Related papers: Interpreting the Synchronization Gap: The Hidden M…
Diffusion Transformers (DiT) have revolutionized high-fidelity image and video synthesis, yet their computational demands remain prohibitive for real-time applications. To solve this problem, feature caching has been proposed to accelerate…
Initially introduced as a machine translation model, the Transformer architecture has now become the foundation for modern deep learning architecture, with applications in a wide range of fields, from computer vision to natural language…
Diffusion Transformers (DiTs) have achieved state-of-the-art (SOTA) image generation quality but suffer from high latency and memory inefficiency, making them difficult to deploy on resource-constrained devices. One major efficiency…
Temporal link prediction in dynamic graphs is a fundamental problem in many real-world systems. Existing temporal graph neural networks mainly focus on learning representations of historical interactions. Despite their strong performance,…
We investigate the statistical and computational limits of latent Diffusion Transformers (DiTs) under the low-dimensional linear latent space assumption. Statistically, we study the universal approximation and sample complexity of the DiTs…
An important development in deep learning from the earliest MLPs has been a move towards architectures with structural inductive biases which enable the model to keep distinct sources of information and routes of processing well-separated.…
Digital Twin (DT) technology revolutionizes industrial processes by enabling the representation of physical entities and their dynamics to enhance productivity and operational efficiency. It has emerged as a vital enabling technology in the…
The emergence of collective behaviors in networks of dynamical units in pairwise interaction has been explained as the effect of diffusive coupling. How does the presence of higher-order interaction impact the onset of spontaneous or…
Pre-trained stable diffusion models (SD) have shown great advances in visual correspondence. In this paper, we investigate the capabilities of Diffusion Transformers (DiTs) for accurate dense correspondence. Distinct from SD, DiTs exhibit a…
Diffusion Transformers (DiTs) achieve state-of-the-art performance in high-fidelity image and video generation but suffer from expensive inference due to their iterative denoising structure. While prior methods accelerate sampling by…
We hypothesize that a key bottleneck in generalizable robot manipulation is not solely data scale or policy capacity, but a structural mismatch between current visual backbones and the physical requirements of closed-loop control. While…
Diffusion models have shown exceptional performance in visual generation tasks. Recently, these models have shifted from traditional U-Shaped CNN-Attention hybrid structures to fully transformer-based isotropic architectures. While these…
Recent advances in diffusion transformers (DiTs) have set new standards in image generation, yet remain impractical for on-device deployment due to their high computational and memory costs. In this work, we present an efficient DiT…
Diffusion models have significantly reshaped the field of generative artificial intelligence and are now increasingly explored for their capacity in discriminative representation learning. Diffusion Transformer (DiT) has recently gained…
Diffusion Transformers (DiTs) excel in generative tasks but face practical deployment challenges due to high inference costs. Feature caching, which stores and retrieves redundant computations, offers the potential for acceleration.…
Here we show that for coupled-map systems, the length of the transient prior to synchronization is both dependant on the coupling strength and dynamics of connections: systems with fixed connections and with no self-coupling display…
Diffusion Transformers (DiTs) have demonstrated remarkable performance in visual generation tasks. However, their low inference speed limits their deployment in low-resource applications. Recent training-free approaches exploit the…
Diffusion Transformers (DiTs) have recently achieved remarkable success in text-guided image generation. In image editing, DiTs project text and image inputs to a joint latent space, from which they decode and synthesize new images.…
Diffusion models trained on different, non-overlapping subsets of a dataset often produce strikingly similar outputs when given the same noise seed. We trace this consistency to a simple linear effect: the shared Gaussian statistics across…
While diffusion Multimodal Large Language Models (dMLLMs) have recently achieved remarkable strides in multimodal generation, the development of interpretability mechanisms has lagged behind their architectural evolution. Unlike traditional…