Reasoning Shift: How Context Silently Shortens LLM Reasoning
Bio
I am a PhD student in Computer Science at New York University, extremely lucky to be advised by Pratyusha Sharma and Pavel Izmailov. I work on reasoning, reinforcement learning, and related areas.
Previously, I was an ML Research Resident at Yandex Research × HSE, working on large language models, reasoning, and efficient inference.
I hold Bachelor's and Master's degrees in Computer Science from ITMO University, Saint Petersburg.
I also have two years of industry experience as an ML Engineer.
Updates
- Sep 2026: Started my PhD at NYU, advised by Pratyusha Sharma and Pavel Izmailov.
- Aug 2026: Reasoning Shift: How Context Silently Shortens LLM Reasoning was accepted to the COLM 2026 Workshop on Efficient Reasoning as a Spotlight.
- Sep 2025: Hogwild! Inference was accepted to NeurIPS 2025 as a Spotlight.
- Sep 2025: AutoJudge was accepted to NeurIPS 2025.
Research Interests
Reasoning, Reinforcement Learning, Large Language Models, Language Model Architectures, and related areas.