3 papers
cs.LG2026
From Zero to Hero: An Open LLM Ecosystem for Armenian
Erik Arakelyan, Khatun Avetisyan, Meri Davtyan +5
Pretraining data for Armenian, a morphologically rich and low-resource language, is scarce, and no open Armenian LLM has been released with the data and recipe needed to reproduce…
cs.LG2025
Piecewise Constant Spectral Graph Neural Network
Vahan Martirosyan, Jhony H. Giraldo, Fragkiskos D. Malliaros
Graph Neural Networks (GNNs) have achieved significant success across various domains by leveraging graph structures in data. Existing spectral GNNs, which use low-degree polynomia…
math.CA2022
On the convergence of multiple ergodic means
Grigori A. Karagulyan, Michael T. Lacey, Vahan A. Martirosyan
Given sequence of measure preserving transformations on a measurable space . We prove a.e. convergence of the ergodic means \begin{equation} \fra…