Maxim Rakhuba

Часть публикаций из профиля, по году от новых к старым. Это не полный список последних работ автора; работы без года — в конце.

• OrthoFuse: Training-free Riemannian Fusion of Orthogonal Style-Concept Adapters for Diffusion Models2026 · аннотациясвернуть

  In a rapidly growing field of model training there is a constant practical interest in parameter-efficient fine-tuning and various techniques that use a small amount of training data to adapt the model to a narrow task.…

• Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation2026
• Optimization on the extended tensor-train manifold with shared factors2026
• Supplementary material to the manuscriptSI : Low-rank matrix and tensor approximations for compression of machine-learning interatomic potentials2025
• Low-rank matrix and tensor approximations for compression of machine-learning interatomic potentials2025
• T-Basis: a Compact Representation for Neural Networks2020 · аннотациясвернуть

  We introduce T-Basis, a novel concept for a compact representation of a set of tensors, each of an arbitrary shape, which is often seen in Neural Networks. Each of the tensors in the set is modeled using Tensor Rings,…

• Low-rank Riemannian eigensolver for high-dimensional Hamiltonians2019
• QTT-finite-element approximation for multiscale problems I: model problems in one dimension2016