activity
20242026
collaborators

8 papers

cs.LG2026

Super-Tuning: From Activation-Aware Pruning to Sparse Fine-Tuning

Ivan Ilin, Philip Zmushko, Peter Richtárik

Large language models (LLMs) remain expensive to fine-tune because full-parameter updates require substantial memory, compute, and per-task storage. We study whether saliency signa…

cs.LG2026

One-Step Gradient Delay is Not a Barrier for Large-Scale Asynchronous Pipeline Parallel LLM Pretraining

Philip Zmushko, Egor Petrov, Nursultan Abdullaev +2

Modern large-scale LLM pretraining benefits from utilizing Pipeline Parallelism; however, synchronous implementations leave GPUs idle during pipeline bubbles, wasting computational…

cs.LG2026

DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers

Ionut-Vlad Modoranu, Philip Zmushko, Erik Schultheis +2

Shampoo is one of the leading approximate second-order optimizers: a variant of it has won the MLCommons AlgoPerf competition, and it has been shown to produce models with lower ac…

cs.LG2026

Sign-SGD via Parameter-Free Optimization

Daniil Medyakov, Sergey Stanko, Gleb Molodtsov +4

Large language models have achieved major advances across domains, yet training them remains extremely resource-intensive. We revisit Sign-SGD, which serves both as a memory-effici…

cs.LG2025

FRUGAL: Memory-Efficient Optimization by Reducing State Overhead for Scalable Training

Philip Zmushko, Aleksandr Beznosikov, Martin Takáč +1

With the increase in the number of parameters in large language models, the process of pre-training and fine-tuning increasingly demands larger volumes of GPU memory. A significant…

math.OC2025

Sign Operator for Coping with Heavy-Tailed Noise in Non-Convex Optimization: High Probability Bounds Under -Smoothness

Nikita Kornilov, Philip Zmushko, Andrei Semenov +3

In recent years, non-convex optimization problems are more often described by generalized -smoothness assumption rather than standard one. Meanwhile, severely corrupted…