Research map / Image generation / Fast diffusion sampling
DISCO: DISCrete nOise for Conditional Control in Text-to-Image Diffusion Models
Longquan Dai, Wu Ming, Dejiao Xue, He Wang, Jinhui Tang
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes DISCO, a method for conditional control of text-to-image diffusion models that addresses alignment of outputs to user-defined conditions without differentiable guidance or joint retraining.
Problem
Aligning text-to-image diffusion outputs with user-defined conditions; existing classifier-based guidance requires differentiable target models and gradient correction, while classifier-free guidance needs expensive joint training and architectural coupling.
Approach
Uses discrete noise as a means of conditional control in diffusion models, avoiding both differentiable guidance and joint training with architectural coupling (based only on the title and abstract; details not available).
Experimental setup
Builds on: classifier guidance classifier-free guidance
Closest papers by meaning
- Elucidating the design space of classifier-guided diffusion generation (ICLR 2024 · similarity 0.91)
- EMControl: Adding Conditional Control to Text-to-Image Diffusion Models via Expectation-Maximization (AAAI 2025 · similarity 0.91)
- NoiseCtrl: A Sampling-Algorithm-Agnostic Conditional Generation Method for Diffusion Models (CVPR 2025 · similarity 0.90)
- Simplifying Control Mechanism in Text-to-Image Diffusion Models (AAAI 2025 · similarity 0.90)
- CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Sampling (ICLR 2024 · similarity 0.90)
- A Simple Approach to Unifying Diffusion-based Conditional Generation (ICLR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
