Viktor Moskvoretskii

I am a Fellowship PhD Student in Computer Science at EPFL DLab. I am now working on LLMs safety under the supervision of Prof. Robert West. We are largely targeting pre-training phase for safety.

My research interests are Personas inside of the LLM and their connections to safety and general LLM behavior.

I have previously worked as a Research Lead at Skoltech NLP Lab focusing on QA and uncertainty estimation.

I have pursued my M.S. in Machine Learning at HSE University under the supervision of Prof. Chris Biemann and Irina Nikishina.

Google Scholar

selected publications

Tracing Persona Vectors Through LLM Pretraining

Viktor Moskvoretskii^*, Dominik Glandorf^*, 3 more authors
* Equal contribution

arXiv preprint, 2026

arXiv Bib

@misc{moskvoretskii2026persona,
  title        = {Tracing Persona Vectors Through LLM Pretraining},
  author       = {Moskvoretskii, Viktor and Glandorf, Dominik and Medina Moreira, Jorge and Käser, Tanja and West, Robert},
  year         = {2026},
  eprint       = {2605.13329},
  archivePrefix= {arXiv},
  primaryClass = {cs.CL},
  url          = {https://arxiv.org/abs/2605.13329}
}

Synthetic Persona Pretraining: Alignment from Token Zero

Julian Minder^*, Viktor Moskvoretskii^*, Raghav Singhal^*, 8 more authors
* Equal contribution

LessWrong (blog post; paper in preparation), 2026

Blog Bib

@misc{minder2026spp,
  title        = {Synthetic Persona Pretraining: Alignment from Token Zero},
  author       = {Minder, Julian and Moskvoretskii, Viktor and Singhal, Raghav and Jiao, Difan and Bali, Kartik and Borjigin, Yiderigun and Cui, Shaobo and Krsteski, Stefan and Anderson, Ashton and Aydin, Roland and West, Robert},
  year         = {2026},
  howpublished = {LessWrong blog post},
  url          = {https://www.lesswrong.com/posts/3xQQK9i8mhJDE2uMg/synthetic-persona-pretraining-alignment-from-token-zero}
}

Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home

Viktor Moskvoretskii, Maria Marina, Mikhail Salnikov, 6 more authors

In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2025

ACL Anthology Bib

@inproceedings{moskvoretskii-etal-2025-adaptive,
  title     = {Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home},
  author    = {Moskvoretskii, Viktor and Marina, Maria and Salnikov, Mikhail and Ivanov, Nikolay and Pletenev, Sergey and Galimzianova, Daria and Krayko, Nikita and Konovalov, Vasily and Nikishina, Irina and Panchenko, Alexander},
  editor    = {Che, Wanxiang and Nabende, Joyce and Shutova, Ekaterina and Pilehvar, Mohammad Taher},
  booktitle = {Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  month     = jul,
  year      = {2025},
  address   = {Vienna, Austria},
  publisher = {Association for Computational Linguistics},
  url       = {https://aclanthology.org/2025.acl-long.319/},
  doi       = {10.18653/v1/2025.acl-long.319},
  pages     = {6355--6384},
  ISBN      = {979-8-89176-251-0}
}