TTP-SSFL: Test-Time Personalization Self-Supervised Federated Learning for Accelerating MR Image Reconstruction [0.03%]
基于测试时间个性化自监督的联邦学习加速MR图像重建
Chenghu Geng,Mingfeng Jiang,Dongsheng Ruan et al.
Chenghu Geng et al.
Federated learning (FL) has emerged as a promising paradigm for accelerating magnetic resonance (MR) image reconstruction while preserving data privacy in multicenter collaborations. However, existing FL-based reconstruction methods face tw...
ARKG: Adversarially Residual Knowledge Generalization to Open-Set Domain Adaptation [0.03%]
用于开放集领域适应的残差知识对抗性泛化（ARKG）
Reyhane Ghaffari,Mohammad Sadegh Helfroush,Kamran Kazemi et al.
Reyhane Ghaffari et al.
Open-set domain adaptation (OSDA) aims to bridge the gap between labeled source and unlabeled target domains while separating unknown data in the target domain. Recent works have addressed the OSDA setting with notable results, yet they hav...
RoCA: Robust Contrastive One-Class Time Series Anomaly Detection With Contaminated Data [0.03%]
基于受污染数据的稳健对比一次时间序列异常检测方法
Xudong Mou,Rui Wang,Bo Li et al.
Xudong Mou et al.
The increasing volume of time series signals and the scarcity of labels make time series anomaly detection (TSAD) a natural fit for self-supervised deep learning. However, existing normality-based approaches face two key limitations: relyin...
Hongjie Jia,Junyi Chen,Qirong Mao et al.
Hongjie Jia et al.
The rise of e-commerce and social media has overwhelmed systems with image data, challenging real-time clustering and recommendation. Although multistage or large-pretrained-model (LPM) assisted clustering methods achieve high accuracy, the...
CausalPD: Joint Causal Discovery and Intervention for Large-Scale Pavement Distress Distribution Data [0.03%]
因果关系发现与干预联合模型揭示路面损坏分布机制
Xuesong Wu,Tianlu Pan,Xueying Chen et al.
Xuesong Wu et al.
Large-scale pavement distress distribution modeling is vital for optimizing pavement inspection systems (PISs), preventive maintenance, and overall infrastructure resilience. Generic urban computing, correlation-based, or spatiotemporal met...
Triangular Adaptive Low-Rank Adaptation for Parameter-Efficient Fine-Tuning [0.03%]
自适应低秩调整的三角形适应性方法参数高效微调
Yao Liang,Yuwei Wang,Yi Zeng
Yao Liang
Parameter efficiency and adaptability are key challenges in fine-tuning large language models (LLMs). Existing parameter-efficient fine-tuning (PEFT) methods such as low-rank adaptation (LoRA) reduce training cost but rely on fixed low-rank...
HAD: Hierarchical Asymmetric Distillation to Bridge Spatio-Temporal Gaps in Event-Based Object Tracking [0.03%]
基于事件的物体跟踪中用于弥合时空差距的层次化非对称蒸馏方法(HAD)
Yao Deng,Xian Zhong,Wenxuan Liu et al.
Yao Deng et al.
RGB cameras capture rich texture with high spatial resolution, whereas event cameras offer superior temporal resolution and high dynamic range (HDR). Exploiting their complementarity can significantly improve object tracking in challenging ...
Hybrid Transfer Active Learning for Multistream Processes With Within-Process and Cross-Process Correlation Modeling and Online Updating [0.03%]
具有过程内和跨过程相关性建模和在线更新的混合迁移主动学习用于多流过程
Zhiyong Hu,Chao Wang
Zhiyong Hu
Active learning for regression (ALR) is a prevalent tool for learning functional relationships by selectively incorporating the most informative data. However, existing ALR methods suffer from the cold-start problem and focus solely on lear...
Learning Dual Transformers for All-in-One Image Restoration From a Frequency Perspective [0.03%]
从频域角度学习双Transformer进行一站式图像复原
Jie Chu,Tong Su,Pei Liu et al.
Jie Chu et al.
This work aims to tackle the all-in-one image restoration task, which seeks to handle multiple types of degradation with a single model. The primary challenge is to extract degradation representations from the input degraded images and use ...
Hypercube Neural Topologies: Enhancing Depth Efficiency and Gradient Flow in Deep Networks [0.03%]
超立方体神经拓扑：增强深度网络的深度效率和梯度流动
Byeong-Jun Park,Dong Seog Han
Byeong-Jun Park
We propose a neural network architecture grounded in high-dimensional hypercube topology. In contrast to conventional sequential or skip-connected designs, the proposed approach maps layers to the vertices of an $n$ -dimensional hypercube a...