Zobrazeno 1 - 10
of 17 251
pro vyhledávání: '"Hua, Yang"'
Few-shot fine-tuning of Diffusion Models (DMs) is a key advancement, significantly reducing training costs and enabling personalized AI applications. However, we explore the training dynamics of DMs and observe an unanticipated phenomenon: during the
Externí odkaz:
http://arxiv.org/abs/2405.19931
Autor:
Qin, Zhen-Hui, Wu, Shu-Mao, Hao, Chen-Bei, Chen, Hua-Yang, Liang, Sheng-Nan, Yu, Si-Yuan, Chen, Yan-Feng
This work proposes a double-layer thin-film lithium niobate (LiNbO3) longitudinally excited shear wave resonator with a theoretical electromechanical coupling coefficient exceeding 60%, RaR close to 28%, and no spurious modes. This ultra-large electr
Externí odkaz:
http://arxiv.org/abs/2405.17168
Heterogeneous Federated Learning (HtFL) enables collaborative learning on multiple clients with different model architectures while preserving privacy. Despite recent research progress, knowledge sharing in HtFL is still difficult due to data and mod
Externí odkaz:
http://arxiv.org/abs/2403.15760
Diffusion Models (DMs) have evolved into advanced image generation tools, especially for few-shot generation where a pretrained model is fine-tuned on a small set of images to capture a specific style or object. Despite their success, concerns exist
Externí odkaz:
http://arxiv.org/abs/2403.11162
We provide sparse estimates for gradients of solutions to divergence form elliptic partial differential equations in terms of the source data. We give a general result of Meyers (or Gehring) type, a result for linear equations with VMO coefficients a
Externí odkaz:
http://arxiv.org/abs/2402.16213
Deep video models, for example, 3D CNNs or video transformers, have achieved promising performance on sparse video tasks, i.e., predicting one result per video. However, challenges arise when adapting existing deep video models to dense video tasks,
Externí odkaz:
http://arxiv.org/abs/2402.09257
Recently, one-stage detectors have achieved competitive accuracy and faster speed compared with traditional two-stage detectors on image data. However, in the field of video object detection (VOD), most existing VOD methods are still based on two-sta
Externí odkaz:
http://arxiv.org/abs/2402.09241
Publikováno v:
2023 International Conference on Computer Vision (ICCV) 13541-13551
Frame quality deterioration is one of the main challenges in the field of video understanding. To compensate for the information loss caused by deteriorated frames, recent approaches exploit transformer-based integration modules to obtain spatio-temp
Externí odkaz:
http://arxiv.org/abs/2402.02574
Publikováno v:
In Proceedings of the AAAI Conference on Artificial Intelligence 2021 (Vol. 35, No. 3, pp. 2620-2627)
State-of-the-art video object detection methods maintain a memory structure, either a sliding window or a memory queue, to enhance the current frame using attention mechanisms. However, we argue that these memory structures are not efficient or suffi
Externí odkaz:
http://arxiv.org/abs/2401.09923
Recently, Heterogeneous Federated Learning (HtFL) has attracted attention due to its ability to support heterogeneous models and data. To reduce the high communication cost of transmitting model parameters, a major challenge in HtFL, prototype-based
Externí odkaz:
http://arxiv.org/abs/2401.03230