2 papers
cs.IR2026
Exploring LLM biases to manipulate AI search overview
Roman Smirnov
Modern large language models (LLMs) are used in many business applications in general, and specifically in web search systems and applications that generate overviews of search res…
cs.LG2024
Classifier-free guidance in LLMs Safety
Roman Smirnov
The paper describes LLM unlearning without a retaining dataset, using the ORPO reinforcement learning method with inference enhanced by modified classifier-free guidance. Significa…