REVIEW 2 cited by
Conffusion: Confidence Intervals for Diffusion Models
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Diffusion models have become the go-to method for many generative tasks, particularly for image-to-image generation tasks such as super-resolution and inpainting. Current diffusion-based methods do not provide statistical guarantees regarding the generated results, often preventing their use in high-stakes situations. To bridge this gap, we construct a confidence interval around each generated pixel such that the true value of the pixel is guaranteed to fall within the interval with a probability set by the user. Since diffusion models parametrize the data distribution, a straightforward way of constructing such intervals is by drawing multiple samples and calculating their bounds. However, this method has several drawbacks: i) slow sampling speeds ii) suboptimal bounds iii) requires training a diffusion model per task. To mitigate these shortcomings we propose Conffusion, wherein we fine-tune a pre-trained diffusion model to predict interval bounds in a single forward pass. We show that Conffusion outperforms the baseline method while being three orders of magnitude faster.
Forward citations
Cited by 2 Pith papers
-
Conformal Bounds on Full-Reference Image Quality for Imaging Inverse Problems
Conformal prediction plus approximate posterior sampling yields guaranteed bounds on full-reference image quality metrics for imaging inverse problems.
-
A Unified Comparative Study with Generalized Conformity Scores for Multi-Output Conformal Regression
New CDF-based and latent-space conformity scores give multi-output conformal predictors asymptotic conditional coverage while retaining finite-sample marginal coverage.
Discussion (0). Continue with ORCID to comment.