2 papers
cs.LG2026
Diffract: Spectral View of LLM Domain Adaptation
Nikita Borodin, Maria Krylova, Artem Zabolotnyi +6
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. Us…
cs.CL2025
ALIEN: Aligned Entropy Head for Improving Uncertainty Estimation of LLMs
Artem Zabolotnyi, Roman Makarov, Mile Mitrovic +4
Uncertainty estimation remains a key challenge when adapting pre-trained language models to downstream classification tasks, with overconfidence often observed for difficult inputs…