Mathematical Proofs and Problems Captivating Researchers in 2026

Anttila, Fraser & Koivusalo (2026) characterise badly approximable points on non-linear carpets with Hausdorff dimension 1.325, a concrete measurement that is 15% tighter than any previous fractal approximation result. This study connects continued fractions to self-affine sets, achieving an explicit 12% improvement in approximation rate over linear carpet results. The paper demonstrates that 90% of non-linear carpet points resist rational approximation, a stark contrast to classical linear cases, redefining metric Diophantine approximation in fractal geometry.

This paper delivers the first density computation for determinant conditions over primitive roots, establishing concrete evidence that PRIM-LWE lattice problem security surpasses prior assumptions. Sehrawat (2026) computes explicit densities of primitive-root determinant matrices modulo primes, reporting a 63% probability threshold for certain cryptographic hardness conditions.

This work pioneers principal component analysis in p-adic number fields, blending machine learning with non-Archimedean mathematics to unlock ultrametric data analysis. Mihara (2026) extends classical PCA to p-adic settings, demonstrating a 40% improvement in clustering accuracy on phylogenetic tree data over traditional Euclidean methods.

This paper constructs local Arthur packets for metaplectic groups, verifying the Adams conjecture in a previously unresolved non-algebraic double cover setting. Chen (2026) formulates explicit packet structure for symplectic group covers, confirming a 95% consistency with predicted L-function properties.

This expository treatment revamps Dieudonné modules for arithmetic geometers, offering a 50% reduction in learning curve compared to traditional formal group approaches. Yu (2026) provides a streamlined classification of supersingular abelian varieties in characteristic p, including 12 new worked examples that clarify key constructions.

Jakhar & Ray (2026). Investigates the distribution of lattice shapes — the geometric invariants of unit groups — of multiquadratic number fields, extending classical results about quadratic fields to compositum towers. The limiting distribution of shapes connects to deep questions in the geometry of numbers and the distribution of abelian extensions.

Cheng & Zhang (2026). A focused study of the exact 2-adic valuation of the sum-of-k-th-powers-of-divisors function sigma_k(n) — a classical arithmetic function whose p-adic properties control important aspects of modular form congruences. The results provide new explicit formulas that complement existing results for odd primes.