Prachi Garg

World models · Vision language action models · Autoregressive video models · 3D scene understanding · Continual learning · parameter efficient fine-tuning · Video-LLMs · Streaming Videos · Robotics (Bimanual ALOHA · Imitation Learning) · 3D Skeleton-activity and gesture recognition · prompting for large vision-language models · few-shot learning · continual learning · continual transformers · meta-learning · semantic segmentation · multi-domain and multi-task learning · domain adaptation · incremental/lifelong/continual learning · generalization in neural networks · memorization in neural networks · domain generalization for object detection