A certificate-based regret analysis framework for guided-diffusion black-box optimization is introduced, with mass lift as the central quantity explaining convergence from pretrained generators.
Canonical reference
Title resolution pending
Canonical reference. 71% of citing Pith papers cite this work as background.
17
Pith papers citing it
Background
71% of classified citations
citation-role summary
background 5
method 2
citation-polarity summary
verdicts
UNVERDICTED 17representative citing papers
The paper introduces penalty-based and randomized-exploration adaptations to flow matching for improved constraint satisfaction in generative models while matching target distributions.
GORMPO uses generative models for density-based regularization in model-based offline RL, outperforming baselines by 17% on a medical dataset while providing theoretical guarantees under mild assumptions.
SOM is an actor-critic algorithm that constructs the target velocity field for one-step MeanFlow policies directly from the Q-function via score estimation and probability flow ODE, achieving claimed SOTA on locomotion tasks with reduced training and inference time.
PACE recovers geometry-consistent continuous transport dynamics from destructive single-cell snapshots via anisotropic metrics and neural bridges, reducing reconstruction distances by 23.7% on average across seven datasets.
Derives optimal inference-time guidance for stochastic interpolant policies via Kolmogorov equation analysis, enabling reactive streaming robot control with training-free and training-based mechanisms.
CrystalREPA closes the representation gap between crystal generators and universal MLIPs via contrastive alignment, yielding more stable and valid generated crystals while revealing that MLIP teacher quality is better predicted by representation distinguishability than by leaderboard accuracy.
SymDrift makes drifting models produce symmetry-invariant samples in one step via symmetrized coordinate drifts or G-invariant embeddings, outperforming prior one-shot baselines on molecular benchmarks and cutting compute by up to 40x.
TRFP combines rectified flow models with truncation to support multimodal policies in MaxEnt RL while allowing fast one-step sampling and stable training.
Diffeomorphisms and vector fields are uniquely identifiable from finitely many pushforward densities or weighted divergences, with the number of required observations determined by embedding theorems.
MUST-FM is a simulation-free multiscale supervised framework that scales unbalanced optimal transport flow matching for trajectory inference in single-cell data by exploiting hierarchical structure and transition priors.
Register tokens improve pixel-space Diffusion Transformers by cleaning high-noise feature maps, and Register Guidance amplifies that effect.
EponaV2 advances perception-free driving world models by forecasting comprehensive future 3D geometry and semantic representations, achieving SOTA planning performance on NAVSIM benchmarks.
SharpEuler estimates a sharpness profile via finite differences on calibration trajectories, smooths it, and applies a quantile transform to generate adaptive timestep grids that improve Euler sampling quality in flow matching models at fixed budgets.
The paper characterizes stability of the Kim-Milman flow map with respect to target measure variations measured in relative Fisher information.
The survey frames VLA models as pipelines that generate progressively grounded action tokens and classifies those tokens into eight types to guide future development.
Show-o2 unifies text, image, and video understanding and generation in a single autoregressive-plus-flow-matching model built on 3D causal VAE representations.
citing papers explorer
-
Regret Analysis of Guided Diffusion for Black-Box Optimization over Structured Inputs
A certificate-based regret analysis framework for guided-diffusion black-box optimization is introduced, with mass lift as the central quantity explaining convergence from pretrained generators.
-
Constraint-Aware Flow Matching via Randomized Exploration
The paper introduces penalty-based and randomized-exploration adaptations to flow matching for improved constraint satisfaction in generative models while matching target distributions.
-
Generative OOD-regularized Model-based Policy Optimization
GORMPO uses generative models for density-based regularization in model-based offline RL, outperforming baselines by 17% on a medical dataset while providing theoretical guarantees under mild assumptions.
-
Score-Based One-step MeanFlow Policy Optimization
SOM is an actor-critic algorithm that constructs the target velocity field for one-step MeanFlow policies directly from the Q-function via score estimation and probability flow ODE, achieving claimed SOTA on locomotion tasks with reduced training and inference time.
-
PACE: Geometry-Aware Bridge Transport for Single-Cell Trajectory Inference
PACE recovers geometry-consistent continuous transport dynamics from destructive single-cell snapshots via anisotropic metrics and neural bridges, reducing reconstruction distances by 23.7% on average across seven datasets.
-
Guided Streaming Stochastic Interpolant Policy
Derives optimal inference-time guidance for stochastic interpolant policies via Kolmogorov equation analysis, enabling reactive streaming robot control with training-free and training-based mechanisms.
-
CrystalREPA: Transferring Physical Priors from Universal MLIPs to Crystal Generative Models
CrystalREPA closes the representation gap between crystal generators and universal MLIPs via contrastive alignment, yielding more stable and valid generated crystals while revealing that MLIP teacher quality is better predicted by representation distinguishability than by leaderboard accuracy.
-
SymDrift: One-Shot Generative Modeling under Symmetries
SymDrift makes drifting models produce symmetry-invariant samples in one step via symmetrized coordinate drifts or G-invariant embeddings, outperforming prior one-shot baselines on molecular benchmarks and cutting compute by up to 40x.
-
Truncated Rectified Flow Policy for Reinforcement Learning with One-Step Sampling
TRFP combines rectified flow models with truncation to support multimodal policies in MaxEnt RL while allowing fast one-step sampling and stable training.
-
On the Unique Recovery of Transport Maps and Vector Fields from Finite Measure-Valued Data
Diffeomorphisms and vector fields are uniquely identifiable from finitely many pushforward densities or weighted divergences, with the number of required observations determined by embedding theorems.
-
Multiscale Supervised Unbalanced Optimal Transport Flow Matching
MUST-FM is a simulation-free multiscale supervised framework that scales unbalanced optimal transport flow matching for trajectory inference in single-cell data by exploiting hierarchical structure and transition priors.
-
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.
-
EponaV2: Driving World Model with Comprehensive Future Reasoning
EponaV2 advances perception-free driving world models by forecasting comprehensive future 3D geometry and semantic representations, achieving SOTA planning performance on NAVSIM benchmarks.
-
Sharpen Your Flow: Sharpness-Aware Sampling for Flow Matching
SharpEuler estimates a sharpness profile via finite differences on calibration trajectories, smooths it, and applies a quantile transform to generate adaptive timestep grids that improve Euler sampling quality in flow matching models at fixed budgets.
-
Stability of the Kim--Milman flow map
The paper characterizes stability of the Kim-Milman flow map with respect to target measure variations measured in relative Fisher information.
-
A Survey on Vision-Language-Action Models: An Action Tokenization Perspective
The survey frames VLA models as pipelines that generate progressively grounded action tokens and classifies those tokens into eight types to guide future development.
-
Show-o2: Improved Native Unified Multimodal Models
Show-o2 unifies text, image, and video understanding and generation in a single autoregressive-plus-flow-matching model built on 3D causal VAE representations.