Sergey Pletenev

Will It Still Be True Tomorrow? Multilingual Evergreen Question Classification to Improve Trustworthy QA

arXiv 2025

How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM?

how-much-knowledge-can-you-pack-into-a-lora

Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home

arXiv 2025

LLM-Independent Adaptive RAG: Let the Question Speak for Itself

arXiv 2025

SynthDetoxM: Modern LLMs are Few-Shot Parallel Detoxification Data Annotators

arXiv 2025

<think> So let's replace this phrase with insult... </think> Lessons learned from generation of toxic texts with LLMs

arXiv 2025

Alexander Panchenko

Mikhail Salnikov

Vasily Konovalov

Daniil Moskovskiy

Daria Galimzianova

Maria Marina

Nikita Krayko

Nikolay Ivanov

Viktor Moskvoretskii

Elena Tutubalina