Zobrazeno 1 - 10
of 37 851
pro vyhledávání: '"zhang, Kai"'
Classifier-free guidance (CFG) is widely used in diffusion models but often introduces over-contrast and over-saturation artifacts at higher guidance strengths. We present EP-CFG (Energy-Preserving Classifier-Free Guidance), which addresses these iss
Externí odkaz:
http://arxiv.org/abs/2412.09966
Recent advances in diffusion and flow-based generative models have demonstrated remarkable success in image restoration tasks, achieving superior perceptual quality compared to traditional deep learning approaches. However, these methods either requi
Externí odkaz:
http://arxiv.org/abs/2412.09465
Protein inverse folding is a fundamental problem in bioinformatics, aiming to recover the amino acid sequences from a given protein backbone structure. Despite the success of existing methods, they struggle to fully capture the intricate inter-residu
Externí odkaz:
http://arxiv.org/abs/2412.09380
A spectrum of new hardware has been studied to accelerate database systems in the past decade. Specifically, CUDA cores are known to benefit from the fast development of GPUs and make notable performance improvements. The state-of-the-art GPU-based i
Externí odkaz:
http://arxiv.org/abs/2412.09337
Generative models, particularly diffusion models, have made significant success in data synthesis across various modalities, including images, videos, and 3D assets. However, current diffusion models are computationally intensive, often requiring num
Externí odkaz:
http://arxiv.org/abs/2412.05899
Autor:
Hu, Hanzhe, Yin, Tianwei, Luan, Fujun, Hu, Yiwei, Tan, Hao, Xu, Zexiang, Bi, Sai, Tulsiani, Shubham, Zhang, Kai
We present Turbo3D, an ultra-fast text-to-3D system capable of generating high-quality Gaussian splatting assets in under one second. Turbo3D employs a rapid 4-step, 4-view diffusion generator and an efficient feed-forward Gaussian reconstructor, bot
Externí odkaz:
http://arxiv.org/abs/2412.04470
Topological semimetals exhibit protected band crossings in momentum space, accompanied by corresponding surface states. Non-Hermitian Hamiltonians introduce geometry-sensitive features that dissolve this bulk-boundary correspondence principle. In thi
Externí odkaz:
http://arxiv.org/abs/2412.02782
Autor:
Pang, Ziqi, Zhang, Tianyuan, Luan, Fujun, Man, Yunze, Tan, Hao, Zhang, Kai, Freeman, William T., Wang, Yu-Xiong
We introduce RandAR, a decoder-only visual autoregressive (AR) model capable of generating images in arbitrary token orders. Unlike previous decoder-only AR models that rely on a predefined generation order, RandAR removes this inductive bias, unlock
Externí odkaz:
http://arxiv.org/abs/2412.01827
Autor:
Shao, Qian, Zhang, Kai, Du, Bang, Li, Zepeng, Wu, Yixuan, Chen, Qiyuan, Wu, Jian, Chen, Jintai, Gao, Honghao, Xu, Hongxia
Deep learning models are widely used to process Computed Tomography (CT) data in the automated screening of pulmonary diseases, significantly reducing the workload of physicians. However, the three-dimensional nature of CT volumes involves an excessi
Externí odkaz:
http://arxiv.org/abs/2412.01525
The intensive study of non-collinear magnets promotes an urgent demand for the quantitative characterization of the non-collinear magnetic structures, which host numerous exotic phenomena. Here we systematically study the non-collinear magnetic struc
Externí odkaz:
http://arxiv.org/abs/2412.01178