文献类型：期刊文献

英文题名：EfficientFSL: Enhancing Few-Shot Classification via Query-Only Tuning in Vision Transformers

作者：Liao, Wenwen[1]; Ruan, Hang[1]; Yu, Jianbo[2]; Song, Bing[3]; Wang, Yuansong[4]; Yang, Xiaofeng[2]

机构：[1] College of Intelligent Robotics and Advance Manufacturing, Fudan University, Shanghai, China; [2] School of Microelectronics, Fudan University, Shanghai, China; [3] School of Information Science and Engineering, East China University of Science and Technology, Shanghai, China; [4] Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen, China

年份：2026

外文期刊名：arXiv

收录：EI(收录号：20260057930)

语种：英文

外文关键词：Classification (of information) - Machine vision - Program processors - Query processing - Structured Query Language - Tuning

摘要：Large models such as Vision Transformers (ViTs) have demonstrated remarkable superiority over smaller architectures like ResNet in few-shot classification, owing to their powerful representational capacity. However, fine-tuning such large models demands extensive GPU memory and prolonged training time, making them impractical for many real-world low-resource scenarios. To bridge this gap, we propose EfficientFSL, a query-only fine-tuning framework tailored specifically for few-shot classification with ViT, which achieves competitive performance while significantly reducing computational overhead. EfficientFSL fully leverages the knowledge embedded in the pre-trained model and its strong comprehension ability, achieving high classification accuracy with an extremely small number of tunable parameters. Specifically, we introduce a lightweight trainable Forward Block to synthesize task-specific queries that extract informative features from the intermediate representations of the pre-trained model in a query-only manner. We further propose a Combine Block to fuse multi-layer outputs, enhancing the depth and robustness of feature representations. Finally, a Support-Query Attention Block mitigates distribution shift by adjusting prototypes to align with the query set distribution. With minimal trainable parameters, EfficientFSL achieves state-of-the-art performance on four in-domain few-shot datasets and six cross-domain datasets, demonstrating its effectiveness in real-world applications. Copyright ? 2026, The Authors. All rights reserved.