MM-SOLD is a training-free particle sampler whose large-particle limit converges to a moment-matched Gibbs distribution obtained by exponentially tilting a score-smoothed target.
hub
Generative modeling by estimating gradients of the data distribution.Advances in neural information processing systems, 32
22 Pith papers cite this work. Polarity classification is still indexing.
22
Pith papers citing it
hub tools
citation-role summary
background 2
method 1
citation-polarity summary
representative citing papers
JEDI is the first online end-to-end latent diffusion world model that trains latents from denoising loss rather than reconstruction, achieving competitive Atari100k results with 43% less VRAM and over 3x faster sampling than pixel diffusion baselines.
Constraint-Aware Flow Matching integrates constraint projections into the flow matching training objective to align model dynamics with constrained sampling and reduce distributional shift.
ArenaPO infers Gaussian capability distributions from pairwise preferences and applies truncated-normal latent inference to derive fine-grained offline rewards for preference optimization of text-to-image diffusion models.
SMoDP routes action chunks in a diffusion policy to semantically specialized experts via a VLM-supervised skill predictor and dual contrastive alignment, achieving better efficiency and compositional transfer than baselines.
SiLD is a score-matching framework that learns both manifold projection and intrinsic density from a single objective, with proven sample complexity depending only on intrinsic dimension.
Flash-GRPO is a one-step GRPO framework for video diffusion alignment that applies iso-temporal grouping and temporal gradient rectification to achieve higher alignment quality and stability than full-trajectory training under low compute budgets on 1.3B-14B models.
Slowly Annealed Langevin Dynamics provides non-asymptotic KL-based convergence guarantees for tracking moving targets and enables training-free guided generation via a velocity-aware correction that accounts for pretrained marginals.
Distilled one-step consistency model from optimal-transport flow-matching teacher reconstructs high-fidelity dynamical system flows from low-fidelity data with 12x speedup, half the parameters, and 23.1% better SSIM than scratch-trained baselines.
An ensemble-based information-theoretic active learning method using ensemble Kalman inversion selects valuable tasks to optimize communication structures in LLM multi-agent systems more reliably than random sampling under limited training budgets.
Outlier tokens in DiTs are addressed with Dual-Stage Registers, which reduce artifacts and improve image generation on ImageNet and text-to-image tasks.
Uniform-based discrete diffusion models behave as associative memories that retrieve unseen data, with a dataset-size-driven memorization-to-generalization transition detectable via conditional entropy of token predictions.
Probability Flow Matching learns biophysically consistent stochastic processes for gene regulation from time-resolved single-cell measurements, where only the biophysical versions accurately capture lineage transitions, fate specification, and perturbation responses despite similar data fit.
Embedding Loss aligns feature distributions via MMD in random network embeddings to boost one-step diffusion distillation, reaching SOTA FID of 1.475 on CIFAR-10 unconditional generation.
EnFlow integrates flow-based conformer generation with energy landscape modeling to enable joint ensemble generation and ground-state identification using only 1-2 ODE steps.
Cross-matrix Krylov projection reuses shared subspaces from seed matrices to accelerate score pre-computation in diffusion models, delivering 15.8-43.7% time savings and up to 115x speedup versus DDPM baselines.
DanceGRPO applies GRPO to visual generation tasks to achieve stable policy optimization across diffusion models, rectified flows, multiple tasks, and diverse reward models, outperforming prior RL methods.
Register tokens improve pixel-space Diffusion Transformers by cleaning high-noise feature maps, and Register Guidance amplifies that effect.
Flow matching generative models preserve sample quality, diversity, and latent representations despite pruning 50% of the CelebA-HQ dataset or altering architecture and training configurations.
One-step pixel-MeanFlow models recover key galaxy morphology statistics at orders-of-magnitude lower computational cost than standard DDPM sampling while remaining weaker on fine-grained structure.
Extends score relations for tilted distributions to constant negative diagonal tilts by linking denoisers via Tweedie's formula, yielding location and time shifts in the score operator.
citing papers explorer
-
Training-Free Generative Sampling via Moment-Matched Score Smoothing
MM-SOLD is a training-free particle sampler whose large-particle limit converges to a moment-matched Gibbs distribution obtained by exponentially tilting a score-smoothed target.
-
JEDI: Joint Embedding Diffusion World Model for Online Model-Based Reinforcement Learning
JEDI is the first online end-to-end latent diffusion world model that trains latents from denoising loss rather than reconstruction, achieving competitive Atari100k results with 43% less VRAM and over 3x faster sampling than pixel diffusion baselines.
-
Constraint-Aware Flow Matching: Decision Aligned End-to-End Training for Constrained Sampling
Constraint-Aware Flow Matching integrates constraint projections into the flow matching training objective to align model dynamics with constrained sampling and reduce distributional shift.
-
Arena as Offline Reward: Efficient Fine-Grained Preference Optimization for Diffusion Models
ArenaPO infers Gaussian capability distributions from pairwise preferences and applies truncated-normal latent inference to derive fine-grained offline rewards for preference optimization of text-to-image diffusion models.
-
Semantically Structured Mixture-of-Experts for Compositional Robotic Manipulation
SMoDP routes action chunks in a diffusion policy to semantically specialized experts via a VLM-supervised skill predictor and dual contrastive alignment, achieving better efficiency and compositional transfer than baselines.
-
Provably Learning Diffusion Models under the Manifold Hypothesis: Collapse and Refine
SiLD is a score-matching framework that learns both manifold projection and intrinsic density from a single objective, with proven sample complexity depending only on intrinsic dimension.
-
Flash-GRPO: Efficient Alignment for Video Diffusion via One-Step Policy Optimization
Flash-GRPO is a one-step GRPO framework for video diffusion alignment that applies iso-temporal grouping and temporal gradient rectification to achieve higher alignment quality and stability than full-trajectory training under low compute budgets on 1.3B-14B models.
-
Slowly Annealed Langevin Dynamics: Theory and Applications to Training-Free Guided Generation
Slowly Annealed Langevin Dynamics provides non-asymptotic KL-based convergence guarantees for tracking moving targets and enables training-free guided generation via a velocity-aware correction that accounts for pretrained marginals.
-
Physical Fidelity Reconstruction via Improved Consistency-Distilled Flow Matching for Dynamical Systems
Distilled one-step consistency model from optimal-transport flow-matching teacher reconstructs high-fidelity dynamical system flows from low-fidelity data with 12x speedup, half the parameters, and 23.1% better SSIM than scratch-trained baselines.
-
Active Learning for Communication Structure Optimization in LLM-Based Multi-Agent Systems
An ensemble-based information-theoretic active learning method using ensemble Kalman inversion selects valuable tasks to optimize communication structures in LLM multi-agent systems more reliably than random sampling under limited training budgets.
-
Taming Outlier Tokens in Diffusion Transformers
Outlier tokens in DiTs are addressed with Dual-Stage Registers, which reduce artifacts and improve image generation on ImageNet and text-to-image tasks.
-
Language Diffusion Models are Associative Memories Capable of Retrieving Unseen Data
Uniform-based discrete diffusion models behave as associative memories that retrieve unseen data, with a dataset-size-driven memorization-to-generalization transition detectable via conditional entropy of token predictions.
-
Learning biophysical models of gene regulation with probability flow matching
Probability Flow Matching learns biophysically consistent stochastic processes for gene regulation from time-resolved single-cell measurements, where only the biophysical versions accurately capture lineage transitions, fate specification, and perturbation responses despite similar data fit.
-
Efficient Diffusion Distillation via Embedding Loss
Embedding Loss aligns feature distributions via MMD in random network embeddings to boost one-step diffusion distillation, reaching SOTA FID of 1.475 on CIFAR-10 unconditional generation.
-
Energy-Guided Generative Modeling for Low-Energy Molecular Structure Discovery
EnFlow integrates flow-based conformer generation with energy landscape modeling to enable joint ensemble generation and ground-state identification using only 1-2 ODE steps.
-
Efficient Score Pre-computation for Diffusion Models via Cross-Matrix Krylov Projection
Cross-matrix Krylov projection reuses shared subspaces from seed matrices to accelerate score pre-computation in diffusion models, delivering 15.8-43.7% time savings and up to 115x speedup versus DDPM baselines.
-
DanceGRPO: Unleashing GRPO on Visual Generation
DanceGRPO applies GRPO to visual generation tasks to achieve stable policy optimization across diffusion models, rectified flows, multiple tasks, and diverse reward models, outperforming prior RL methods.
-
Registers Matter for Pixel-Space Diffusion Transformers
Register tokens improve pixel-space Diffusion Transformers by cleaning high-noise feature maps, and Register Guidance amplifies that effect.
-
The Amazing Stability of Flow Matching
Flow matching generative models preserve sample quality, diversity, and latent representations despite pruning 50% of the CelebA-HQ dataset or altering architecture and training configurations.
-
Accelerating Redshift-Conditioned Galaxy Image Synthesis with One-step Generative Modeling
One-step pixel-MeanFlow models recover key galaxy morphology statistics at orders-of-magnitude lower computational cost than standard DDPM sampling while remaining weaker on fine-grained structure.
-
Technical Note on Relating Scores of Tilted Distributions
Extends score relations for tilted distributions to constant negative diagonal tilts by linking denoisers via Tweedie's formula, yielding location and time shifts in the score operator.
- VRAG: Learning World Models for Interactive Video Generation