An exhaustive scan of Δ(96) Modular Littlest Seesaw models yields 35 viable residual-symmetry patterns with new fixed PMNS columns and sharp, testable predictions beyond TM1.
Regression-based inverse distance weighting with applications to computer experiments
3 Pith papers cite this work, alongside 67 external citations. Polarity classification is still indexing.
3
Pith papers citing it
67
external citations · Pith
abstract
We study the moduli stabilization from the viewpoint of modular flavor symmetries. We systematically analyze stabilized moduli values in possible configurations of flux compactifications, investigating probabilities of moduli values and showing which moduli values are favorable from our moduli stabilization. Then, we examine their implications on modular symmetric flavor models. It is found that distributions of complex structure modulus $\tau$ determining the flavor structure are clustered at a fixed point with the residual $\mathbb{Z}_3$ symmetry in the $SL(2,\mathbb{Z})$ fundamental region. Also, they are clustered at other specific points such as intersecting points between $|\tau|^2=k/2$ and ${\rm Re}\,\tau=0,\pm 1/4, \pm1/2$, although their probabilities are less than the $\mathbb{Z}_3$ fixed point. In general, CP-breaking vacua in the complex structure modulus are statistically disfavored in the string landscape. Among CP-breaking vacua, the values ${\rm Re}\,\tau=\pm 1/4$ are most favorable in particular when the axio-dilaton $S$ is stabilized at ${\rm Re}\,S=\pm 1/4$. That shows a strong correlation between CP phases originated from string moduli.
representative citing papers
A finite modular symmetric model generates inflation via a Coleman-Weinberg potential from vector-like quarks, with Im(τ) as inflaton and Re(τ) as heavy axion, matching cosmological observations and predicting possible isocurvature perturbations.
A joint optimization of neural manifold learning and active-learning-guided Gaussian process regression in latent space outperforms random sampling on synthetic data for complex functions.
citing papers explorer
-
Active Learning for Manifold Gaussian Process Regression
A joint optimization of neural manifold learning and active-learning-guided Gaussian process regression in latent space outperforms random sampling on synthetic data for complex functions.