Research
Efficient Text-to-Image Generation: An Adaptive Step Schedule Controller for Diffusion Models
arXiv:2609.16572v1 Announce Type: cross Abstract: Text-to-image diffusion models often use a fixed number of denoising steps, balancing time costs and image quality. However, the optimal number of ste
arXiv:2609.16572v1 Announce Type: cross Abstract: Text-to-image diffusion models often use a fixed number of denoising steps, balancing time costs and image quality. However, the optimal number of steps depends on the complexity of the input text prompt. We propose an adaptive diffusion controller that dynamically adjusts the number of steps to generate high-quality images efficiently, without additional model training. By leveraging a mixture of step schedules with varying step sizes and evaluating the error term discrepancy at each timestep, our method transitions between schedules to optimize performance. Experiments on COCO and DiffusionDB show that our approach reduces inference time while maintaining visual fidelity, offering a more efficient alternative for text-to-image diffusion models.
Related
- LENS: Low-Frequency Eigen Noise Shaping for Efficient Diffusion Sampling
- When Few Steps Are Enough: Training-Free Acceleration of Identity-Preserved Generation
- Beta Sampling is All You Need: Efficient Image Generation Strategy for Diffusion Models using Stepwise Spectral Analysis
- Naive PAINE: Lightweight Text-to-Image Generation Improvement with Prompt Evaluation
- IB-Flow: Information Bottleneck-Guided CFG Distillation for Few-Step Text-to-Image Generation
Source: arXiv cs.AI | 2026-09-16