Theoretical analysis reveals MaskGIT's implicit temperature sampling in masked diffusion; proposes equivalent moment sampler and efficiency techniques for adaptive unmasking with image and text experiments.
A pytorch reproduction of masked generative image transformer
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
verdicts
UNVERDICTED 5roles
method 1polarities
use method 1representative citing papers
A parallel-in-time τ-leaping sampler for absorbing discrete diffusion models is introduced, with an exponential-factorial convergence proof and empirical speedups of 7-9× on synthetic tasks and 1.45-1.86× on image/text tasks while using 50% fewer NFE.
FR3D predicts a persistent 3D latent representation for future dynamic reconstruction by explicitly disentangling scene evolution from agent trajectory and using foundation-model distillation.
Re2Pix decomposes video prediction into semantic feature forecasting followed by representation-conditioned diffusion synthesis, with nested dropout and mixed supervision to handle prediction errors.
Fixed-Point Distillation constructs one-step correction targets for discrete diffusion generators via partial corruption and single teacher refinement, lifted into continuous features with a multi-bandwidth drift loss and straight-through estimation.
citing papers explorer
-
Demystifying MaskGIT Sampler and Beyond: Adaptive Order Selection in Masked Diffusion
Theoretical analysis reveals MaskGIT's implicit temperature sampling in masked diffusion; proposes equivalent moment sampler and efficiency techniques for adaptive unmasking with image and text experiments.
-
Accelerating Discrete Diffusion Models with Parallel-In-Time Sampling
A parallel-in-time τ-leaping sampler for absorbing discrete diffusion models is introduced, with an exponential-factorial convergence proof and empirical speedups of 7-9× on synthetic tasks and 1.45-1.86× on image/text tasks while using 50% fewer NFE.
-
Future Dynamic 3D Reconstruction: A 3D World Model with Disentangled Ego-Motion
FR3D predicts a persistent 3D latent representation for future dynamic reconstruction by explicitly disentangling scene evolution from agent trajectory and using foundation-model distillation.
-
Representations Before Pixels: Semantics-Guided Hierarchical Video Prediction
Re2Pix decomposes video prediction into semantic feature forecasting followed by representation-conditioned diffusion synthesis, with nested dropout and mixed supervision to handle prediction errors.
-
One-Step Distillation of Discrete Diffusion Image Generators via Fixed-Point Iteration
Fixed-Point Distillation constructs one-step correction targets for discrete diffusion generators via partial corruption and single teacher refinement, lifted into continuous features with a multi-bandwidth drift loss and straight-through estimation.