Unlocking Few-Step Diffusion for Faithful Previews
cs.LG, cs.CV, stat.ML
Submitted: 2026-09-28
Updated: 2026-09-28
License: http://creativecommons.org/licenses/by/4.0/
The gist: Sampling latency compounds in diffusion workflows, where users generate and discard many candidates before keeping one.
Terminology
Abstract
Sampling latency compounds in diffusion workflows, where users generate and discard many candidates before keeping one. Surprisingly, the poor outputs of standard few-step samplers do not reflect a lack of reconstruction capacity: by optimizing only the initial noise, frozen 3-4-step samplers can closely reproduce their corresponding full-step outputs. Building on this finding, we learn corrections to the initial noise and denoising updates using endpoint supervision, improving correspondence with full-step outputs generated from the same noise and prompt. The resulting previews allow users to screen candidates cheaply and reserve full-step generation for promising ones. Input correction also transfers across sampling budgets without retraining. Experiments show substantial improvements in reference fidelity, including 53-78% lower reconstruction MSE than retrained LD3 on unconditional benchmarks, alongside improved ranking preservation and candidate selection on SD1.5, SDXL, and FLUX.1-dev.
Sources
- FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space
- SDXL-Lightning: Progressive Adversarial Diffusion Distillation
- Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis
- LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
- Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks