A calibratable jet-free HH(4b) search framework at the LHC
Abstract
Contributed talk at the Higgs Pairs Workshop 2025. Update (August 2025): This work has been released as a paper on arXiv: 2508.15048. Indico link https://indico.cern.ch/event/1399335/contributions/6384808 Talk abstract A calibratable experimental strategy is proposed to enhance the HH(4b) search sensitivity via full-particle classification. Inspired by the competitive performance from the boosted-topology HH analysis, which uses state-of-the-art jet neural networks to analyze o(100) particles within large-R jets, this approach aims to extend its strong signal-to-background discrimination power beyond the boosted regime to a broader phase space accessible through conventional HH(4b) triggers. The approach involves training a universal classifier to distinguish X → Y₁Y₂ → bb̅bb̅ signals from QCD and tt̅ multijet backgrounds across a wide range of X and Y₁,₂ mass values, and simultaneously estimating the Y₁,₂ masses via a multiclass classification technique. Results demonstrate that the background suppression capability matches that of identifying two boosted X → bb̅ jets, revealing a scaling law governing signal and background yields in both cases. The framework is complemented by a robust signal calibration and validation procedure: event-level classifier calibration is performed using an orthogonal dimuon-triggered phase space and an "event hemisphere mixing" technique to construct fake ZZ(4b) events; validation is then conducted using genuine ZZ(4b) data passing the analysis trigger. With combined Run 2 and 3 datasets, the proposed strategy can achieve the first observation of the ZZ(4b) process and deliver a search sensitivity for HH(4b) comparable to HL-LHC projection. This approach holds a great premise to accelerate the pace of HH search at the LHC and advance our understanding of the Higgs self-coupling.
Full text
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 1 A calibratable jet-free HH(4b) search framework at the LHC Tianyi Yang and Congqiao Li (Peking University) Higgs Pairs Workshop 2025, Isola d'Elba 16 May, 2025
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 2 b ! b b ! b b b ! b ! b cartoons credit to link HH→4b at the LHC See detail in Marina and Yizhou’s talk (Mon)
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Boosted regime as a booster — even in ggF? 3 10−5−0 5 10 15 20 λ κ 10 2 10 3 10 4 10 ) [fb]b bb(b Β HH)→(pp σ95% CL limit on Observed σ 1±Expected Theory prediction σ 2±Expected CMS Supplementary = 1 V κ = 2V κ = t κ (13 TeV) -1 138 fb 10−5−0 5 10 λ κ 1000 2000 3000 4000 HH) [fb]→ (pp ggF+VBF σ CMS (13 TeV) -1 138 fb 95% CL upper limits Observed Median expected 68% expected 95% expected Theoretical prediction bbb b→HH All categories PRL 131 (2023) 041803 PRL 129 (2022) 081802 results websiteresults website • Boosted regime can only focus on high-pT behaviour • can only detect a tiny amount of signals ( GeV) • but their sensitivity is close!? mHH ≳600 Obs (exp) 95% CL: −2.3 (−5.0) < κλ< 9.4 (12.0) Obs (exp) 95% CL: −9.9 (−5.1) < κλ< 16.9 (12.2) Taking CMS results as examples
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Boosted regime as a booster — even in ggF? 4 PRL 131 (2023) 041803 PRL 129 (2022) 081802 results websiteresults website • Boosted regime can only focus on high-pT behaviour • can only detect a tiny amount of signals ( GeV) • but their sensitivity is close!? mHH ≳600 60 80 100 120 140 160 180 200 220 [GeV] reg m 2 j 0.0 0.5 1.0 1.5 2.0 Data / pred. 0 2 4 6 8 10 12 14 16 18 20 Events / 10 GeV Data QCD, ggF H, VBF H VH = 3.5)µHH ( +jetstt Htt V+jets, VV Total unc. CMS ggF cat. 1 (13 TeV) -1 138 fb 0 50 100 150 200 250 300 350 400 Events 2016 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 36 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. 0 100 200 300 400 500 600 700 Events 2017-2018 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 102 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. the high-mHH region 2017-2018 plots Counting the last 8 bins: S=6.7, B=950 S/√B = 0.22 Counting the 4 bins in the mass window: S=1.63, B=9.5, σB=4.9 Z = 0.27 0.51 (if B is certain) Taking CMS results as examples
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Boosted regime as a booster — even in ggF? 5 PRL 131 (2023) 041803 PRL 129 (2022) 081802 results websiteresults website • Boosted regime can only focus on high-pT behaviour • can only detect a tiny amount of signals ( GeV) • but their sensitivity is close!? mHH ≳600 60 80 100 120 140 160 180 200 220 [GeV] reg m 2 j 0.0 0.5 1.0 1.5 2.0 Data / pred. 0 2 4 6 8 10 12 14 16 18 20 Events / 10 GeV Data QCD, ggF H, VBF H VH = 3.5)µHH ( +jetstt Htt V+jets, VV Total unc. CMS ggF cat. 1 (13 TeV) -1 138 fb 0 50 100 150 200 250 300 350 400 Events 2016 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 36 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. 0 100 200 300 400 500 600 700 Events 2017-2018 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 102 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. the high-mHH region 2017-2018 plots Counting the last 8 bins: S=6.7, B=950 S/√B = 0.22 Counting the 4 bins in the mass window: S=1.63, B=9.5, σB=4.9 Z = 0.27 0.51 (B is certain) It's the tagger(!) (ParticleNet/GloParT) that pushes boosted-regime sensitivity to this exceptional level. Can we bring its superb background suppression power to more general regimes?
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Boosted regime as a booster — even in ggF? 5 PRL 131 (2023) 041803 PRL 129 (2022) 081802 results websiteresults website • Boosted regime can only focus on high-pT behaviour • can only detect a tiny amount of signals ( GeV) • but their sensitivity is close!? mHH ≳600 60 80 100 120 140 160 180 200 220 [GeV] reg m 2 j 0.0 0.5 1.0 1.5 2.0 Data / pred. 0 2 4 6 8 10 12 14 16 18 20 Events / 10 GeV Data QCD, ggF H, VBF H VH = 3.5)µHH ( +jetstt Htt V+jets, VV Total unc. CMS ggF cat. 1 (13 TeV) -1 138 fb 0 50 100 150 200 250 300 350 400 Events 2016 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 36 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. 0 100 200 300 400 500 600 700 Events 2017-2018 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 102 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. the high-mHH region 2017-2018 plots Counting the last 8 bins: S=6.7, B=950 S/√B = 0.22 Counting the 4 bins in the mass window: S=1.63, B=9.5, σB=4.9 Z = 0.27 0.51 (B is certain) It's the tagger(!) (ParticleNet/GloParT) that pushes boosted-regime sensitivity to this exceptional level. Can we bring its superb background suppression power to more general regimes? ❖Yes, we do. We expect a >5x significance(!) improvement in HH(4b) ❖Our analysis, presented in this talk, is based on our realistic CMS-like simulation using Delphes
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Interlude: Our journey toward a “more realistic Delphes” 6
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Interlude: Our journey toward a “more realistic Delphes” 6 H. Qu*, CL, S. Qian, ICML 2022 JetClass [data card] February ‘22 Realistic particle-level features: (+impact params d0, dz, d0_err, dz_err) May ‘24 CL* et al. arXiv:2405.12972 JetClass-II [data card] emulate pileup (<μ>=50) & pileup mitigation with PUPPI enable realistic flavour-tagging
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Interlude: Our journey toward a “more realistic Delphes” 6 H. Qu*, CL, S. Qian, ICML 2022 JetClass [data card] February ‘22 Realistic particle-level features: (+impact params d0, dz, d0_err, dz_err) May ‘24 CL* et al. arXiv:2405.12972 JetClass-II [data card] emulate pileup (<μ>=50) & pileup mitigation with PUPPI enable realistic flavour-tagging March ‘25 JetClass-II + realistic small/large-R jet tagger (anti-kT R=0.4/0.8) Y. Zhao, CL* et al. arXiv:2503.00118 [SophonAK4], [Sophon (AK8)]
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 10 100 fb-1 Further require all four leading jets to be b-tagged at medium WP (εl = 1%) Then train a ParT classifier, using four jets’ kinematic inputs to distinguish signal events vs QCD. (ParT algorithm constructs ΔR, mij of all jet pairs, effectively enabling jet-pair association.) 0 50 100 150 200 250 300 350 400 Events 2016 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 36 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. 0 100 200 300 400 500 600 700 Events 2017-2018 Data Bkg. model Bkg. unc. SM ggF-HH x 100 =2) x 100 2V κVBF-HH ( bbb b→HH region 4b SR AHH ggF high-m CMS (13 TeV) -1 102 fb 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 BDT Output 0.6 0.8 1 1.2 1.4 Data/Bkg. PRL 129 (2022) 081802 S=6.7 B=950 Reproduced CMS result!
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 11 100 fb-1 ① Events (100 fb-1) passing boosted-jet triggers Mimic CMS’s typical boosted-jet trigger: ✦HT (AK8 pT>200 GeV) > 800 GeV ✦leading AK8 jet trimmed mass > 50 GeV ② And further require two leading jets to have soft-drop masses within the circle: • centered at (125, 115) with R=25 GeV Then, we reproduce boosted HH4b results Both signal and bkg. events suppressed by x100 !
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 12 100 fb-1 But the QCD suppression power is astounding… using the sum of two AK8 jets’ Sophon (AK8) XbbvsQCD scores as the discriminator
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 12 100 fb-1 But the QCD suppression power is astounding… using the sum of two AK8 jets’ Sophon (AK8) XbbvsQCD scores as the discriminator 60 80 100 120 140 160 180 200 220 [GeV] reg m 2 j 0.0 0.5 1.0 1.5 2.0 Data / pred. 0 2 4 6 8 10 12 14 16 18 20 Events / 10 GeV Data QCD, ggF H, VBF H VH = 3.5)µHH ( +jetstt Htt V+jets, VV Total unc. CMS ggF cat. 1 (13 TeV) -1 138 fb PRL 131 (2023) 041803 (for 100 fb-1) S=1.18 B(QCD)≈3.0 CMS’s performance is even stronger (may be due to the use of an additional BDT)
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 13 100 fb-1 Remarks from comparison of CMS’s resolved & boosted analyses Draw contour lines of s/√b …and we see that the boosted regime even has more potential 😲
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 14 100 fb-1 But… can we improve the resolvedregime strategy? 😕 Unfortunately, if we insist on using AK4 jets in the resolved regime, we yield only marginal gains. Benchmarking the best(!) achievable performance (illustrative only; too aggressive for practical use): Train a ParT on events passing resolved triggers, utilizing full AK4 jet information—including tagging scores Remarks from comparison of CMS’s resolved & boosted analyses
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 15 100 fb-1 🤔 Naively, the strong discriminative power should generalize across all boosting conditions. For the resolved topology, this requires exposing the full event information to the NN! Can we achieve somewhere close to this line? Remarks from comparison of CMS’s resolved & boosted analyses
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Reproducing CMS’s HH→4b results ➔Using our realistic Delphes simulation + the Sophon and SophonAK4 taggers, let us now reproduce the CMS’s HH→4b results 16 100 fb-1 Yes. This is achievable.
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Introducing our HH4b strategy 17
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Analysis selection ➔Train “X→Y1Y2→4b” signal versus QCD and ttbar (passing the “4j3b” resolved trigger!) ❖Y1 and Y2 masses: uniformly distributed in [40, 200] GeV ❖X mass? taking various HH→4b topologies as a reference ❖using full-event particle-level inputs 18 mY1+mY2 figure credit 1.6 (mY1+mY2) mX Take mHH spectrum as a reference
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Further investigation into the classifier 24 Is the QCD background truly irreducible? -No • From signal vs QCD ROC curves, we observe that we can suppress QCD bkgs to a non-negligible level, even in the resolved regime. • Using solely jet-based info, it’s challenging to reach the same stage •Soft radiation pattern matters in vetoing QCD! One reason for the superb performance of boosted X→bb taggers is that “soft radiation patterns” are merged into the large-R jet. For the resolved regime, we need to take them to NN training manually 3
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Further investigation into the classifier 25 Should HH4b vs bkgs have a similar separation power across all Lorentz boost conditions? –Yes 4 ❖We focus on a subset of the QCD sample containing exactly 4 gen-bhadrons, as these events are the most challenging to distinguish from signal ❖Use the maximum pT vector sum of all b-hadron pairs as an indicator to characterize the event’s Lorentz boost ❖Results show that the separation ability is consistent across different boosting conditions!
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Further investigation into the classifier 26 Alternative mass-grid filling method: choose the bin with maximum probability 5 ❖Since the scores reveal the underlying “discrete” probability density, choosing the maximumprobability bin is equivalent to a discrete-value regression if the PDF is Gaussian-like ❖But our 2D mass-grid PDF may not be 2D Gaussian-like ❖Changing to this strategy further improves bkg. suppression by 3x, but introduces complexity in bkg estimation…(more to be investigated) Resolved: all particles (SophonHH-13M) Resolved: all particles (SophonHH-2.6M) Resolved: all particles (SophonHH-13M) Resolved: all particles (SophonHH-2.6M) Instead, fill the mass grids via maximum prob score
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Calibration strategy 27 ZZ→4b HH→4b We assume their MC-to-data scale factors, peak shifts, and smearing effects are the same ZZ→4b extraction (with SF, peak shifts and smearing obtained from the fit) HH→4b extraction (using shared factors) A valid strategy An even safer strategy ZZ→4b extraction (using shared factors) HH→4b extraction (using shared factors) extracting fake “ZZ→4b” obtained from event mixing (with SF, peak shifts and smearing obtained from the fit in a different phase space) Validation region
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Calibration: event mixing overview 28 ➔Using dimuon-triggered events with mμμ within the Z mass window. All b-tagged jets should be well separated from the reconstructed Z candidate (to veto tt () ➔Transverse plane splitting: divide the transverse plane into two semi-circles, placing both muons in the same semi-circle. ➔Replace the semi-circle containing the two muons with a corresponding non-dimuon semi-circle from a different event, matching on key kinematic variables: pT, Δφ, pz, and mass ➔Process the mixed events through the same classifier and mass-grid filling workflow as the main analysis. ➔The mixing can be multifaceted: ❖Z(μμ)+jets ⊗ Z(μμ)+jets, Z(μμ)+jet ⊗ ttbar, … ❖Only Z(μμ)Z(bb) ⊗ Z(μμ)Z(bb) mixing generates a ZZ(4b) events and thus produces a peak at (90, 90) on the spectrum ➔Method still under investigation! Orange: leptons Grey: R=0.4 jets Blue: particle-flow particles Black: reconstructed Z boson from μμ
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Summary ➔The remarkable background suppression power at the boosted regime suggests that it can be extended to a broader phase space ➔Several findings that challenge the conventional understanding of HH measurement ❖even in the resolved topology, QCD multijet can be suppressed to a negligible level—but only achievable if going beyond jets and utilizing full-event particles ❖signal vs. QCD suppression should be similar at all Lorentz boosts ❖AI engineering (e.g. effective algorithms, increased training size) is now the key driving force of the sensitivity ➔Proposed a methodology in calibration: a safer calibration–validation–measurement workflow, where calibration is delivered on fake “mixed” events (still in progress) ➔A significant leap in HH sensitivity! ❖current expected significance under 450 fb-1 (work in progress): ZZ→4b: 4.5σ, HH→4b: 2.2σ (>5x enhancement than established strategy!) ❖further improvements are anticipated: better mass-grid filling strategy; improved Run3 triggers 29
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 30 Backup
A calibratable jet-free HH(4b) search framework at the LHC Congqiao Li (Peking University) 16 May, 2025 Higgs Pairs Workshop 2025 Higgs Pairs Workshop 2025 Sample production ➔QCD: ❖Pythia-generated ❖for resolved studies: only kept events passing the resolved “4j3b” trigger → enabled GEN-level filters to improve event-generation efficiency: GEN HT and GEN jet pT cuts (looser than reco-level cuts to avoid overremoving phase space) ❖use the Pythia calculated cross-section ➔Z(qq)+jets ❖generated by MadGraph LO with the addition +0/1/2 jets and MLM matching with partons. Z→qq decay implemented in MadGraph ❖for resolved studies: only kept events passing the resolved “4j3b” trigger → use matrix-element-level cuts, on pT,j1/2/3 and HT ❖use the MadGraph calculated cross-section ➔ ttbar, single-top, tW, tt (W, tt (Z, WW, WZ, ZZ ❖generated by MadGraph LO with the addition +0/1/2 jets, or +0/1 jet, or 0 jet (depending on the processes) ❖consider inclusive “boson -> fermion fermion” decay modes ❖use the highest-order cross sections recommended ➔ggF Higgs, VBF Higgs, VH, tt (H & ggHH signal ❖generated by Powheg at NLO ❖consider inclusive decay ❖use the Powheg cross sections 31