Author

Anna Choromanska

3 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#machine learning Mar 2026

Understanding Quantization of Optimizer States in LLM Pre-training: Dynamics of State Staleness and Effectiveness of State Resets

A simple theory-guided method for choosing useful reset periods is derived, showing that in low precision the key question is not only whether resets help, but when they should be applied.

Kristi Topollai, Anna Choromanska · arXiv.org · 6 citations · ⚡2

#machine learning Preprint Sep 2026

Cost-free Spectral Estimation for Adaptive Newton--Schulz in Matrix Optimizers

The results suggest that matrix-function operations inside optimizers need not be designed for a conservative worst-case spectrum: they can cheaply measure the spectrum they are already processing and specialize computations accordingly.

Kristi Topollai, Anna Choromanska · 0 citations

#machine learning Preprint Sep 2026

Balancing Early Performance Sacrifices with Long-Term Gains: Scaling Learning-Rate Warmup Duration Across Training Horizons

A quadratic model whose modes respond differently to the peak learning rate yields a compact horizon scaling law that captures regimes ranging from essentially no warmup, through fixed-duration warmup, to durations that grow with the training horizon and explains how the preferred regime changes with peak learning rate...

Kristi Topollai, Anna Choromanska · 0 citations