Compress Guidance in Conditional Diffusion Sampling
This addresses a major challenge in applying guidance effectively to generative tasks, but it is incremental as it builds on existing diffusion methods.
The paper tackles the problem of guidance in conditional diffusion sampling being counterproductive due to model-fitting issues, and by reducing or excluding guidance at many timesteps, it achieves significant improvements in image quality and diversity while cutting required guidance timesteps by nearly 40%.
We found that enforcing guidance throughout the sampling process is often counterproductive due to the model-fitting issue, where samples are 'tuned' to match the classifier's parameters rather than generalizing the expected condition. This work identifies and quantifies the problem, demonstrating that reducing or excluding guidance at numerous timesteps can mitigate this issue. By distributing a small amount of guidance over a large number of sampling timesteps, we observe a significant improvement in image quality and diversity while also reducing the required guidance timesteps by nearly 40%. This approach addresses a major challenge in applying guidance effectively to generative tasks. Consequently, our proposed method, termed Compress Guidance, allows for the exclusion of a substantial number of guidance timesteps while still surpassing baseline models in image quality. We validate our approach through benchmarks on label-conditional and text-to-image generative tasks across various datasets and models.