Flavour at the End of Inflation: Koide's Form, Trimaximal Mixing and a Quantised θ̄ from the Two Zero-Energy Vacua of an SL(2,Z) Plateau

v2 (19 Nov 2026): the slow-modulus basin result is in. Under a pre-registered lock, 2000 starting points distributed uniformly in hyperbolic area over the fundamental domain (y ≤ 3) were evolved with the gradient flow of the same potential. 69.0% drain to i and 31.0% to ρ; none ends elsewhere. The ρ basin is a strip around the CP-symmetric line x = ½ (the inflaton's line); along y = 3, x ∈ [0, 0.355] reaches i and [0.358, 0.5] reaches ρ. The criterion set in advance (f_i ≥ 0.25) is met, so the neutrino modulus at i is the natural outcome of the same potential (for a slow modulus obeying it) rather than an independent assumption. No other numbers changed. v1.1 (19 Nov 2026, referee pass): the statement that the potential vanishes only at ρ and i is now backed by a numerical scan of the fundamental domain (no other zeros); credit added for residual-symmetry methods in modular flavour models (Novichkov–Penedo–Petcov, JHEP 04 (2021) 206; Feruglio, arXiv:2211.00659), with the novelty limited to the dynamical reason for the two vacua. No numbers changed. The slow-modulus basin computation is still running and will appear in v2. Flavour at the end of inflation (draft, 19 Nov 2026). In the companion paper (doi:10.5281/zenodo.24377930) inflation is driven by the running of the species scale along the type IIB axio-dilaton, with V₀ ∝ |∇ ln E(τ,3/2)|². This potential vanishes at exactly two points of the fundamental domain, the SL(2,Z) fixed points ρ = e^(iπ/3) and i. We ask what the flavour sector looks like if its modular fields come to rest at these two vacua. (i) Two-field simulations show the inflaton itself ends at ρ for every transverse kick tested, although the line it falls along has a transverse instability band (0.96 ≲ y ≲ 1.9), a "Y-fork" towards i. (ii) The residual Z₃ at ρ forces a circulant √M_ℓ whose eigenvalues are exactly the Koide–Brannen form; Q = 2/3 is equal norm of the singlet and doublet components. The data give amplitude √2 and δ = 2/9 = Q/3 within 1σ; minimal A₄ weight-2 constructions exactly at ρ do not produce these numbers. (iii) Charged leptons at ρ and neutrinos at i give TM2 mixing (second PMNS column (1/3,1/3,1/3)): θ₁₃ is freed (exact-ρ tribimaximal mixing is excluded at 38σ), sin²θ₁₂ = 0.3408 (+2.8σ), and δ_CP is predicted: 236° or 125° (sin²θ₂₃ = 0.561), 287° or 74° (sin²θ₂₃ = 0.470). (iv) At ρ the A₄ quark determinant is the real number 27/8; with CP-symmetric couplings θ̄ ∈ {0, π}, but the CKM phase vanishes (|J| < 5×10⁻¹⁵), so the observed phase needs a quark modulus at a generic point with anomaly-free weights (Feruglio–Strumia–Titov). Status of the claims. Derived: Koide form from the Z₃ at ρ; Q ⇔ equipartition; TM2 from the (ρ, i) pair; δ_CP band; det = 27/8 and θ̄ quantisation; the end point ρ (two-field run). Principles (assumed): the species-gradient potential, equipartition with δ = Q/3, lepton moduli at the fixed points (neutrinos at i), anomaly-free quark weights with a generic quark modulus. Failed locked tests are reported (minimal A₄ at exact ρ; tribimaximal mixing; CKM phase at the fixed points). Whether slow moduli drain to i is under a pre-registered computation, to be reported in a revision. All computations pre-registered (locks D1595–D1607). The δ_CP values are predictions to be scored on DUNE and Hyper-Kamiokande. Not peer reviewed. License: CC BY-NC-SA 4.0 under the AEGIS Community License — free for individuals, scientists and students; any business or for-profit use needs the AEGIS Commercial License (https://github.com/dannyeldridge54/AEGIS/blob/master/licensing/COMMERCIAL_LICENSE.md). Built by one independent researcher with the help of AI. Code and pre-registration files: https://github.com/dannyeldridge54/AEGIS (branch unify-bg). Prepared with an AI assistant (Claude, Anthropic).

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-19
DOI
https://doi.org/10.5281/zenodo.24377944
Primary Topic
Particle physics theoretical and experimental studies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Flavour at the End of Inflation: Koide's Form, Trimaximal Mixing and a Quantised θ̄ from the Two Zero-Energy Vacua of an SL(2,Z) Plateau

Danny Lee Eldridge
Zenodo (CERN European Organization for Nuclear Research)
Particle physics theoretical and experimental studies
preprint

Flavour at the End of Inflation: Koide's Form, Trimaximal Mixing and a Quantised θ̄ from the Two Zero-Energy Vacua of an SL(2,Z) Plateau

Danny Lee Eldridge
preprint en

Abstract

v2 (19 Nov 2026): the slow-modulus basin result is in. Under a pre-registered lock, 2000 starting points distributed uniformly in hyperbolic area over the fundamental domain (y ≤ 3) were evolved with the gradient flow of the same potential. 69.0% drain to i and 31.0% to ρ; none ends elsewhere. The ρ basin is a strip around the CP-symmetric line x = ½ (the inflaton's line); along y = 3, x ∈ [0, 0.355] reaches i and [0.358, 0.5] reaches ρ. The criterion set in advance (f_i ≥ 0.25) is met, so the neutrino modulus at i is the natural outcome of the same potential (for a slow modulus obeying it) rather than an independent assumption. No other numbers changed. v1.1 (19 Nov 2026, referee pass): the statement that the potential vanishes only at ρ and i is now backed by a numerical scan of the fundamental domain (no other zeros); credit added for residual-symmetry methods in modular flavour models (Novichkov–Penedo–Petcov, JHEP 04 (2021) 206; Feruglio, arXiv:2211.00659), with the novelty limited to the dynamical reason for the two vacua. No numbers changed. The slow-modulus basin computation is still running and will appear in v2. Flavour at the end of inflation (draft, 19 Nov 2026). In the companion paper (doi:10.5281/zenodo.24377930) inflation is driven by the running of the species scale along the type IIB axio-dilaton, with V₀ ∝ |∇ ln E(τ,3/2)|². This potential vanishes at exactly two points of the fundamental domain, the SL(2,Z) fixed points ρ = e^(iπ/3) and i. We ask what the flavour sector looks like if its modular fields come to rest at these two vacua. (i) Two-field simulations show the inflaton itself ends at ρ for every transverse kick tested, although the line it falls along has a transverse instability band (0.96 ≲ y ≲ 1.9), a "Y-fork" towards i. (ii) The residual Z₃ at ρ forces a circulant √M_ℓ whose eigenvalues are exactly the Koide–Brannen form; Q = 2/3 is equal norm of the singlet and doublet components. The data give amplitude √2 and δ = 2/9 = Q/3 within 1σ; minimal A₄ weight-2 constructions exactly at ρ do not produce these numbers. (iii) Charged leptons at ρ and neutrinos at i give TM2 mixing (second PMNS column (1/3,1/3,1/3)): θ₁₃ is freed (exact-ρ tribimaximal mixing is excluded at 38σ), sin²θ₁₂ = 0.3408 (+2.8σ), and δ_CP is predicted: 236° or 125° (sin²θ₂₃ = 0.561), 287° or 74° (sin²θ₂₃ = 0.470). (iv) At ρ the A₄ quark determinant is the real number 27/8; with CP-symmetric couplings θ̄ ∈ {0, π}, but the CKM phase vanishes (|J| < 5×10⁻¹⁵), so the observed phase needs a quark modulus at a generic point with anomaly-free weights (Feruglio–Strumia–Titov). Status of the claims. Derived: Koide form from the Z₃ at ρ; Q ⇔ equipartition; TM2 from the (ρ, i) pair; δ_CP band; det = 27/8 and θ̄ quantisation; the end point ρ (two-field run). Principles (assumed): the species-gradient potential, equipartition with δ = Q/3, lepton moduli at the fixed points (neutrinos at i), anomaly-free quark weights with a generic quark modulus. Failed locked tests are reported (minimal A₄ at exact ρ; tribimaximal mixing; CKM phase at the fixed points). Whether slow moduli drain to i is under a pre-registered computation, to be reported in a revision. All computations pre-registered (locks D1595–D1607). The δ_CP values are predictions to be scored on DUNE and Hyper-Kamiokande. Not peer reviewed. License: CC BY-NC-SA 4.0 under the AEGIS Community License — free for individuals, scientists and students; any business or for-profit use needs the AEGIS Commercial License (https://github.com/dannyeldridge54/AEGIS/blob/master/licensing/COMMERCIAL_LICENSE.md). Built by one independent researcher with the help of AI. Code and pre-registration files: https://github.com/dannyeldridge54/AEGIS (branch unify-bg). Prepared with an AI assistant (Claude, Anthropic).

Zenodo (CERN European Organization for Nuclear Research)
Particle physics theoretical and experimental studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon