{"id":"db4499a8-a172-4193-9c51-c750dfa65344","arxiv_id":"2411.18859","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new 2HDM renormalization scheme fixes the mixing-angle counterterms so that h to ZZ* and h to tau tau NLO rates equal the tree-level scaled SM rates, giving unambiguous NLO predictions for other Higgs decays.","lead":"This paper presents a renormalization scheme for two Higgs doublet models that defines the Higgs mixing angles by requiring two measured decay channels to match their tree-level scaled Standard Model predictions. The scheme offers a practical way to make unambiguous next-to-leading-order predictions for other Higgs decay channels at future lepton colliders.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The two input conditions fix δα and δβ only if the 2×2 system is nonsingular; the large Type-X h→bb 'prediction' near alignment could be a scheme artifact rather than physics.","rationale":"The paper is a proceedings summary of a published longer article, and its central idea—fixing mixing-angle counterterms by requiring two NLO decay rates to equal κ² times the SM NLO rates—is a legitimate, process-defined renormalization scheme. The input channels are indeed fixed by construction, so the predictive content lives in the other channels, especially h→bb in Type-X. The reader's weakest assumption (existence and uniqueness of the counterterm solution) is the right place to look. My stress-test sharpens it: the concern is not just algebraic solvability but whether the solution remains perturbative and nonsingular in the alignment limit, where the Type-X bb deviation becomes large. The paper's own explanation involving tan(β−α) enhancement is suspect because tan(β−α)=c/s is small when c≪1, so independent checking is warranted. Lack of self-contained formulas is secondary because Ref. [8] is published; credit is due to the public H-COUP version 3, which makes the numerics reproducible in principle. Overall, the conditional verdict is appropriate: the scheme is plausible and potentially useful, but the key demonstration should be backed by a diagnostic of the counterterm solution. I therefore see no reason to change the reader's verdict.","tokens_in":8069,"tokens_out":8593,"duration_ms":90368,"concrete_test":"Using the explicit expressions in Ref. [8], compute the 2×2 Jacobian M = ∂(Δ^{Zℓℓ}_{EW}, Δ^{τ}_{EW})/∂(δ(β−α), δβ) and the solved counterterms δα and δβ for the same M2 scan as Fig. 2, focusing on small δ. If det M stays bounded away from zero and δα, δβ remain perturbative with no 1/c_{β−α} growth, the Type-X h→bb curve is a physical prediction; if det M→0 or the counterterms diverge, the scheme is singular and the central demonstration fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim stands or falls on whether Eqs. (17)–(18) actually determine δα and δβ, and on whether the other channels are then genuine predictions. This proceedings text does not prove existence or uniqueness; it refers to Ref. [8] for the explicit δ(β−α) and δβ expressions. The concrete risk is that the system becomes nearly singular in the alignment limit, so the solved counterterms grow as c_{β−α}→0 and force the input channels to match by large cancellations. The paper's own explanation of the striking Type-X h→bb deviation in Fig. 2 is that a δZh term is 'enhanced by the factor of tan(β−α)' for c_{β−α}≪1, but tan(β−α)=c_{β−α}/s_{β−α} is small, not enhanced, in that limit; the mechanism is at least not transparent. If δβ and δ(β−α) become large or nearly cancel the physical loop corrections, then ΔR(h→bb) in Type-X measures the scheme's normalization rather than the 2HDM. The scan obeys unitarity and vacuum-stability constraints, but no plot of δα, δβ, or of the determinant of the linear system is provided, so this cannot be checked from the manuscript alone.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript proposes a new renormalization scheme for the mixing angles α and β in the 2HDM with a softly-broken Z2 symmetry. The counterterms δα and δβ are fixed by requiring that the NLO partial widths of h → Zℓ⁺ℓ⁻ and h → τ⁺τ⁻ are equal to the SM NLO widths multiplied by the squared tree-level scaling factors (κ_V)² and (κ_τ)² (Eqs. (17) and (18)). The scheme is implemented in the H-COUP package, and NLO predictions for h → WW*, h → bbar b, etc. are shown for Type-I and Type-X 2HDMs, comparing with the earlier KOSY scheme. The paper is a proceedings contribution; detailed expressions for the counterterms are deferred to Ref. [8].","tokens_in":8257,"tokens_out":4652,"duration_ms":39091,"significance":"If the proposed scheme is consistent, it provides a practical loop-level definition of the mixing angles that keeps the tree-level κ-structure intact for the input channels, enabling future Higgs factories to use measured h → ZZ and h → ττ rates as inputs and to predict other decay modes at NLO. The use of the public H-COUP code and the clear statement of the renormalization conditions are strengths. The main risk is whether the two conditions in Eqs. (17) and (18) actually determine δα and δβ without large or singular solutions, particularly near the alignment limit; this is not demonstrated in the present text.","major_comments":[{"comment":"The central claim that the conditions (17) and (18) uniquely determine δα and δβ requires that the two equations are independent and that a solution exists. The manuscript does not give the explicit expressions for δ(β−α) and δβ, nor does it provide any information about the determinant of the 2×2 linear system that must be inverted; it only refers to Ref. [8]. Since the two input channels are forced to match by construction, the predictive status of the other channels in Figs. 1 and 2 depends on the counterterms not being driven to large or singular values by a near-degenerate system. Please provide evidence for the existence and uniqueness of the solution across the scanned parameter space, for instance plots of δα, δβ, δ(β−α), or of the determinant, especially near c_{β−α} → 0.","section":"Section 3, Eqs. (17) and (18)"},{"comment":"The explanation of the large Type-X h → bbar b deviation is internally inconsistent. The text states that a δZh contribution is 'enhanced by the factor of tan(β−α)' in the nearly alignment case c_{β−α} ≪ 1, but tan(β−α) = c_{β−α}/s_{β−α} is small, not enhanced, in that limit. Either the expression is a typo (perhaps cot(β−α) or 1/tan(β−α) is meant) or the stated mechanism is incorrect. As written, the explanation does not account for the numerical result, and this is the only explanation offered for a striking prediction of the scheme. Please correct the parametric statement and clarify the actual mechanism.","section":"Section 3, paragraph after Fig. 2"},{"comment":"The numerical results are shown only for one benchmark point (mH± = mH = mA = 300 GeV, tanβ = 2, cos(β−α) > 0) with M2 scanned. Given that the main message is the predictive power of the new scheme, the robustness of the results would be significantly strengthened by showing at least one additional benchmark with a different tanβ or different additional-Higgs masses, or by explicitly stating that such scans are presented in Ref. [8]. This is a request for more evidence, not a claim that the current scan is incorrect.","section":"Section 3, Figs. 1 and 2"}],"minor_comments":[{"comment":"There is a typo: 'renormlization' should be 'renormalization'.","section":"Conclusions"},{"comment":"The sign convention for c_{β−α} should be stated explicitly. The text later enforces cos(β−α) > 0 in the numerical scans, but the equations as written hold for either sign, and the κ factors change sign accordingly.","section":"Eq. (19)"},{"comment":"The sentence 'the former (latter) includes the counterterm δ(β−α) (δβ and δ(β−α))' is ambiguous. Please specify explicitly that Δ^{Zℓℓ}_{EW} depends on δ(β−α) and that Δ^{τ}_{EW} depends on both δβ and δ(β−α), to avoid confusion about which channel receives which counterterm.","section":"After Eq. (22)"},{"comment":"Since the proceedings relies on Ref. [8] for the detailed expressions, it would be helpful to include a short sentence summarizing the structure of the solution (e.g., that the two conditions yield a 2×2 linear system whose coefficients are known functions of the input parameters) so that the reader can assess the existence issue without consulting the external reference.","section":"Section 3, Eq. (17)–(18)"}],"recommendation":"major_revision","confidential_remarks":"This is a proceedings contribution that summarizes Ref. [8] (Phys. Lett. B 858, 139050), where the full technical derivation presumably appears. The present text is therefore not expected to contain all derivations, but the two points raised in the major comments are central to the credibility of the numerical demonstration and should be addressed in the proceedings itself: evidence that the counterterm system is solvable, and a correct explanation of the Type-X h→bb behavior. If these are fixed, the paper would be a useful and appropriately scoped contribution."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a proceedings summary of the authors' PLB paper [8] that defines a new renormalization scheme for the 2HDM mixing angles α and β. The core idea is sensible—fix δα and δβ by requiring the NLO rates for h→ZZ* and h→ττ to match the tree-level κ-scaled SM rates—and the implementation in H-COUP with parameter scans under unitarity and vacuum stability is solid. What this submission adds beyond [8] is negligible; it is a compact presentation, not new physics.\n\nThe scheme itself is a legitimate renormalization choice. The two input channels are inputs by construction, which is the nature of renormalization conditions; the other decays, h→WW* and h→bb, are genuine predictions. The Type-X h→bb deviation near alignment is the most interesting output. The paper attributes it to δZh enhanced by tan(β−α); that is correct since tan(β−α) becomes large when c_{β−α}→0. The stress-test note's worry that tan is small in that limit is a mistake—cot would be small. So that particular objection does not land.\n\nThe real soft spots are two. First, the paper is not self-contained: the explicit counterterm expressions are only referenced to [8]. For a proceedings, that is acceptable, but it means the reader cannot verify the scheme from this text alone. Second, the existence and uniqueness of the solution for δα and δβ from Eqs. (17)-(18) is assumed, not shown. If the two conditions become linearly dependent in some limit, the scheme would fail. The numerical scans suggest solutions exist for the points shown, but the paper shows no plot of the counterterms or of the determinant of the 2×2 system. That would be a cheap and informative addition. The concern is moderate, not fatal—the results look plausible—but without it, one cannot tell whether the large Type-X bb effect is a robust prediction or an artifact of large counterterm cancellations.\n\nWho is this for? Researchers doing precision 2HDM phenomenology for future lepton colliders. It is a useful entry point to the scheme, but the cited PLB paper is the actual reference. I would not cite this proceedings in my own work; I would cite [8] and the H-COUP papers. For peer review, I would send it to a referee if the venue requires one, because the underlying scheme is meaningful and the summary is accurate, but the lack of self-containedness should be flagged. For a proceedings, it is fine as is.","headline":"Sensible proceedings summary of a useful 2HDM renormalization scheme, but not self-contained; the two defining channels are inputs, and the existence/uniqueness of the counterterm solution is not demonstrated.","tokens_in":8874,"tokens_out":3190,"would_cite":false,"duration_ms":28202,"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":"This paper proposes a renormalization scheme for the two-Higgs-doublet mixing angles in which the counterterms δα and δβ are fixed by demanding that the next-to-leading-order decay rates of h→ZZ* and h→ττ equal the SM rates times the…","keywords":["two Higgs doublet model","renormalization scheme","mixing angle counterterms","Higgs precision measurements","electroweak radiative corrections","scaling factors","alignment limit","future lepton colliders"],"falsifier":"Compute the two equations (22) as functions of δα and δβ across the 2HDM parameter space allowed by unitarity and vacuum stability; any parameter point where the Jacobian determinant with respect to δα and δβ vanishes, or where no real solution exists, would show the scheme is not globally defined. The Type-X near-alignment region is the natural place to look, because the ττ equation's δZ_h term changes with tan(β−α).","tokens_in":7779,"feed_emoji":"⚛️","tokens_out":7588,"duration_ms":63479,"temperature":0.7,"pith_summary":"The paper aims to define the two mixing angles α and β of the two-Higgs-doublet model at one-loop order in a way that keeps their tree-level physical meaning: how close the 125 GeV Higgs is to the Standard Model Higgs. It proposes fixing the counterterms δα and δβ by requiring that the NLO decay rates of h→ZZ*→Zℓ+ℓ− and h→τ+τ− equal the corresponding SM NLO rates multiplied by the squared scaling factors κ_V² and κ_τ². Because these two channels are used as inputs, the NLO predictions for other channels—such as h→WW*, h→bb, and h→ττ itself—become well-defined predictions of the scheme. The motivation is that future Higgs factories will measure Higgs couplings to better than one percent, and a scheme of this type lets measured deviations be directly mapped onto tree-level κ factors without contamination from loop corrections.","feed_headline":"Two Higgs decays set the loop-level mixing angles","feed_subtitle":"Future Higgs-factory data on h→ZZ* and h→ττ become inputs that pin down the other NLO rates.","key_machinery":"The load-bearing object is the pair of equations (17)–(18) (rewritten as (22)), which impose that the electroweak corrections to the two selected decay channels match the SM electroweak corrections; solving these two equations fixes δα and δβ. The rest of the counterterms are handled by standard on-shell conditions supplemented by the alternative tadpole scheme, and the scheme's numerical behavior is exhibited by scanning M² under perturbative-unitarity and vacuum-stability constraints. The mechanism that produces the distinctive Type-X h→bb effect is the appearance of δZ_h multiplied by tan(β−α) after substituting the solved counterterms, which becomes large as cos(β−α)→0.","core_discovery":"The central claim is that the two mixing-angle counterterms can be determined by the two renormalization conditions Γ(h→Zℓℓ)^NLO = (κ_V)² Γ(h→Zℓℓ)^NLO_SM and Γ(h→ττ)^NLO = (κ_τ)² Γ(h→ττ)^NLO_SM, equivalently $Δ^{{Zℓℓ}}$_EW = $Δ^{{Zℓℓ}}$_EW|_SM and $Δ^{{τ}}$_EW = $Δ^{{τ}}$_EW|_SM. In the near-alignment regime this scheme keeps the NLO decay rates of h→ZZ* and h→WW* close to their tree-level κ-factor description, and in Type-I models the fermionic rates h→ff follow κ_f² as well. In Type-X models, by contrast, h→bb can receive a large NLO correction near the alignment limit, traced to the δZ_h wavefunction counterterm entering with a tan(β−α) enhancement when ζ_τ and ζ_f differ.","pith_inferences":["Choosing different input channels (for instance h→WW* and h→bb instead of h→ZZ* and h→ττ) would define a different, but equally consistent, scheme; predictions for the non-input channels would shift by finite one-loop terms, so any κ extraction from data is only meaningful once the input pair is specified.","The near-alignment Type-X result suggests that in models with flavor-dependent ζ_f, the tree-level κ parametrization can break down at the loop level for fermionic channels, so precision fits should either use NLO-defined κ factors or include the scheme dependence explicitly.","The same input-channel strategy could be transplanted to other extended Higgs sectors (e.g., models with more doublets or with triplet representations) whenever the number of mixing parameters equals the number of channels that future colliders can measure cleanly, provided the one-loop corrections depend on the counterterms in a non-degenerate way."],"forward_implications":["In the new scheme, ΔR(h→ZZ*) and ΔR(h→ττ) coincide with the LO κ-factor predictions by construction, so future measurements of these channels can be used as inputs rather than tests.","In Type-I 2HDMs, the NLO predictions for h→WW* and h→bb remain close to the tree-level κ² scaling, so the κ description of these channels survives at one loop.","In Type-X 2HDMs, h→bb at NLO can deviate strongly from its tree-level κ_b² prediction near the alignment limit, especially at smaller cos(β−α) and small M²/v².","Because the scheme fixes the mixing counterterms by physical inputs rather than by wavefunction-renormalization conventions, it removes a major source of scheme ambiguity in NLO 2HDM Higgs-precision calculations."],"supporting_citations":[{"why":"Sets out the full formulation of the new scheme and gives the explicit expressions for δ(β−α) and δβ used here.","marker":"[8]"},{"why":"Provides the earlier mixing-counterterm scheme against which the new scheme's predictions are compared.","marker":"[18]"},{"why":"Defines the Standard Model electroweak renormalization used to fix δv.","marker":"[17]"},{"why":"Supplies the alternative tadpole scheme that avoids explicit tadpole counterterms.","marker":"[15, 16]"},{"why":"Defines the four Yukawa types and the ζ_f factors entering κ_τ and the fermionic decay rates.","marker":"[12–14]"},{"why":"Provides the numerical NLO decay-rate computation used to demonstrate the scheme.","marker":"[25–27]"}],"fun_headline_variants":["Two decays dictate loop-level mixing angles","Renormalization from h→ZZ* and h→ττ","NLO mixing angles from Higgs decay inputs","New scheme keeps NLO rates close to tree-level"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The scheme assumes that the two conditions (17) and (18) can actually be solved for δα and δβ, meaning the two equations are independent and consistent at every parameter point where the scheme is used.","fun_headline_variants_meta":{"raw":{"variants":["Two decays dictate loop-level mixing angles","Renormalization from h→ZZ* and h→ττ","NLO mixing angles from Higgs decay inputs","New scheme keeps NLO rates close to tree-level"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00025,"raw_usage":{"total_tokens":1548,"prompt_tokens":935,"completion_tokens":613,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":551,"completion_tokens_details":{"reasoning_tokens":552}},"tokens_in":551,"tokens_out":613,"duration_ms":6142,"temperature":1.0,"reasoning_tokens":552,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:48:48.971805+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the two equations (22) as functions of δα and δβ across the 2HDM parameter space allowed by unitarity and vacuum stability; any parameter point where the Jacobian determinant with respect to δα and δβ vanishes, or where no real solution exists, would show the scheme is not globally defined. The Type-X near-alignment region is the natural place to look, because the ττ equation's δZ_h term changes with tan(β−α).","supporting_citations":[],"review_version":1}