Hi Everyone, In the reading discussion this week, we will discuss: Unified Language-Vision Pretraining in LLM with Dynamic Discrete Visual Tokenization https://openreview.net/forum?id=FlvtjAB0gl This paper proposed to unify the pre-training objectives for visual and textual tokens while adapting the LLM for vision-language tasks. The reading discussion is at *4-5 pm on Jan 5 *this Friday. You may attend the reading discussion in Room 366 in the Data & Decision Building, or on Zoom https://virginiatech.zoom.us/j/5961845863. Enjoy reading! Thank you for your time! Best, ======================================== Jingyuan Qi Ph.D. student, Department of Computer Science Natural Language Processing Lab Virginia Tech (0710) Blacksburg, VA 24061