couplings to the top quark, the heaviest elementary particle of the SM.

    3 TeV.

  • Overall Reach: The results collectively enhance existing exclusion limits across a broad range of models predicting heavy resonances decaying into top quark pairs, covering the entire mass range from 0.4 TeV to 5.0 TeV.

In summary, this paper provides a robust and highly sensitive search utilizing the full Run 2 dataset with complementary single-lepton and dilepton channels to constrain the parameter space for heavy spin-1 and spin-2 particles decaying into top quark pairs, setting some of the most powerful exclusion limits currently available in this sector. p. 2

Where the paper says it

The results in this paper consider spin-1 and spin-2 signal models and are valid for the case when interference is negligible. It is the first time in Run 2 that the dilepton channel has been included in an ATLAS search for these types of resonances, and both the single-lepton and dilepton channels use the full Run-2 data for the first time.

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. Enhanced Signal Discrimination for Heavy Resonances: The improved AI can perform highly precise discrimination between spin-1 and spin-2 resonances by analyzing angular variables sensitive to top quark spin correlations in both 1-lepton and 2-lepton channels. This allows the system to better separate the signal from the Standard Model background, as stated in "The angular variables provide additional discrimination power between the signal process, in which the tt¯system is produced via the s-channel diagram, and the background from SM tt¯ production, which mostly proceeds via the t-channel diagram."

  2. Advanced Background Modeling: The system can accurately estimate and correct background contributions using data-driven methods like matrix method estimations for fakes and data-driven reweighting for Z plus jets. This ensures that Data-driven corrections are applied to the Sherpa MC simulation samples for the Z + jets background in the 2-lepton channel before the final fit to correct the overall normalisation of this background component.

  3. Improved Uncertainty Quantification: The AI can incorporate complex uncertainty correlation schemes into its statistical analysis, treating experimental uncertainties as fully correlated across samples and fitting them using a Beeston-Barlow lite technique for MC statistics. This provides a more robust determination of the total uncertainty by summing in quadrature the impacts of all relevant NP groups.

Abstract

A search for heavy spin-1 and spin-2 resonances decaying into a top-antitop-quark pair has been performed with 140 fb-1 of proton-proton collision data collected by the ATLAS experiment at the Large Hadron Collider at a centre-of-mass energy of sqrt s =13 TeV. Final states with either exactly one electron or muon, or exactly two leptons (ee, μμ or eμ), large missing transverse momentum, and two jets, at least one of which must be identified as likely containing a b-hadron decay, are considered. The search targets resonances with both narrow and broad widths relative to the detector resolution, and with masses in the range of 0.4-5.0 TeV. No significant deviation from the Standard Model prediction is observed. Exclusion limits are set on the production cross-section times branching ratio for hypothetical Z' bosons, Kaluza-Klein gravitons, and Kaluza-Klein gluons that decay into top-quark pairs.

Sources

Related papers