x

LESSWRONG

LW

Viktor Moskvoretskii — LessWrong

Viktor Moskvoretskii

Viktor Moskvoretskii

Message

PhD @ EPFL DLab with Robert West

I do research on LLM Personas - where to they come from, how to apply them, how to make models safer.

130

3

3mo

Viktor Moskvoretskii

PhD @ EPFL DLab with Robert West

I do research on LLM Personas - where to they come from, how to apply them, how to make models safer.

Synthetic Persona Pretraining: Alignment from Token Zero

by Julian Minder, Raghav Singhal, Viktor Moskvoretskii, Stefan Krsteski, ashtonanderson, rolandaydin, and Robert West

Update 14.08.2026: We have since released the full paper with many important new results, in-depth analysis, and larger models. We recommend that readers read the paper or the release blog instead of this post: modelraising.ai/spp. Julian Minder, Viktor Moskvoretskii, Raghav Singhal, Difan Jiao, Kartik Bali, Yiderigun Borjigin, Shaobo Cui, Stefan...