寻找
发表于 2025-3-30 11:36:04
,Static and Dynamic Concepts for Self-supervised Video Representation Learning,pose to first learn general visual concepts then attend to discriminative local areas for video understanding. Specifically, we utilize static frame and frame difference to help decouple static and dynamic concepts, and respectively align the concept distributions in latent space. We add diversity a
flammable
发表于 2025-3-30 14:10:16
SphereFed: Hyperspherical Federated Learning,challenge is the handling of non-. (independent identically distributed) data across multiple clients that may induce disparities of their local features. We introduce the Hyperspherical Federated Learning (SphereFed) framework to address the non-. issue by constraining learned representations of da
显而易见
发表于 2025-3-30 20:14:30
http://reply.papertrans.cn/24/2343/234272/234272_53.png
Texture
发表于 2025-3-30 23:11:41
,Posterior Refinement on Metric Matrix Improves Generalization Bound in Metric Learning,trained on finite known data can achieve similitude performance on infinite unseen data. While considerable efforts have been made to bound the generalization gap by enhancing the model architecture and training protocol a priori in the training phase, none of them notice that a lightweight posterio
贪婪性
发表于 2025-3-31 01:27:03
http://reply.papertrans.cn/24/2343/234272/234272_55.png
OCTO
发表于 2025-3-31 05:22:38
http://reply.papertrans.cn/24/2343/234272/234272_56.png
intolerance
发表于 2025-3-31 11:12:26
,CoSCL: Cooperation of Small Continual Learners is Stronger Than a Big One, general, learning all tasks with a shared set of parameters suffers from severe interference between tasks; while learning each task with a dedicated parameter subspace is limited by scalability. In this work, we theoretically analyze the generalization errors for learning plasticity and memory sta
新娘
发表于 2025-3-31 16:34:34
,Manifold Adversarial Learning for Cross-domain 3D Shape Representation,generalization to out-of-distribution 3D point clouds remains challenging for DNNs. As annotating large-scale point clouds is prohibitively expensive or even impossible, strategies for generalizing DNN models to unseen domains of point clouds without access to those domains during training are urgen
Confirm
发表于 2025-3-31 20:44:30
http://reply.papertrans.cn/24/2343/234272/234272_59.png
大喘气
发表于 2025-3-31 23:02:48
,LoRD: Local 4D Implicit Representation for High-Fidelity Dynamic Human Modeling, to missing surface details and accumulating tracking error. While many deep local representations have shown promising results for 3D shape modeling, their 4D counterpart does not exist yet. In this paper, we fill this blank by proposing a novel .cal 4D implicit .epresentation for .ynamic clothed h