Hogwild! Inference: Parallel LLM Generation via Concurrent Attention
Bio
I am an ML Research Resident at Yandex Research × HSE Joint Laboratory, where I work on LLM, LLM Efficiency and Reasoning.
Previously, I was a Data Scientist at OneMarketData, an ML Engineer at VK.com and an SWE Intern at Yandex.
I hold a Bachelor’s Degree in Computer Science (Applied Mathematics and Programming) and a Master’s Degree in Computer Science (Programming and Artificial Intelligence), both from ITMO University, Saint Petersburg.
Research Interests
Large Language Models (LLMs), LLM Efficiency and Capabilities, Reasoning, Language Model Architectures, Reinforcement Learning.