Test-Time Low Rank Adaptation Via Confidence Maximization for Zero-Shot Generalization of Vision-Language Models
IEEE/CVF Winter Conference on Applications of Computer Vision（2025）
关键词
Zero-shot Generalization,Vision-language Models,Trainable Parameters,Entropy Loss,Attention Weights,Strict Set,Augmented Samples,Training Data,Test Samples,Domain Shift,Baseline Methods,Performance Gain,Projection Matrix,Average Prediction,Domain Generalization,Low Entropy,Update Step,Pre-trained Weights,Minimum Entropy,Text Encoder,Visual Encoding,Self-attention Module,Image Encoder,Average Confidence,Efficient Adaptation,Standard Entropy,ImageNet
AI 理解论文
溯源树
样例
生成溯源树，研究论文发展脉络
Chat Paper
正在生成论文摘要