Shuzhen Li

parameter-efficient fine-tuning (2) prompt learning (2) information bottleneck (1) in-context learning (1) multimodal learning (1) semantic alignment (1) parameter-efficient learning (1) orthogonal decomposition (1) vision language model (1) cross-modal alignment (1) image-text pair (1) low-rank adaptation (1) vision-language model (1) weight matrix (1) fine-grained understanding (1) representation space (1) gradient-free optimization (1) multimodal sarcasm detection (1) multi-modal sarcasm detection (1) text-image pair (1)