Junyang Lin

A new approach to enhance language models by focusing on diverse and complex training data.

A voice-driven model transforming audio interaction with technology.

CC-OCR sets a new standard for evaluating text recognition systems.