Abstract: Self-supervised learning guided by masked image modeling, such as masked autoencoder (MAE), has attracted wide attention for pretraining vision transformers in remote sensing. However, MAE ...

Abstract: This paper introduces a novel Large Language Model (LLM)-based system designed to enhance learning effect through Socratic inquiry, thereby fostering deep understanding and longterm ...