← Search

Maria Krylova

2 accepted papers

2026

Diffract: Spectral View of LLM Domain Adaptation

ICML 2026oral

We study continual pre-training (CPT) as a mechanism for adapting general-purpose large language models to specialized domains: mathematics, instruction, code, and natural text. Using singular value decomposition of weight matrices, we find that CPT leaves singular value spectra largely invariant, w…

Cited by 0SourceScholar
2025

GIFT-SW: Gaussian noise Injected Fine-Tuning of Salient Weights for LLMs

ACL 2025long

Parameter Efficient Fine-Tuning (PEFT) methods have gained popularity and democratized the usage of Large Language Models (LLMs). Recent studies have shown that a small subset of weights significantly impacts performance. Based on this observation, we introduce a novel PEFT method, called Gaussian n…