{"id":"18f031eb-1434-407a-9adc-10a7b799d15d","arxiv_id":"hep-ph/0512103","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A four-dimensional supersymmetric A4 model reproduces tri-bimaximal neutrino mixing naturally, but the same A4 structure applied to quarks cannot generate the observed quark mixing.","lead":"The authors build a four-dimensional supersymmetric model in which the discrete symmetry A4 forces neutrino mixing into the tri-bimaximal pattern, and they show the leading corrections stay small over a stated range of parameters. They also note that A4 is a quotient of the modular group, and that extending the same A4 assignments to quarks leaves the quark mixing matrix too close to the identity.","discovery_kind":"extension","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"Eq. (40) for |U_e3| appears dimensionally inconsistent: unlike eqs. (41)-(42) it carries no denominator, so as printed it yields |U_e3| ~ (v/Λ)^5 rather than the ~v/Λ level the text claims for all corrections.","rationale":"The reader's weakest_assumption — the assumed stabilization of u and the U(1)_F flavon — is correct as far as it goes, but Section 4 explicitly states both assumptions ('we assume to be eventually stabilized by one-loop radiative corrections'; 'We implicitly assume that such field(s) σ ... are included'), so it is an acknowledged gap rather than an unacknowledged error. The more concrete, unacknowledged issue is the normalization of eq. (40): the correction formulas (41)-(42) are dimensionally consistent with the text's ~v/Λ claim, but (40) is not, suggesting a missing denominator. This is a fixable technical error, not a structural flaw, so it does not change the CONDITIONAL verdict; it does mean the quantitative θ13 prediction should be re-checked before the NLO results are trusted. The 'completely natural' wording in abstract:s0 is also in mild tension with the conclusion's admitted 'moderate fine tuning' for r, but the authors flag this, so it is a framing issue rather than a new finding.","tokens_in":20206,"tokens_out":13102,"duration_ms":239093,"concrete_test":"Re-derive |U_e3| from eqs. (36)-(39) by diagonalizing m_ν+δm_ν and m_l†m_l to first order in δz_i and δv_T, keeping all denominators. Check whether the result carries a factor 1/(āb+b̄a-|b|²) (or its square) and whether the coefficient matches eq. (40). If the re-derived expression differs from (40) by ~(v/Λ)⁴, the printed formula is wrong and the θ13 prediction must be corrected; if it matches, the concern does not land.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central quantitative claim is that NLO corrections stay at the ~v/Λ level (text: 'deviations ... at approximately the same level ... not larger than about 0.05'). Eqs. (41)-(42) are consistent with this: their correction terms have denominators ~(āb+b̄a±|b|²) ~ (v/Λ)² and numerators ~(v/Λ)³, giving ~v/Λ. Eq. (40) for |U_e3| has the same numerator structure but no denominator. With a,b ~ v/Λ and δz_i ~ (v/Λ)² (eqs. 37-38, 77), the printed bracket is ~(v/Λ)⁵, i.e. |U_e3| ~ 10⁻⁶ for v/Λ ~ 0.03 — four to five orders below the claimed level. Either a denominator ~(āb+b̄a-|b|²)² is missing or the formula is wrong. This matters because θ13 (|U_e3|) is the sharpest TBM prediction and the model's falsifiability rests on its NLO size. The reader's concern — that the flat direction u is only 'assumed' to be stabilized (Section 4) — is real but explicitly acknowledged by the authors, so it is not a new finding; it compounds the eq. (40) issue because the correction size scales with u/Λ.","agreement_with_reader":"partial"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript presents a four-dimensional supersymmetric version of the A4 model for tri-bimaximal (Harrison-Perkins-Scott) neutrino mixing. Left-handed lepton doublets are assigned to a triplet and the right-handed charged leptons to 1, 1'' and 1', with the flavour symmetry broken by two triplets φ_T and φ_S and by singlets ξ and ξ̃; a Z3 × U(1)_R symmetry forbids the unwanted (ll) and φ_T ↔ φ_S terms. Driving fields φ_T0, φ_S0 and ξ0 generate a superpotential whose F-term conditions align ⟨φ_T⟩ = (v_T,0,0) and ⟨φ_S⟩ = (v_S,v_S,v_S), giving diagonal charged-lepton masses and the HPS matrix at leading order. The authors classify the next-to-leading operators, recompute the shifted vacuum in Appendix B, and derive the resulting corrections to |U_e3|, tan²θ12 and tan²θ23 (eqs. 40-42). They also present A4 as a subgroup of the modular group PSL(2,Z), classify the three-dimensional unitary irreps in Appendix A, and show that extending the A4 assignment to quarks gives a diagonal CKM matrix whose next-to-leading corrections are too small to reproduce the observed mixing.","tokens_in":20589,"tokens_out":32554,"duration_ms":576047,"significance":"If correct, the paper provides a natural four-dimensional alternative to the extra-dimensional construction of ref. [8], with an explicit and controlled 1/Λ expansion. I checked the leading-order mass matrices (15)-(16) and the eigenvalues (20), and I verified that the correction matrix (36) reproduces the eigenvalues (39) under the TBM eigenvectors; the charged-lepton rotation (35) is also correct, since the off-diagonal entries of m'†m' are not suppressed by y_e/y_μ. The quark-sector analysis is a useful negative result: it shows that the same A4 classification cannot generate CKM mixing, which is a real constraint on unified A4 constructions. The modular-group observation is a representation-theory remark rather than a derivation of A4, and the authors present it as such.","major_comments":[{"comment":"The naturalness claim in the abstract (abstract:s0, 'completely natural') rests on two stabilizations that the paper assumes rather than computes: the flat direction u of eq. (29), for which the text says 'u slides to a large scale, which we assume to be eventually stabilized by one-loop radiative corrections', and the U(1)_F flavon VEV ⟨θ⟩, for which the text says 'we implicitly assume that such field(s) σ,... are included'. The authors acknowledge both, so this is not a hidden gap, but the abstract's unqualified 'completely natural' is stronger than what is derived. Either the one-loop stabilization should be sketched, at least enough to show that a minimum exists with u/Λ inside the window of eq. (43), or the abstract and Section 9 should state that naturalness is conditional on it.","section":"Section 4, eq. (29) and final paragraph","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The upper bound 0.05 in eq. (43) is not a prediction but a self-imposed consistency requirement: the text says 'if we require that the subleading terms do not spoil the leading order picture, these deviations should not be larger than about 0.05'. The lower bound is also tan β-dependent: the text quotes v_T/Λ > 0.0022 at tan β = 2.5 and 0.024 at tan β = 30, so at large tan β the allowed window is only a factor of about two. Since the naturalness argument rests on the window being comfortably wide, the paper should present eq. (43) explicitly as a consistency window as a function of tan β, and should not let the 0.05 end read as a derived number.","section":"Section 5.4, eq. (43)","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The factors printed as '13', '12' and '14' are the paper's flattened notation for 1/3, 1/2 and 1/4 (compare eq. (9) and eq. (28)), and in eqs. (40)-(42) they must be read as denominators. With that reading |U_e3| scales as (v/Λ)³/(v/Λ)² = v/Λ, consistent with the text; read as multiplicative prefactors the result would be (v/Λ)⁵, four orders below the claimed level. Please make sure the typeset fractions are unambiguous.","section":"Eqs. (40)-(42)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"With B = −2A + z and the eigenvalues (48), the quoted |m_2|² ≈ Δm²_atm(9/8 + 13r/12) and |m_3|² ≈ Δm²_atm(1/8 + r/12) are correct, but |m_1|² ≈ Δm²_atm(9/8 + 11r/12) is not: the same expansion gives (9/8 + r/12), and the printed coefficient would imply Δm²_sol = |m_2|² − |m_1|² = r Δm²_atm/6 rather than r Δm²_atm.","section":"Eq. (49)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"For B = A + z the quoted |m_1|² ≈ (r/3)Δm²_atm and |m_2|² ≈ (4r/3)Δm²_atm are correct, but |m_3|² ≈ Δm²_atm(1 − r/3) should be (1 + r/3); with the printed sign, Δm²_atm = |m_3|² − |m_1|² would not reproduce Δm²_atm.","section":"Eq. (50)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The operator (30) is of order 1/Λ², but after inserting ⟨φ_T⟩ = v_T its contribution to m_l is of relative order v_T/Λ, not 1/Λ²; what is true is that it does not introduce a new flavour structure. Please rephrase to avoid confusion with the relative-order counting used in Section 5.3.","section":"Section 5.1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The text says the corrections 'can be kept small, below the percent level', but the bound subsequently adopted in eq. (43) is 0.05, i.e. five percent. Please reconcile the two statements.","section":"Section 5.4","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Eq. (22) quotes 0.004 < v_S/Λ ≈ v_T/Λ ≈ u/Λ < 1, while eq. (43) quotes 0.0022 < ... < 0.05; the text explains the change of lower bound but eq. (22) is left standing. Please update eq. (22) or add a forward reference.","section":"Eq. (22) versus eq. (43)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The see-saw table assigns Z3 charge ω² to φ_S, ξ and ξ̃, whereas the table of Section 4 assigns them ω. The statement that the driving superpotential is 'unchanged' is correct in form, but the reader should be told that the charges have changed and that the invariants still close.","section":"Section 6, table","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The caption describes six allowed regions and a star marking the representation used in the model, but the figure as reproduced is not legible at this size. Please check the reproduction.","section":"Figure 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The parenthetical argument that no two-dimensional unitary representation exists ('only det(T³) = −1 is in this case compatible with S² = (ST)³ = 1') is elliptically worded; a one-line expansion would help.","section":"Section 2","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The modular-group connection is a representation-theory observation, not a derivation of A4, and the authors present it as such; consider whether the title's 'and the Modular Symmetry' overstates its role in the construction.","section":"Title and Section 8","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"The paper is a follow-up to ref. [8]; the genuinely new content is the four-dimensional SUSY alignment with driving fields, the explicit next-to-leading analysis, and the modular-group observation, which the authors appropriately flag as speculative. The quark-sector negative result is a useful addition. I checked the concern that eq. (40) is dimensionally inconsistent with the claimed ~v/Λ correction level; it is not, provided the printed '13' and '12' are read as denominators, which is the paper's convention elsewhere. The remaining issues are local and fixable."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Altarelli and Feruglio take their extra-dimensional A4 model and rebuild it in four dimensions with SUSY. That is the substance: driving fields, a Z3 x U(1)_R to forbid the unwanted (ll) and phi_T <-> phi_S terms, and an explicit F-term minimization that lands on (1,0,0) and (1,1,1). The alignment is clean, and the correction analysis in Sec. 5 and App. B is done properly — every higher-dimension operator is classified and its effect on the mixing angles computed. The quark section is a straight negative result: the same A4 assignments give CKM = 1 at leading order, and the NLO corrections cancel between up and down sectors, so new breaking is needed. They say so rather than burying it. The modular-group section (App. A) is a nice side result — a classification of three-dimensional unitary irreps of PSL(2,Z), with A4 sitting at the symmetric point (beta_2, beta_3) = (2/3, 1/3). That is the part that aged best; modular flavor symmetry became a whole industry a decade later. Soft spots. Eq. (40) for |U_e3| looks dimensionally wrong as printed: with a, b ~ v/Lambda and delta z ~ (v/Lambda)^2, the bracket is ~(v/Lambda)^5, so |U_e3| ~ 10^-7 rather than the ~v/Lambda the text claims for all corrections. Eqs. (41)-(42) have the right structure — the theta_23 formula carries a 1/(abar b + bbar a) denominator. Almost certainly a missing denominator in (40), possibly a transcription slip, but theta_13 is the sharpest TBM prediction, so it should be checked against the published version. Second, the naturalness claim rides on u being stabilized at a large scale by one-loop effects; the authors flag this explicitly (we assume), so it is not a hidden gap, but it is the load-bearing assumption, and the same goes for the U(1)_F flavon. Third, the upper bound 0.05 in eq. (43) is chosen because larger corrections spoil the leading-order picture — a consistency requirement, not a prediction. Worth a referee report? Yes. It is careful, self-aware model building with a real negative result and a genuinely useful math appendix. And note the obvious: the leading-order TBM prediction is now excluded by theta_13, so read it as a 2005 snapshot and as the ancestor of the modular-symmetry program, not as a live model.","headline":"A careful, honest 4D SUSY A4 model with a genuinely useful modular-group appendix — but eq. (40) for |U_e3| looks dimensionally wrong as printed and should be checked.","tokens_in":21143,"tokens_out":3329,"would_cite":false,"duration_ms":55360,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["14.60.Pq","11.30.Hv","12.60.Jv"],"model":"selfhosted/deepseek-v4.1-flash","headline":"A four-dimensional supersymmetric model in which a broken A4 flavour symmetry produces tri-bimaximal neutrino mixing, with controlled corrections and a quark-sector mismatch.","keywords":["tri-bimaximal neutrino mixing","A4 discrete flavour symmetry","neutrino mass matrix","vacuum alignment","supersymmetric flavour model","modular group","CKM matrix","flavon"],"falsifier":"A measurement of the reactor mixing angle $|U_{e3}|$ substantially above the correction size the model allows, or a value of $tan^2 θ_{12}$ differing from $1/2$ by more than the percent-level corrections computed in eqs. (40)–(42), would break the leading-order tri-bimaximal prediction; conversely, confirmation of $|U_{e3}|$ at or below the flavon-to-cutoff ratio would support it.","tokens_in":21598,"feed_emoji":"⚛️","tokens_out":3156,"duration_ms":51487,"temperature":0.7,"pith_summary":"The paper asks whether the observed neutrino mixing pattern — the tri-bimaximal matrix of Harrison, Perkins and Scott — can be produced by a broken discrete flavour symmetry in four flat dimensions with supersymmetry, rather than by the extra-dimensional construction the authors used earlier. It answers yes: two triplet flavons, $A_4$ fields $S$ and $T$, are aligned by a supersymmetric driving sector so that one points along $(1,0,0)$ and the other along $(1,1,1)$, which makes the charged-lepton mass matrix diagonal and gives a neutrino mass matrix whose diagonalizing matrix is exactly the tri-bimaximal one. The authors then enumerate every higher-dimensional operator allowed by the symmetries and show that, once the ratios of symmetry-breaking vacuum expectation values to the cutoff sit in the window $0.0022 < v_S/Λ ≈ v_T/Λ ≈ u/Λ < 0.05$, all corrections stay below the percent level. A second thread re-expresses $A_4$ through the two generators of the modular group, so that the basis in which the symmetry is written is already the basis in which charged leptons are diagonal. The last thread is a negative result: extending the same $A_4$ assignment to quarks makes the CKM matrix the identity at leading order and leaves corrections far too small to reproduce the observed quark mixing, so quarks need new sources of symmetry breaking.","feed_headline":"A4 symmetry yields tri-bimaximal neutrino mixing in 4D","feed_subtitle":"Broken A4 flavour symmetry plus supersymmetric vacuum alignment reproduces the observed neutrino mixing pattern.","key_machinery":"The load-bearing object is the pair of $A_4$ triplet flavons and their vacuum alignment. $A_4$ is presented through two generators $S$ and $T$ obeying $S^2 = (ST)^3 = T^3 = 1$, with $T$ diagonal and $S$ the symmetric matrix with $-1$ on the diagonal and $2$ off it (divided by three). A triplet vacuum along $(1,0,0)$ preserves the $Z_3$ generated by $T$; a triplet vacuum along $(1,1,1)$ preserves the reflection generated by $S$. The driving superpotential, linear in the driving fields $ϕ_T^0$, $ϕ_S^0$ and $ξ^0$ and written with all contractions allowed by $A_4$, $Z_3$ and $U(1)_R$, has its minimum exactly at those two directions, with an extra singlet $ξ$ required so that the $ϕ_S$ alignment has a non-trivial solution at all. The same $S$–$T$ presentation is what connects the construction to the modular group, whose defining relations are $s^2 = (st)^3 = 1$ with $t$ of infinite order; $A_4$ is the finite group obtained by truncating $t$ to order three, and the paper classifies the three-dimensional unitary representations of the modular group to show where the $A_4$ representation sits among them.","core_discovery":"The central claim is that a four-dimensional supersymmetric $A_4$ model with a driving-field sector, a $Z_3$ symmetry and a continuous $U(1)_R$ symmetry reproduces the tri-bimaximal mixing matrix at leading order without extra dimensions, and that it is natural in the technical sense: no parameter must be tuned to obtain the mixing angles. The mechanism is vacuum alignment. The driving superpotential forces the flavon $S$ to $(v_S,v_S,v_S)$ and the flavon $T$ to $(v_T,0,0)$, up to equivalent $A_4$ images. The first direction leaves unbroken the subgroup generated by $S$, which is the low-energy symmetry of the neutrino sector; the second leaves unbroken the subgroup generated by $T$, which is the low-energy symmetry of the charged-lepton sector. The resulting charged-lepton mass matrix is diagonal and the neutrino mass matrix has eigenvalues $a+b$, $a$, $-a+b$ with the tri-bimaximal matrix as its diagonalizer. The authors then classify the next-to-leading operators in three groups — corrections to the charged-lepton masses, to the neutrino masses, and to the vacuum alignment — and derive the induced shifts in $|U_{e3}|$, $tan^2 θ_{12}$ and $tan^2 θ_{23}$. All are suppressed by ratios of flavon vacuum expectation values to the cutoff, so the leading-order picture survives provided those ratios lie in the stated interval. The paper also shows that the same field assignment applied to quarks yields a diagonal CKM matrix at leading order and, because the first-order vacuum corrections are nearly identical in the up and down sectors, they cancel in the mixing matrix; the departures from the identity are therefore far too small to accommodate the observed quark mixing angles.","pith_inferences":["Because the leading-order prediction is $|U_{e3}| = 0$ with corrections of order the flavon-to-cutoff ratio, a measured reactor angle near the current experimental bound would push the cutoff-suppression window to its upper edge or beyond, giving a direct handle on the scale of flavour breaking.","The same $S$–$T$ presentation that makes $A_4$ a truncation of the modular group suggests the symmetry could arise from a compactification modulus rather than being imposed by hand; the paper stops at noting the coincidence and does not build such a model.","The quark-sector failure is informative in the reverse direction: it says that any unified flavour scheme must break the symmetry differently in the up and down sectors, which is a constraint on grand-unified embeddings of this construction.","The classification of three-dimensional unitary representations of the modular group places the $A_4$ representation at a symmetric interior point of the allowed region, which invites the question of whether a dynamical principle selects that point."],"forward_implications":["The tri-bimaximal mixing pattern follows from a broken discrete symmetry plus supersymmetric vacuum alignment alone, with no extra spatial dimensions required, so the earlier extra-dimensional version and this version are two realizations of the same low-energy mixing.","Next-to-leading corrections to $|U_{e3}|$, $tan^2 θ_{12}$ and $tan^2 θ_{23}$ are all of the same parametric size, roughly the ratio of flavon vacuum expectation values to the cutoff, so a measurement of any one of them bounds the others.","The window $0.0022 < v_S/Λ ≈ v_T/Λ ≈ u/Λ < 0.05$ is fixed from two sides: the lower end by keeping the tau Yukawa coupling perturbative, the upper end by not spoiling the leading-order agreement with $Δm^2_{sol}/Δm^2_{atm}$ and $tan^2 θ_{12}$.","The construction carries over to a see-saw realization with right-handed neutrino triplets, where both normal and inverted hierarchies are attainable depending on the relative phase between the two mass parameters.","Extending the leptonic $A_4$ assignment literally to quarks gives an identity CKM matrix at leading order and first-order corrections that cancel between up and down sectors, so quark mixing must come from new symmetry-breaking sources specific to quarks."],"supporting_citations":[{"why":"Supplies the tri-bimaximal mixing matrix of Harrison, Perkins and Scott that the model is built to reproduce.","marker":"[2]"},{"why":"Earlier $A_4$ construction whose 12 group matrices, obtained by the basis change in eq. (5), the paper reproduces and extends.","marker":"[3]"},{"why":"The authors' previous extra-dimensional $A_4$ model, which sets the parameter ranges and the vacuum-alignment problem this paper solves in four dimensions.","marker":"[8]"},{"why":"Source for the presentation of $A_4$ through the generators $S$ and $T$ with $S^2 = (ST)^3 = T^3 = 1$.","marker":"[10]"},{"why":"The experimental determinations of neutrino mixing used to justify that the observed matrix is compatible with tri-bimaximal form.","marker":"[1]"},{"why":"Classification of low-dimensional unitary representations of the braid group, used in Appendix A for the three-dimensional modular-group representations.","marker":"[12]"},{"why":"Background on the modular group and duality transformations in string theory, the context in which the $A_4$-from-modular-group connection is motivated.","marker":"[9]"},{"why":"Abelian flavour models that motivate the alternative treatment of quarks through a $U(1)_F$ symmetry alone.","marker":"[11]"}],"fun_headline_variants":["A4 vacuum alignment fixes tri-bimaximal neutrino mixing in 4D","Driving fields align A4 flavons for tri-bimaximal neutrino mixing","A4 modular link sets basis with diagonal charged leptons","Next-to-leading A4 shifts stay small only in fixed VEV interval","Quark A4 extension leaves CKM too close to identity"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The naturalness claim depends on the singlet vacuum expectation value $u$, which the leading-order potential leaves undetermined, actually being stabilized at a large scale by one-loop radiative corrections and by unspecified additional fields — the paper states this as an assumption rather than deriving it, so if that stabilization fails the ratio window that keeps corrections small is not fixed.","fun_headline_variants_meta":{"raw":{"variants":["A4 vacuum alignment fixes tri-bimaximal neutrino mixing in 4D","Driving fields align A4 flavons for tri-bimaximal neutrino mixing","A4 modular link sets basis with diagonal charged leptons","Next-to-leading A4 shifts stay small only in fixed VEV interval","Quark A4 extension leaves CKM too close to identity"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1650,"prompt_tokens":1069,"completion_tokens":581,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":499}},"tokens_in":1069,"tokens_out":581,"duration_ms":10617,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":499,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-29T16:01:38.920401+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"A measurement of the reactor mixing angle $|U_{e3}|$ substantially above the correction size the model allows, or a value of $tan^2 θ_{12}$ differing from $1/2$ by more than the percent-level corrections computed in eqs. (40)–(42), would break the leading-order tri-bimaximal prediction; conversely, confirmation of $|U_{e3}|$ at or below the flavon-to-cutoff ratio would support it.","supporting_citations":[{"cited_title":"Tri-bimaximal mixing and the neutrino oscillation data","cited_arxiv_id":"hep-ph/0202074","evidence_quote":"Supplies the tri-bimaximal mixing matrix of Harrison, Perkins and Scott that the model is built to reproduce."},{"cited_title":"Softly broken <mml:math xmlns:mml=\"http://www.w3.org/1998/Math/MathML\" display=\"inline\"> <mml:mrow> <mml:msub> <mml:mrow> <mml:mi>A</mml:mi> </mml:mrow> <mml:mrow> <mml:mn>4</mml:mn> </mml:mrow> </mml:msub> </mml:mrow> </mml:math> symmetry for nearly degenerate neutrino masses","cited_arxiv_id":"hep-ph/0106291","evidence_quote":"Earlier $A_4$ construction whose 12 group matrices, obtained by the basis change in eq. (5), the paper reproduces and extends."},{"cited_title":"Tri-Bimaximal Neutrino Mixing from Discrete Symmetry in Extra Dimensions","cited_arxiv_id":"hep-ph/0504165","evidence_quote":"The authors' previous extra-dimensional $A_4$ model, which sets the parameter ranges and the vacuum-alignment problem this paper solves in four dimensions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Source for the presentation of $A_4$ through the generators $S$ and $T$ with $S^2 = (ST)^3 = T^3 = 1$."},{"cited_title":"Neutrino oscillations: Current status and prospects","cited_arxiv_id":"hep-ph/0510331","evidence_quote":"The experimental determinations of neutrino mixing used to justify that the observed matrix is compatible with tri-bimaximal form."},{"cited_title":"Tuba, Low dimensional unitary representations of B3, Proc","cited_arxiv_id":null,"evidence_quote":"Classification of low-dimensional unitary representations of the braid group, used in Appendix A for the three-dimensional modular-group representations."},{"cited_title":"Target Space Duality in String Theory","cited_arxiv_id":"hep-th/9401139","evidence_quote":"Background on the modular group and duality transformations in string theory, the context in which the $A_4$-from-modular-group connection is motivated."},{"cited_title":"Models of Neutrino Masses and Mixings","cited_arxiv_id":"hep-ph/0405048","evidence_quote":"Abelian flavour models that motivate the alternative treatment of quarks through a $U(1)_F$ symmetry alone."}],"review_version":1}