REVIEW 2 major objections 4 minor 1 cited by
Probing a 2HDM Type-I light Higgs state via $H_{\rm SM} \to hh \to b\bar b\gamma \gamma$ at the LHC
T0 review · 2 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read In a Type-I 2HDM, the gg→H→hh→bbγγ chain is discoverable at Run 3 for five benchmark points.
desk verdict A useful, honestly-written search proposal for H->hh->bbγγ in 2HDM Type-I, whose 5-sigma claims rest on an unquantified zero-background assumption for the irreducible pp→bbγγ continuum. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing identity is the equal-mass correlation of the two reconstructed decay pairs: in the signal both $b\bar b$ and $\gamma\gamma$ come from the same on-shell light scalar $h$, so their invariant masses agree, while the Standard Model backgrounds have no such correlation. The analysis captures this with $\Delta m_h = (m_{b\bar b}-m_{\gamma\gamma})/m_{\gamma\gamma}<0.25$, together with $m_{b\bar b}<62$ GeV, $m_{\gamma\gamma}<62$ GeV, and $m_{b\bar b\gamma\gamma}<150$ GeV. The 2HDM Type-I inverted hierarchy matters because it makes $H$ the 125 GeV state while $h$ retains $\mathrm{BR}(h\to b\bar b)\approx 80\%$ and a few-percent $\mathrm{BR}(h\to\gamma\gamma)$. The selection also relies on two $b$-tagged jets with $p_T>20$ GeV and two barrel photons with thresholds 22/14 GeV, a trigger choice tuned to the soft photons from a light scalar; after the mass cuts the modeled backgrounds vanish, so the significance is approximately $\sqrt{N_S}$.
What would settle it
Generate the irreducible $pp\to b\bar b\gamma\gamma$ continuum at $\sqrt{s}=14$ TeV, apply the identical event selection and 300 fb${}^{-1}$ luminosity, and count survivors: more than about one event would invalidate $\Sigma=\sqrt{N_S}$. A sideband measurement in real Run-3 data, selecting events near $m_{b\bar b\gamma\gamma}=125$ GeV with $m_{b\bar b}$ and $m_{\gamma\gamma}$ below 62 GeV but off the h-mass peak, would settle the same question empirically.
Extended reading notes
Core claim
The paper's central claim is that, in the 2HDM Type-I with an inverted mass hierarchy, the process $gg\to H(125)\to hh\to b\bar b\gamma\gamma$ is not just allowed but discoverable at the LHC. For six benchmark points with $m_h = 20, 25, 30, 35, 40, 44$ GeV that pass theoretical and experimental constraints, the Monte Carlo simulation yields 1.91, 20.57, 25.18, 20.75, 13.10, and 18.55 signal events after all cuts at 300 fb${}^{-1}$, with essentially zero modeled background, giving significances of 2.18, 7.07, 7.9, 7.2, 5.72, and 6.8 $\sigma$ after including a K-factor. The paper concludes that five of the six points are within discovery reach at Run 3, and that the roughly tenfold larger HL-LHC luminosity would bring the $m_h=20$ GeV point above $5\sigma$ as well.
Load-bearing premise
The decisive assumption is that the uncomputed backgrounds, especially genuine $pp\to b\bar b\gamma\gamma$ events, leave fewer than about one event after the final cuts, and that the fast detector simulation correctly handles the very soft b-jets and photons produced by a 20 to 44 GeV scalar.
Editorial extensions
If this is right
- A dedicated Run-3 search for $b\bar b\gamma\gamma$ with $m_{b\bar b}, m_{\gamma\gamma}<62$ GeV could discover the Type-I 2HDM light Higgs for $m_h$ between 25 and 44 GeV with 300 fb${}^{-1}$.
- With the HL-LHC's roughly ten times larger integrated luminosity, the quoted significances scale approximately by $\sqrt{10}$, bringing the $m_h=20$ GeV benchmark above the $5\sigma$ threshold.
- Because no upper limit currently exists on $\mathrm{BR}(H\to aa\to b\bar b\gamma\gamma)$, a first dedicated search would set the initial constraint in the 10 to 62 GeV mass window whether or not a signal appears.
- The same $\Delta m_h$ mass-correlation cut should suppress the dominant $t\bar t H$, $ZH$, and $b\bar b H$ backgrounds in other $H\to hh$ channels where both daughter pairs are reconstructable, so the strategy transfers to final states such as $b\bar b\tau^+\tau^-$.
Reading between the lines
- The $\Delta m_h<0.25$ cut is really an equal-mass test: applied to $H\to aa$ cascades with a pseudoscalar daughter, it should work identically, so the search strategy generalizes beyond CP-even scalars to any light resonance decaying to two reconstructable pairs.
- The quoted significances should be treated as an upper bound until the $pp\to b\bar b\gamma\gamma$ continuum is computed with the same cuts; its survival rate, which the paper estimates only by argument, is the single number that would most directly move the discovery claim.
- The barrel-only photon trigger with 22/14 GeV thresholds sets a practical low-mass floor: for $m_h\lesssim 20$ GeV the photons are too soft for the trigger, so the low-mass reach is likely trigger-limited rather than rate-limited.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper studies the LHC discovery potential of the process gg -> H(125) -> hh -> b bbar gamma gamma in the Type-I 2HDM with inverted mass hierarchy, where the heavier CP-even state H is identified with the observed 125 GeV Higgs boson. The authors scan the parameter space subject to theoretical and experimental constraints, select six benchmark points with m_h between 20 and 44 GeV, simulate signal and several H(->gamma gamma) associated backgrounds with MadGraph/Pythia/Delphes, and apply cuts on m_bb, m_gamma gamma, Delta m_h, and the reconstructed m_H. The resulting signal event counts after the full selection are 1.9-25 events at 300 fb^-1, while the simulated backgrounds are found to vanish after the final cuts. Significances of 2.18-7.9 sigma are quoted in Tab. V, with five benchmark points above 5 sigma at sqrt(s)=14 TeV.
Significance. If the zero-background premise is correct, the paper identifies an interesting and largely unexplored final state for light Higgs searches in the 2HDM Type-I, and it provides six explicit benchmark points that can be used by experimental collaborations. The analysis uses public tools in a reproducible way, the benchmark-point constraints are treated with standard codes, and the internal arithmetic of the cutflows and significances is consistent. The main limitation is that the discovery claim rests entirely on the assertion that the irreducible pp -> b bbar gamma gamma continuum and reducible jet backgrounds are negligible; this assertion is not quantified, and for the irreducible component the b-tagging and photon-mistagging arguments given in the Conclusions do not apply. The paper is therefore a useful phenomenological starting point, but the central claim is not yet supported to the standard required for a discovery-level statement.
major comments (2)
- [Section IV, Conclusions (last paragraph before acknowledgments)] The claim that the omitted backgrounds pp -> b bbar gamma gamma, pp -> b bbar jj, pp -> jjjj, and pp -> jj gamma gamma are negligible is load-bearing for all of the Tab. V significances, but no numerical estimate is provided. For pp -> b bbar gamma gamma in particular, both b-jets and both photons are genuine, so the stated reasoning based on b-tagging efficiency and photon mistagging rates does not apply; the only suppression must come from the cuts m_bb < 62 GeV, m_gamma gamma < 62 GeV, Delta m_h < 0.25, and m_H < 150 GeV. Given the signal counts of 1.9-25 events after the full selection at 300 fb^-1 (Tab. III), even a modest survival of the continuum, e.g. about 10 events for BP5, would bring the quoted 5.72 sigma below 5 sigma. The authors should generate pp -> b bbar gamma gamma with the same selection and report the cutflow after each of the final cuts, or provide an analytic estimate of the continuum survival probability. Without this, the discovery claim is not supported.
- [Section III.B, Tab. V and text preceding it] The significance formula is stated as Sigma = N_S/sqrt(N_S + N_B) approximately sqrt(N_S), but the numerical values in Tab. V correspond to sqrt(K*N_S) with K approximately 2.4-2.5. Since the K-factor is applied to the signal but not to the backgrounds, the notation and the definition of Sigma should be clarified. More importantly, with N_B set to zero the quoted significances contain no systematic uncertainties; for a discovery claim in a final state with low-pT b-jets and photons, detector-related systematics on b-tagging, photon identification, and luminosity can be comparable to the statistical error. A brief treatment or at least an explicit statement of the assumed systematic uncertainties is needed before 5 sigma statements can be taken at face value.
minor comments (4)
- [Tab. III] The first row of Tab. III appears garbled: the entries '823.5 27002970.0002591 1980 3027' are not readable as event counts and should be reformatted as six separate numbers.
- [Section III.B, Eq. (4)] The notation pT(gamma[b,j]) is confusing; it would be clearer to write the three separate thresholds, e.g. pT(gamma) > 5 GeV and pT(b), pT(j) > 10 GeV, together with the eta and Delta R requirements.
- [Section III.B, footnote 1] The statement that pp -> t tbar h, Zh, and b bbar h contributions are 'negligibly small' would benefit from a numerical upper bound or a representative cutflow entry, since these processes have the same final state after h -> b bbar or h -> gamma gamma and are not included in Tab. IV.
- [Section II.B] The phrase 'restrict here BR(H -> hh) to below 4%' is stated without a citation or a quantitative derivation; please indicate the source of this bound, as it is relevant for the benchmark-point selection.
Circularity Check
No significant circularity: the signal, backgrounds, and significances are computed from declared model parameters with public tools, and no predicted quantity is fitted to the target observable.
full rationale
The derivation chain is self-contained in the following sense. The benchmark points in Tab. I are inputs, fixed by a random scan over 2HDM Type-I parameters that pass 2HDMC, SuperIso, and HiggsTools constraints; the LO cross sections and branching ratios are computed from these parameters with SusHi and 2HDMC, and the event counts in Tabs. III and IV are obtained by passing MadGraph, Pythia8, and Delphes simulated events through the stated cuts. No parameter is fitted to the final significance, and no equation defines a predicted quantity in terms of the observed event count. The only self-citation, Ref. [15], motivates the 20/20 GeV b-jet pT threshold by reporting that this threshold gave better significance in the related bb tau tau final state; this is a methodological carryover and not a load-bearing reduction, since the threshold is a selection choice rather than a fitted value of the present signal yield. The paper's neglect of pp to bb gamma gamma and reducible backgrounds is an unquantified physics assumption that would affect the significance if wrong, but it is an assumption about background rates, not a circularity: the background is not used to define the signal nor is any parameter adjusted to force the background to zero. No circular step can therefore be exhibited from the text, and the honest finding is a score of 0.
Assumptions & free parameters
free parameters (2)
- Benchmark point parameter set {mh, tan beta, s(beta-alpha), mA, mH+/-} =
BP1-6: mh 20/25/30/35/40/44 GeV; tan beta 8/13/20/17/14/10; s(beta-alpha) -0.12/-0.07/-0.05/-0.06/-0.077/-0.11; mA…
- m2_12 (soft Z2 breaking mass parameter) =
Scanned in [0, mh^2 cos beta sin beta]
assumptions (4)
- domain assumption The 2HDM Type-I with an inverted mass hierarchy and a softly broken Z2 symmetry, as defined in Sect. II (potential in Eq. 1, Yukawa couplings in Eq. 3), is the correct effective model for the light-Higgs scenario.
- ad hoc to paper The un-simulated backgrounds pp to b bbar gamma gamma, pp to b bbar jj, pp to jjjj, and pp to jj gamma gamma contribute less than about one event after the final cuts at 300 fb^-1.
- domain assumption Delphes 3.5 with the standard CMS card, together with the stated barrel-only diphoton trigger thresholds (22/14 GeV), reliably models the detector response for b-jets and photons from a 20 to 44 GeV scalar.
- domain assumption The significance formula Sigma = N_S/sqrt(N_S+N_B) with N_B about 0 and a signal-only K-factor of about 2.4 to 2.5 adequately quantifies the discovery potential.
Cite this review
Pith. "Pith review of Probing a 2HDM Type-I light Higgs state via $H_{\rm SM} \to hh \to b\bar b\gamma \gamma$ at the LHC." pith.science (2026). https://pith.science/paper/ADU2ZDGD
@misc{pith2026241206052,
author = {Pith},
title = {Pith review of: Probing a 2HDM Type-I light Higgs state via $H_\rm SM \to hh \to b\bar b\gamma \gamma$ at the LHC},
year = {2026},
howpublished = {\url{https://pith.science/paper/ADU2ZDGD}},
note = {Machine review of arXiv:2412.06052}
}
abstract
We study the discovery potential for a light Higgs boson via $gg \to H_{\text{SM-like}} \to hh \to b\bar{b}\gamma\gamma$ process at the Large Hadron Collider (LHC). Focusing on the 2-Higgs Doublet Model (2HDM) Type-I, which can accommodate light neutral Higgs states, of $\mathcal{O}(100)$ GeV or less in mass, while agreeing with theoretical and up-to-date experimental constraints, we explore the feasibility of a light CP-even Higgs state $h$ via the largely unexplored final state $b\bar{b}\gamma\gamma$ at Run-3 of the LHC. We further propose a few Benchmark Points (BPs) for future searches.
Figures
Forward citations
Cited by 1 Pith paper
-
Light scalars within the $\mathcal{CP}$-conserving Aligned-two-Higgs-doublet model
All seven light-scalar scenarios of the CP-conserving aligned two-Higgs-doublet model contain sizeable parameter regions compatible with current collider and flavour data, with the lightest new scalars bounded by LEP ...
Reference graph
Works this paper leans on
-
[1]
Georges Aad et al. (ATLAS), “Observation of a new particle in the search for the Standard Model Higgs bo- son with the ATLAS detector at the LHC,” Phys. Lett. B 716, 1–29 (2012), arXiv:1207.7214 [hep-ex]
arXiv 2012
-
[2]
ObservationofaNew Boson at a Mass of 125 GeV with the CMS Experi- ment at the LHC,
SergueiChatrchyan et al. (CMS),“ObservationofaNew Boson at a Mass of 125 GeV with the CMS Experi- ment at the LHC,” Phys. Lett. B 716, 30–61 (2012), arXiv:1207.7235 [hep-ex]
arXiv 2012
-
[3]
A detailed map of Higgs boson interactions by the AT- LAS experiment ten years after the discovery,
“A detailed map of Higgs boson interactions by the AT- LAS experiment ten years after the discovery,” Nature 607, 52–59 (2022), [Erratum: Nature 612, E24 (2022)], arXiv:2207.00092 [hep-ex]
arXiv 2022
-
[4]
A portrait of the Higgs bosonbytheCMSexperimenttenyearsafterthediscov- ery,
Armen Tumasyanet al. (CMS), “A portrait of the Higgs bosonbytheCMSexperimenttenyearsafterthediscov- ery,” Nature 607, 60–68 (2022), arXiv:2207.00043 [hep- ex]
arXiv 2022
-
[5]
Armen Tumasyanet al. (CMS), “Search for the exotic decay of the Higgs boson into two light pseudoscalars with four photons in the final state in proton-proton collisions at √s = 13 TeV,” JHEP 07, 148 (2023), arXiv:2208.01469 [hep-ex]
arXiv 2023
-
[6]
Georges Aad et al. (ATLAS), “Search for short- and long-lived axion-like particles inH → aa → 4γ decays with the ATLAS experiment at the LHC,” Eur. Phys. J. C 84, 742 (2024), arXiv:2312.03306 [hep-ex]
arXiv 2024
-
[7]
Morad Aaboudet al. (ATLAS), “Search for Higgs boson decays into pairs of light (pseudo)scalar particles in the γγjj final state inpp collisions at√s = 13 TeV with the ATLAS detector,” Phys. Lett. B782, 750–767 (2018), arXiv:1803.11145 [hep-ex]
arXiv 2018
-
[8]
Fermiophobic Higgs bosons at the Tevatron,
A. G. Akeroyd, “Fermiophobic Higgs bosons at the Tevatron,” Phys. Lett. B368, 89–95 (1996), arXiv:hep- ph/9511347
Show all 30 references
-
[9]
Fermiophobic and other nonminimal neutral Higgs bosons at the LHC,
A. G. Akeroyd, “Fermiophobic and other nonminimal neutral Higgs bosons at the LHC,” J. Phys. G24, 1983– 1994 (1998), arXiv:hep-ph/9803324
1998 arXiv
-
[10]
Experimental signatures of fermiophobic Higgs bosons,
L. Brucher and R. Santos, “Experimental signatures of fermiophobic Higgs bosons,” Eur. Phys. J. C12, 87–98 (2000), arXiv:hep-ph/9907434
2000 arXiv
-
[11]
Double fermiophobic Higgs boson production at the CERN LHC and LC,
A. G. Akeroyd, Marco A. Diaz, and Francisco J. Pacheco, “Double fermiophobic Higgs boson production at the CERN LHC and LC,” Phys. Rev. D70, 075002 (2004), arXiv:hep-ph/0312231
2004 arXiv
-
[12]
By imposing aZ2 symmetry, these FCNCs can be suppressed, resulting in four main types of 2HDM, where each is defined by how the Higgs dou- blets couple to fermions
(2) The presence of an extra Higgs doublet in the Yukawa sector introduces unwanted Flavour Changing Neutral Currents (FCNCs). By imposing aZ2 symmetry, these FCNCs can be suppressed, resulting in four main types of 2HDM, where each is defined by how the Higgs dou- blets coupl...
1980
-
[13]
Search for a light fermiophobic Higgs boson produced via gluon fusion at Hadron Colliders,
Abdesslam Arhrib, Rachid Benbrik, R. B. Guedes, and R. Santos, “Search for a light fermiophobic Higgs boson produced via gluon fusion at Hadron Colliders,” Phys. Rev. D 78, 075002 (2008), arXiv:0805.1603 [hep-ph]
2008 arXiv
-
[14]
Multi-photon production in the Type-I 2HDM,
Abdesslam Arhrib, Rachid Benbrik, Stefano Moretti, Abdessamad Rouchad, Qi-Shu Yan, and Xian- hui Zhang, “Multi-photon production in the Type-I 2HDM,” JHEP 07, 007 (2018), arXiv:1712.05332 [hep- ph]
2018 arXiv
-
[15]
Fermiophobic light Higgs boson in thetype-Itwo-Higgs-doubletmodel,
Jinheung Kim, Soojin Lee, Jeonghyeon Song, and Prasenjit Sanyal, “Fermiophobic light Higgs boson in thetype-Itwo-Higgs-doubletmodel,” Phys.Lett.B 834, 137406 (2022), arXiv:2207.05104 [hep-ph]
2022 arXiv
-
[16]
Searching for H→hh→bb¯τ τin the 2HDM type-I at the LHC,
A. Arhrib, S. Moretti, S. Semlali, C. H. Shepherd- Themistocleous, Y. Wang, and Q. S. Yan, “Searching for H→hh→bb¯τ τin the 2HDM type-I at the LHC,” Phys. Rev. D 109, 055020 (2024), arXiv:2310.02736 [hep-ph]
2024 arXiv
-
[17]
Theory and phe- nomenology of two-Higgs-doublet models,
G. C. Branco, P. M. Ferreira, L. Lavoura, M. N. Re- belo, Marc Sher, and Joao P. Silva, “Theory and phe- nomenology of two-Higgs-doublet models,” Phys. Rept. 516, 1–102 (2012), arXiv:1106.0034 [hep-ph]
2012 arXiv
-
[18]
2HDMC: Two-Higgs-Doublet Model Calculator Physics and Manual,
David Eriksson, Johan Rathsman, and Oscar Stal, “2HDMC: Two-Higgs-Doublet Model Calculator Physics and Manual,” Comput. Phys. Commun. 181, 189–205 (2010), arXiv:0902.0851 [hep-ph]
2010 arXiv
-
[19]
Review of Par- ticle Physics,
P. A. Zylaet al. (Particle Data Group), “Review of Par- ticle Physics,” PTEP2020, 083C01 (2020)
2020
-
[20]
SuperIso v2.3: A Program for cal- culating flavor physics observables in Supersymme- try,
F. Mahmoudi, “SuperIso v2.3: A Program for cal- culating flavor physics observables in Supersymme- try,” Comput. Phys. Commun.180, 1579–1613 (2009), arXiv:0808.3144 [hep-ph]
2009 arXiv
-
[21]
HiggsTools: BSM scalar phenomenol- ogy with new versions of HiggsBounds and HiggsSig- nals,
Henning Bahl, Thomas Biekötter, Sven Heinemeyer, Cheng Li, Steven Paasch, Georg Weiglein, and Jonas Wittbrodt, “HiggsTools: BSM scalar phenomenol- ogy with new versions of HiggsBounds and HiggsSig- nals,” Comput. Phys. Commun. 291, 108803 (2023), arXiv:2210.09332 [hep-ph]
2023 arXiv
-
[22]
SusHi Bento: Beyond NNLO and the heavy- top limit,
Robert V. Harlander, Stefan Liebler, and Hendrik Mantler, “SusHi Bento: Beyond NNLO and the heavy- top limit,” Comput. Phys. Commun. 212, 239–257 (2017), arXiv:1605.03190 [hep-ph]
2017 arXiv
-
[23]
Report from Working Group 2: Higgs Physics at the HL-LHC and HE-LHC,
M. Cepedaet al., “Report from Working Group 2: Higgs Physics at the HL-LHC and HE-LHC,” CERN Yellow Rep.Monogr. 7,221–584(2019),arXiv:1902.00134[hep- ph]
2019 arXiv
-
[24]
The automated computation of tree-level and next-to-leading order differential cross sections, and theirmatchingtopartonshowersimulations,
J. Alwall, R. Frederix, S. Frixione, V. Hirschi, F. Mal- toni, O. Mattelaer, H. S. Shao, T. Stelzer, P. Torrielli, and M. Zaro, “The automated computation of tree-level and next-to-leading order differential cross sections, and theirmatchingtopartonshowersimulations,” JHEP 07,...
2014 arXiv
-
[25]
PYTHIA 6.4 Physics and Manual,
Torbjorn Sjostrand, Stephen Mrenna, and Peter Z. Skands, “PYTHIA 6.4 Physics and Manual,” JHEP05, 026 (2006), arXiv:hep-ph/0603175
2006 arXiv
-
[26]
DELPHES 3, A modular framework for fast simu- lation of a generic collider experiment,
J. de Favereau, C. Delaere, P. Demin, A. Giammanco, V. Lemaître, A. Mertens, and M. Selvaggi (DELPHES 3), “DELPHES 3, A modular framework for fast simu- lation of a generic collider experiment,” JHEP02, 057 (2014), arXiv:1307.6346 [hep-ex]
2014 arXiv
-
[27]
MadAnalysis 5, A User-Friendly Framework for Col- lider Phenomenology,
Eric Conte, Benjamin Fuks, and Guillaume Serret, “MadAnalysis 5, A User-Friendly Framework for Col- lider Phenomenology,” Comput. Phys. Commun. 184, 222–256 (2013), arXiv:1206.1599 [hep-ph]
2013 arXiv
-
[28]
Performance of Photon Reconstruction andIdentification withthe CMS Detector in Proton-Proton Collisions at sqrt(s) = 8 TeV,
Vardan Khachatryan et al. (CMS), “Performance of Photon Reconstruction andIdentification withthe CMS Detector in Proton-Proton Collisions at sqrt(s) = 8 TeV,” JINST 10, P08010 (2015), arXiv:1502.02702 [physics.ins-det]
2015 arXiv
-
[29]
The anti-kt jet clustering algorithm,
Matteo Cacciari, Gavin P. Salam, and Gregory Soyez, “The anti-kt jet clustering algorithm,” JHEP 04, 063 (2008), arXiv:0802.1189 [hep-ph]
2008 arXiv
-
[30]
Machine Learning Applications for Jet Tag- ging in the CMS Experiment,
Antimo Cagnotta, Francesco Carnevali, and Agostino De Iorio, “Machine Learning Applications for Jet Tag- ging in the CMS Experiment,” Appl. Sciences12, 10574 (2022). 6
2022
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.