Immiscible Diffusion: Accelerating Diffusion Training with Noise Assignment

Feb-17-2026, 04:21:38 GMT–Neural Information Processing Systems

In this paper, we point out that suboptimal noise-data mapping leads to slow training of diffusion models.

artificial intelligence, machine learning, natural language, (19 more...)

Neural Information Processing Systems

Feb-17-2026, 04:21:38 GMT

Conferences PDF

Country:
- North America
  - Canada (0.04)
  - United States > California
    - Alameda County > Berkeley (0.04)

Genre:
- Research Report > Experimental Study (0.93)

Technology:
- Information Technology > Artificial Intelligence
  - Machine Learning > Neural Networks
    - Deep Learning (0.48)
  - Natural Language (0.93)
  - Vision (1.00)