BriefGPT.xyz
Sep, 2023
双对齐下的上下文感知视觉-语言模型提示调优
Context-Aware Prompt Tuning for Vision-Language Model with Dual-Alignment
HTML
PDF
Hongyu Hu, Tiancheng Lin, Jie Wang, Zhenbang Sun, Yi Xu
TL;DR
利用双重对齐提示调整(DuAl-PT),结合大规模视觉语言模型和预训练大型语言模型,在少样本识别和基于新样本泛化上取得了卓越的性能,为未来研究提供了强有力的基准。
Abstract
large-scale vision-language models
(VLMs), e.g., CLIP, learn broad visual concepts from tedious training data, showing superb generalization ability. Amount of
prompt learning
methods have been proposed to effici
→