박은혁 연구실

LLM Optimization research focuses on making enormous language models more efficient, adaptable, and practical. This includes techniques to compress models, merge knowledge from multiple specialized models, and extend their context lengths without sacrificing performance.

Visual Generative Model Acceleration research is about speeding up image and video generation without compromising output quality. Modern generative models like diffusion models and autoregressive image transformers can create stunning visuals but often require lengthy sequential computations.

Efficient Embodied Intelligence research develops AI policies that let robots and embodied systems act naturally and reliably in complex real-world environments, while making the most of limited computational resources.

Performant System Design for AI Applications research focuses on the hardware-software co-design needed to run modern AI models at peak efficiency. This research area addresses how to redesign systems and algorithms together.

Robust and Reliable AI research aims to ensure that AI systems not only perform well under ideal conditions but also maintain their integrity and trustworthiness in the wild. As AI is deployed in high-stakes domains, it must be resilient to noisy inputs and adversarial perturbations.