1 paper
Wen Dong, Zhao Wang, Shuangqing Zhang +5
Multimodal Large Language Models (MLLMs) excel in diverse vision tasks, but full-parameter retraining is computationally expensive as real-world knowledge evolves. Existing continu…