REVIEW 9 cited by
TRACT: Denoising Diffusion Models with Transitive Closure Time-Distillation
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
read the original abstract
Denoising Diffusion models have demonstrated their proficiency for generative sampling. However, generating good samples often requires many iterations. Consequently, techniques such as binary time-distillation (BTD) have been proposed to reduce the number of network calls for a fixed architecture. In this paper, we introduce TRAnsitive Closure Time-distillation (TRACT), a new method that extends BTD. For single step diffusion,TRACT improves FID by up to 2.4x on the same architecture, and achieves new single-step Denoising Diffusion Implicit Models (DDIM) state-of-the-art FID (7.4 for ImageNet64, 3.8 for CIFAR10). Finally we tease apart the method through extended ablations. The PyTorch implementation will be released soon.
Forward citations
Cited by 9 Pith papers
-
Amortized Moment Matching for Visual Generation
Amortized Fréchet Distance uses neural nets to match conditional means and covariances, yielding stronger one-step visual generators than explicit FD-loss or multi-step teachers.
-
IDLM: Inverse-distilled Diffusion Language Models
IDLM distills pretrained discrete diffusion language models into few-step generators, cutting inference steps by 4–64× with roughly matched GenPPL and entropy.
-
Understanding, Accelerating, and Improving MeanFlow Training
Training MeanFlow by first forming instantaneous velocity and short-gap average velocity, then shifting to long gaps, improves 1-NFE ImageNet FID from 3.43 to 2.87 and speeds training by about 2.5x.
-
Distilling Parallel Gradients for Fast ODE Solvers of Diffusion Models
A parallel-gradient ODE solver for diffusion models achieves better image quality at low step counts by learning how to combine multiple intermediate denoising evaluations per step.
-
CoVAE: Consistency Training of Variational Autoencoders
CoVAE trains a time-dependent VAE with a consistency loss so one or few decoder passes generate images, reaching FID 5.62 on MNIST and 11.69 on CIFAR-10 with adversarial loss, without a learned prior.
-
Dual-Expert Consistency Model for Efficient and High-Quality Video Generation
By training a semantic expert and a LoRA-based detail expert, DCM reaches nearly teacher-level VBench scores with 4-step video sampling on HunyuanVideo and CogVideoX.
-
Physics-Informed Distillation of Diffusion Models for PDE-Constrained Generation
Post-hoc distillation with a PDE-residual loss on final samples avoids the Jensen gap and yields one-step physics-constrained generation.
-
Efficient Difficulty-Aware Dynamic Routing for Diffusion-Based Real-World Image Super-Resolution
DDR-SR routes each real-world low-resolution image to one of two diffusion experts based on a high-frequency-loss difficulty score, using a low-compression VAE for hard images and a high-compression VAE for easy image...
-
DiffPR: Diffusion-Based Phase Reconstruction via Frequency-Decoupled Learning
DiffPR couples a quarter-resolution U-Net phase predictor with an unconditional diffusion refiner and reports solid but modest gains over U-Net baselines on four QPI datasets, with the claimed spectral-bias mechanism ...
Discussion (0). Sign in to comment.