DUNE: The major shift for Diffusion Models
Meet DUNE, the training-free framework shaking up diffusion models by refining how deep latents are handled. By targeting early-stage fluctuations, it reduces artifacts and enhances fidelity.
Read the full articleYou might also wanna read
Training-Free Single-Image Diffusion Model Achieves Fast, High-Quality Generation
We consider the problem of generating images whose internal structure -- defined by the distribution of patches across multiple scales -- ma
arxiv.org·1mo ago
Core Principles and Mathematical Foundations of Diffusion Models
This monograph presents the core principles that have guided the development of diffusion models, tracing their origins and showing how dive
arxiv.org·8mo ago
Three training-time interventions improve diffusion-based speculative decoding by 21-76%
Large language models (LLMs) achieve remarkable performance across a wide range of tasks, but their autoregressive decoding process incurs s
arxiv.org·21d ago
Terrain Diffusion: A Diffusion-Based Framework for Infinite, Real-Time Terrain Generation
For decades, procedural worlds have been built on procedural noise functions such as Perlin noise, which are fast and infinite, yet fundamen
arxiv.org·7mo agoTowards Demystifying the Creativity of Diffusion Models
Algorithms & Theory.
LILAC: A Layer-Based Framework for Multi-Concept Customization of Diffusion Models Without Joint Training
Personalizing text-to-image diffusion models to render several specific subjects in a coherent image remains challenging: the model must pre
arxiv.org·9d ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.