{"id":"a2302f24-13bb-4033-9fd8-bc40d77044fb","arxiv_id":"2412.14906","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":11,"one_line_summary":"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 and LHC searches.","lead":"This paper runs global statistical fits of the CP-conserving Aligned-two-Higgs-doublet model allowing new scalars lighter than the 125 GeV Higgs. It finds that all seven possible light-scalar scenarios still have sizeable parameter regions compatible with current data, and it maps the gaps where future searches could find them.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Missing b→cτν observables (RD, RD*) in Sec. 3.4 leaves the high-ςl region of light-H± scenarios untested; adding them may shrink or close the claimed allowed regions.","rationale":"The reader's weakest assumption (Sec. 3.6 likelihood) is legitimate: the truncated-normal mapping from 95% CL upper limits is ad hoc, and the absence of deposited configuration files weakens reproducibility. However, that critique concerns how already-included limits are encoded, and it is unlikely by itself to eliminate an entire scenario unless the variance is badly mistuned. The omission of RD and RD* is more load-bearing because it is a definite, well-established dataset directly sensitive to the same couplings (ς_d, ς_l, M_H±) that the paper identifies as weakly constrained. The published fits therefore cannot fully support 'compatible with all current data' while a relevant current dataset is absent. The proposed test is concrete and feasible: HEPfit already contains the necessary machinery, and Ref. [57], cited by the authors for flavour constraints, provides the charged-Higgs amplitudes. If the test closes the high-ςl region, the practical conclusion about a light charged scalar in the 95–120 GeV window weakens; if it does not, the central claim is robust and the Sec. 3.6 caveat remains the main residual concern. I therefore keep the reader's CONDITIONAL verdict unchanged, and specify the RD/RD* check as the condition that should be satisfied before accepting the claim as stated.","tokens_in":44034,"tokens_out":10206,"duration_ms":86775,"concrete_test":"Add B→Dτν and B→D*τν to the HEPfit A2HDM likelihood using the charged-Higgs contributions from Ref. [57] and current HFLAV world averages for RD and RD*; rerun the seven global fits. As a control, first reproduce the published Sec. 5 results with the original likelihood. Then check whether the 95% volume in the M_H±–ςl plane for scenarios with M_H±<M_h shrinks substantially and whether each scenario's posterior still contains the data within 2σ. If any scenario no longer passes, the central claim fails; if all scenarios remain viable, the omitted observable is not fatal and the reader's verdict stands.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The claim that all seven scenarios have sizeable regions compatible with all current data rests on the flavour constraint list in Sec. 3.4. That list includes B→Xsγ, Bs→μ+μ−, and leptonic decays such as B→τν and D(s)→μν, but omits the semileptonic tree-level decays B→Dτν and B→D*τν (RD and RD*). These are among the most sensitive observables to a light charged scalar with non-negligible leptonic couplings: tree-level H± exchange generates a contribution to b→cτν proportional to ς_d ς_l m_b m_τ/M_{H±}^2 (plus a ς_u m_c term), which is O(10–30%) of the SM amplitude for M_H± ≈ 100 GeV, ς_d ≈ 1, and ς_l ≈ 50–100 — exactly the high-ςl window highlighted for light-H± scenarios. Current HFLAV averages of RD and RD* carry ~10% uncertainties, so such shifts are experimentally significant. The paper's assertion that all flavour observables relevant for CP-conserving NP are included is therefore inaccurate, and the conclusion that the leptonic alignment parameter is only weakly constrained is drawn from an incomplete dataset. This is a concrete missing constraint, not a modelling choice: the Sec. 3.6 truncated-normal treatment of direct-search limits is debatable, but the complete absence of RD/RD* is a definite omission.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a Bayesian global fit of the CP-conserving Aligned-two-Higgs-doublet model (A2HDM) in the regime where at least one of the additional scalars is lighter than the 125 GeV Higgs. Seven mass scenarios (one, two, or three light scalars) are fitted separately with HEPfit. The constraints include vacuum stability, perturbative unitarity and perturbativity of the cubic couplings, the S and T oblique parameters, a list of flavour observables, Higgs signal strengths, many LEP and LHC direct searches for neutral and charged scalars, and reinterpreted slepton searches. The central finding is that all seven scenarios contain sizeable regions compatible with current data at the 2σ level, with light neutral scalars bounded above 60 GeV and light charged scalars above 95 GeV, leaving a weakly constrained window around 95–120 GeV for the leptonic alignment parameter |ςl|. The implications of the muon anomalous magnetic moment are analysed separately using two different SM predictions.","tokens_in":44427,"tokens_out":7458,"duration_ms":63076,"significance":"If the main claim survives scrutiny, this is the first global analysis covering all light-scalar scenarios of the A2HDM and it provides a useful reference for future collider searches. The strengths of the paper are its use of the public HEPfit code with a specified commit, the broad and up-to-date set of direct search constraints, the transparent prior choices, the separate treatment of (g−2)μ, and the careful handling of slepton reinterpretation. The paper also gives a benchmark for how viable the low-mass A2HDM remains. The main concerns are the ad hoc likelihood used for direct upper limits and the omission of tree-level b→cτν observables, both of which bear directly on the highlighted high-|ςl| window.","major_comments":[{"comment":"The flavour-observable list omits the tree-level semileptonic decays B→Dτν and B→D*τν (RD and RD*). The text in Sec. 3.4 states that all flavour observables relevant for CP-conserving NP are included, but b→cτν receives a tree-level charged-Higgs contribution proportional to (ςd mb or ςu mc) ςl mτ / M_H±^2. In the allowed high-|ςl| window reported in Sec. 5.1 and Fig. 5 (|ςl| up to about 100, M_H± between 95 and 120 GeV, with ςd not forced to be negligible), these contributions can be O(10%) or larger relative to the SM and are comparable to the current experimental precision of RD and RD*. The conclusion that ςl is only weakly constrained and that all scenarios contain sizeable allowed regions is therefore drawn from an incomplete dataset. I recommend adding RD, RD* (and, where relevant, Bc→τν) to the fit, or at least quantifying their impact on the high-|ςl| window.","section":"Sec. 3.4, Eq. (2.12)"},{"comment":"The direct-search likelihood is constructed by assigning a truncated normal distribution to Odirect = (σ·B)_A2HDM/(σ·B)_exp, with the variance chosen so that Odirect = 1 is excluded at 95% probability. This is not the experimental likelihood, yet it is used for all direct bounds, including the 60 GeV neutral-scalar lower bound, the 95 GeV charged-scalar lower bound, and the restrictions on |ςl|. The derived mass limits and the shape of the high-|ςl| region are therefore contingent on this modelling choice. I recommend a robustness check with an alternative treatment (for example, a hard 95% CL cutoff with theory uncertainty, or a likelihood based on observed event counts or published p-values) to confirm that the seven scenarios remain compatible and that the 95–120 GeV |ςl| window is not an artifact of the chosen likelihood.","section":"Sec. 3.6"},{"comment":"The claim that the scenarios contain 'sizeable regions' of parameter space is not quantified. The paper reports posterior ranges, 95% probability plots, and IC values, but not the prior-to-posterior volume fraction or any other measure of the size of the allowed region. Since this claim is the headline result, I ask the authors to quantify it (for example, by giving posterior mass fractions, credible-region volumes, or a precise definition of 'sizeable') or to temper the wording so that the qualitative statement is not mistaken for a quantitative result.","section":"Sec. 5, Abstract"}],"minor_comments":[{"comment":"The sentence 'the absolute maxima at a 68% probability: ᾱ ≲ 0.05 rad, ςu ≲ 0.1, ςu ≲ 1.7, and ςl ≲ 42' contains a typo: the second 'ςu ≲ 1.7' should be 'ςd ≲ 1.7'.","section":"Sec. 5.1"},{"comment":"The truncated-normal prescription for Odirect is described only in words; a precise definition of the density and its variance would improve reproducibility.","section":"Sec. 3.6"},{"comment":"The CKM inputs are said to follow the procedure of Ref. [100], but the actual numerical inputs are not given; please provide them or an explicit pointer to the relevant appendix in Ref. [100].","section":"Sec. 3.4"},{"comment":"The phrase 'the A2HDM model is clearly not a great solution' is informal; consider 'the A2HDM does not provide a good solution' or similar.","section":"Sec. 6"},{"comment":"The ranges in the tables for heavy neutral scalars are labelled 'Range (TeV)', but many entries are below 1 TeV; please ensure the units are consistent throughout the tables.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"This is a solid global-fit paper that will be of interest to the 2HDM and Higgs-search communities. The missing RD/RD* observables should be added or explicitly assessed, and the direct-search likelihood should be justified or validated by a robustness check. No concerns about citation practices or scope. I recommend major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The genuinely new thing here is the first global fit of the CP-conserving A2HDM with scalars lighter than 125 GeV. Seven mass scenarios are mapped separately, all with surviving regions, using a broad and current constraint set in HEPfit. The direct-search catalog is extensive, the slepton reinterpretation is done carefully (left-handed only, with branching-fraction caveats), and the separate treatment of (g−2)μ is sensible. For the 2HDM subfield this is a useful reference map.\n\nThe soft spots are real but not fatal. The Sec. 3.6 direct-search likelihood is a truncated normal tuned so that O_direct = 1 is excluded at 95% probability. That is an approximation, not the experimental likelihood, and it can deform the shape and size of allowed regions. Minor point: the configuration files are not deposited, only offered on request.\n\nThe bigger issue is the flavour list. The paper claims all flavour observables relevant for CP-conserving NP are included, but RD and RD* are absent. That is a concrete omission, not a modelling choice. Tree-level H± exchange contributes to b→cτν at a level that matters for a light charged scalar with ς_d ~ 1 and ς_l ~ 50–100, exactly the high-ς_l window the paper highlights as weakly constrained. Adding RD/RD* with current HFLAV averages would likely shrink those regions and could close part of the 95–120 GeV gap in ς_l. The core conclusion — that some parameter space survives in all seven scenarios — probably holds, since ς_l can be small, but the 'sizeable' and 'weakly constrained' claims need qualification.\n\nSend this to a serious referee. The referee should ask for RD/RD* to be added, and for the direct-search likelihood to be justified or replaced. With those changes the paper would be a solid, citable result.","headline":"First global low-mass A2HDM fit, useful and mostly sound, but the flavour set omits RD/RD* and the direct-search likelihood is ad hoc, so the size of the claimed allowed regions should be treated with caution.","tokens_in":44924,"tokens_out":1980,"would_cite":true,"duration_ms":18900,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The CP-conserving aligned two-Higgs-doublet model can still hide neutral scalars between 60 and 125 GeV and a charged scalar between 95 and 125 GeV, with all seven light-mass patterns surviving global fits to current data.","keywords":["aligned two-Higgs-doublet model","light scalars","charged Higgs boson","global Bayesian fit","LEP and LHC searches","Yukawa alignment","flavour constraints","muon g-2"],"falsifier":"A dedicated LHC search for $H^\\pm\\to\\tau^\\pm\\nu$ with $M_{H^\\pm}$ between 95 and 120 GeV, with sensitivity to the $\\sigma\\cdot B$ values predicted by this paper's allowed regions, would either confirm or close the large-$\\varsigma_l$ window; equally, an experiment that excludes $\\sigma\\cdot B$ below the A2HDM prediction across that entire window would falsify the claim that the light-charged-scalar scenarios remain viable.","tokens_in":43839,"feed_emoji":"⚛️","tokens_out":8027,"duration_ms":55678,"temperature":0.7,"pith_summary":"The paper asks whether the LHC could have missed additional scalars lighter than the 125 GeV Higgs boson, and answers that within the CP-conserving aligned two-Higgs-doublet model (A2HDM), a Standard Model extension with a second scalar doublet whose Yukawa couplings are aligned with fermion masses, they are not excluded. Depending on which of the three new scalars is light, there are seven distinct scenarios; the authors subject each to global Bayesian fits that combine vacuum stability, perturbativity, electroweak precision data, flavour observables, Higgs rates, and direct LEP and LHC searches. In every scenario the posteriors of all observables agree with measurements within two standard deviations, and the fits' information criteria are similar, with the one-light-neutral cases performing slightly better. This matters because it maps which light-mass windows are still open: neutral scalars between roughly 60 and 125 GeV and a charged scalar between 95 and 125 GeV, with the charged scalar's leptonic coupling weakly constrained in the 95-120 GeV region.","feed_headline":"Seven light-scalar scenarios survive global fits","feed_subtitle":"After all data, neutral scalars down to 60 GeV and a charged scalar near 95-125 GeV remain allowed.","key_machinery":"The load-bearing construction is the CP-conserving A2HDM parameter space with ten beyond-Standard-Model parameters: the three new scalar masses $M_H$, $M_A$, $M_{H^\\pm}$, the CP-even mixing angle $\\tilde\\alpha$, three quartic couplings $\\lambda_2,\\lambda_3,\\lambda_7$, and three Yukawa alignment parameters $\\varsigma_u,\\varsigma_d,\\varsigma_l$. The argument is carried by a Bayesian global fit whose likelihood combines theoretical requirements, precision and flavour observables, Higgs signal strengths, and direct searches; experimental upper limits enter through the ratio $O_{\\rm direct}=(\\sigma\\cdot B)_{\\rm A2HDM}/(\\sigma\\cdot B)_{\\rm exp}$ with a truncated normal distribution tuned so that $O_{\\rm direct}=1$ is excluded at 95% probability. Two features make light scalars viable: the direct searches leave a gap for $M_{H^\\pm}\\in[95,120]$ GeV where the leptonic coupling is nearly unconstrained, and the oblique parameters force the charged-scalar mass to align with one of the neutral masses, which explains the mass correlations among the seven scenarios.","core_discovery":"On its own terms, the paper's central claim is that the CP-conserving A2HDM accommodates all seven possible arrangements of light new scalars alongside the 125 GeV Higgs. A global Bayesian fit yields posterior probabilities for every observable that are compatible with the measured values within two standard deviations, so each scenario retains sizeable allowed regions of parameter space. The fit pins the light neutral scalars above roughly 60 GeV, because below half the Higgs mass the decay $h\\to\\phi\\phi$ is strongly constrained by LHC searches, and the light charged scalar above roughly 95 GeV, from LEP pair-production limits. In the 95-120 GeV window the direct-search constraints on the leptonic alignment parameter $\\varsigma_l$ are essentially absent, allowing values as large as about 100, while flavour constraints force the quark alignment parameters to be small whenever the charged scalar is light. The paper also shows that using the 2020 white-paper value of $(g-2)_\\mu$ degrades the fit substantially, while the newer lattice-based prediction leaves the fit unchanged.","pith_inferences":["If the 95-120 GeV window is where a charged scalar actually hides, the correlation between $M_{H^\\pm}$ and $|\\varsigma_l|$ predicted here discriminates the A2HDM from $Z_2$-symmetric two-Higgs-doublet models, where the leptonic coupling is tied to $\\tan\\beta$.","The near-equal information criteria across the seven scenarios suggest that current data cannot tell which scalar, if any, is light; a future measurement of the $h\\to\\gamma\\gamma$ rate or of triple-Higgs couplings could break this degeneracy.","The reinterpreted slepton searches only start to bite above 115 GeV, so the 95-115 GeV region remains the least explored territory for a light charged Higgs in this model."],"forward_implications":["All seven light-scalar patterns of the CP-conserving A2HDM are experimentally open; none is excluded by current data at 95% probability.","Neutral scalars must sit above about 60 GeV, because below half the Higgs mass the $h\\to\\phi\\phi$ searches cut the space; this gives a concrete 60-125 GeV window for future searches.","The charged scalar must be heavier than about 95 GeV, and in the 95-120 GeV slice its leptonic alignment parameter can be as large as roughly 100, so tauonic charged-Higgs searches in that slice have unique discovery power.","Whenever at least one new scalar is light, the heavier scalars are pushed below about 560 GeV, and if both neutral scalars are light the charged scalar stays below 226 GeV, setting explicit mass targets for collider searches.","A light-scalar explanation of the 2020 $(g-2)_\\mu$ anomaly is strongly disfavoured, whereas the lattice-based prediction is fully consistent with the same fits."],"supporting_citations":[{"why":"Defines the alignment condition $Y_f=\\varsigma_f M_f$ that gives the A2HDM its name and removes tree-level flavour-changing neutral currents.","marker":"[51]"},{"why":"Establishes the Bayesian global-fit framework for the A2HDM that this work extends to light scalars.","marker":"[99]"},{"why":"Supplies the updated heavy-scalar global fit, the oblique-parameter inputs with $R_b$ removed, and the CKM refit procedure reused here.","marker":"[100]"},{"why":"Provides the open-source HEPfit code and its Markov-chain Monte Carlo engine on which all fits run.","marker":"[102]"},{"why":"Derives the one-loop corrected cubic couplings whose perturbativity constraint shapes the allowed $\\lambda_3$, $\\lambda_7$ regions.","marker":"[55]"},{"why":"Computes charged-Higgs contributions to flavour observables, giving the constraints that force $\\varsigma_u$, $\\varsigma_d$ small for light $M_{H^\\pm}$.","marker":"[57]"},{"why":"Motivates the reinterpretation of slepton searches as charged-Higgs searches and reviews $(g-2)_\\mu$ interpretations in two-Higgs-doublet models.","marker":"[139]"},{"why":"CMS stau search that is the only slepton constraint acting on the light charged scalar, binding for $M_{H^\\pm}>115$ GeV.","marker":"[171]"},{"why":"The 2020 white-paper Standard Model value of $(g-2)_\\mu$ used to quantify the tension when included in a separate fit.","marker":"[181]"},{"why":"The BMW lattice value of hadronic vacuum polarisation used to show the negligible impact when the Standard Model prediction agrees with experiment.","marker":"[184]"}],"fun_headline_variants":["All seven light-scalar scenarios pass global fit","Light scalars: all seven scenarios survive data","A2HDM: light scalars down to 60 GeV still allowed","Global fit keeps all seven light-scalar scenarios alive","Seven light-scalar arrangements still fit the data"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire fit's treatment of upper limits assumes that each experimental exclusion can be converted into a truncated Gaussian likelihood centred at zero, with the width chosen so that a prediction equal to the published limit is excluded at 95% probability; the detailed shape and size of every allowed region depends on this mapping.","fun_headline_variants_meta":{"raw":{"variants":["All seven light-scalar scenarios pass global fit","Light scalars: all seven scenarios survive data","A2HDM: light scalars down to 60 GeV still allowed","Global fit keeps all seven light-scalar scenarios alive","Seven light-scalar arrangements still fit the data"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000524,"raw_usage":{"total_tokens":2530,"prompt_tokens":938,"completion_tokens":1592,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":554,"completion_tokens_details":{"reasoning_tokens":1511}},"tokens_in":554,"tokens_out":1592,"duration_ms":10469,"temperature":1.0,"reasoning_tokens":1511,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T11:48:55.930675+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A dedicated LHC search for $H^\\pm\\to\\tau^\\pm\\nu$ with $M_{H^\\pm}$ between 95 and 120 GeV, with sensitivity to the $\\sigma\\cdot B$ values predicted by this paper's allowed regions, would either confirm or close the large-$\\varsigma_l$ window; equally, an experiment that excludes $\\sigma\\cdot B$ below the A2HDM prediction across that entire window would falsify the claim that the light-charged-scalar scenarios remain viable.","supporting_citations":[],"review_version":1}