Zobrazeno 1 - 10
of 25 738
pro vyhledávání: '"HU, WEI"'
Consistency Models (CMs) have significantly accelerated the sampling process in diffusion models, yielding impressive results in synthesizing high-resolution images. To explore and extend these advancements to point-cloud-based 3D shape generation, w
Externí odkaz:
http://arxiv.org/abs/2412.19413
Deep learning models for point clouds have shown to be vulnerable to adversarial attacks, which have received increasing attention in various safety-critical applications such as autonomous driving, robotics, and surveillance. Existing 3D attackers g
Externí odkaz:
http://arxiv.org/abs/2412.12626
With the maturity of depth sensors in various 3D safety-critical applications, 3D point cloud models have been shown to be vulnerable to adversarial attacks. Almost all existing 3D attackers simply follow the white-box or black-box setting to iterati
Externí odkaz:
http://arxiv.org/abs/2412.00404
Dark solitons carrying quantized phase information arouse great interest in different nonlinear systems. A dark soliton in 1D can be stabilized in microcavity polariton condensates as a confinement is imposed on it to prevent its decay. Such a confin
Externí odkaz:
http://arxiv.org/abs/2411.14780
Thermal broadening of the quasi-particle peak in the spectral function is an important physical feature in many statistical systems, but difficult to calculate. Within the projective truncation approximation (PTA) of Green's function equation of moti
Externí odkaz:
http://arxiv.org/abs/2411.06384
Recent analysis on the training dynamics of Transformers has unveiled an interesting characteristic: the training loss plateaus for a significant number of training steps, and then suddenly (and sharply) drops to near--optimal values. To understand t
Externí odkaz:
http://arxiv.org/abs/2410.22244
Autor:
Jia, Hu-Wei, Tong, Ning-Hua
In this paper, we study the thermodynamical properties of the classical one-dimensional Klein-Gordan lattice model ($n \ge 2$) by using the cluster variation method with linear response theory. The results of this method are exact in the thermodynami
Externí odkaz:
http://arxiv.org/abs/2410.19249
This paper tackles the challenging task of 3D visual grounding-locating a specific object in a 3D point cloud scene based on text descriptions. Existing methods fall into two categories: top-down and bottom-up methods. Top-down methods rely on a pre-
Externí odkaz:
http://arxiv.org/abs/2410.15615
Extensive knowledge graphs (KGs) have been constructed to facilitate knowledge-driven tasks across various scenarios. However, existing work usually develops separate reasoning models for different KGs, lacking the ability to generalize and transfer
Externí odkaz:
http://arxiv.org/abs/2410.12288
Autor:
Yang, Yongyi, Park, Core Francisco, Lubana, Ekdeep Singh, Okawa, Maya, Hu, Wei, Tanaka, Hidenori
Prior work has shown that text-conditioned diffusion models can learn to identify and manipulate primitive concepts underlying a compositional data-generating process, enabling generalization to entirely novel, out-of-distribution compositions. Beyon
Externí odkaz:
http://arxiv.org/abs/2410.08309