Nadav Benedek

My research spans image generation and understanding, video analysis and synthesis, fonts, natural language processing (NLP), large language models (LLMs), multimodal learning, and reinforcement learning (RL). I work on advancing parameter-efficient fine-tuning and adaptation methods for large pre-trained models.

I also explore text-to-image and text-to-video generation, image classification and segmentation, and related vision tasks.
In the area of fonts, I study automatic glyph generation and hinting.