Diffusion posterior samplers produce biased outputs that can be expressed as an Ornstein-Uhlenbeck path expectation via a surrogate Gaussian path and Feynman-Kac representation, with STSL flattening the spatially varying bias term.
hub Canonical reference
International Conference on Learning Representations , year=
Canonical reference. 71% of citing Pith papers cite this work as background.
39
Pith papers citing it
Background
71% of classified citations
hub tools
citation-role summary
background 5
method 2
citation-polarity summary
representative citing papers
Quotient-space diffusion models generate correct symmetric distributions by removing redundancy on the quotient space, simplifying learning and improving results on small molecules and proteins under SE(3) symmetry.
Normalizing flows are constructed by learning the velocity of a stochastic interpolant via a quadratic loss derived from its probability current, yielding an efficient ODE-based alternative to diffusion models.
Using adjoint equations, the paper derives S-independent IPM convergence bounds for masked and uniform discrete diffusion, controlled by score-matching losses.
DiHAL uses geometry proxies to pick where to replace the lower layers of a pretrained transformer with a diffusion bridge for hidden-state reconstruction, improving over token-level diffusion baselines on 8B models.
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
Kernel-gradient drifting reformulates drifting models via kernel gradients to yield identifiable one-step generation with smoothed score matching and KL descent on Euclidean, Riemannian, and discrete spaces.
Metropolis-adjusted Langevin correctors using score-based acceptance probabilities, including an exact Bernoulli factory method and a Simpson's rule approximation, reduce sampling bias in diffusion models and improve FID scores.
The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.
STMD distills the full transition map of diffusion sampling SDEs into a conditional Mean Flow model to enable fast one- or few-step stochastic sampling without teacher models or bi-level optimization.
TRACE creates valid conformal prediction sets for complex generative models by scoring outputs via averaged denoising or velocity errors along stochastic transport paths instead of likelihoods.
A 130M-parameter continuous bitstream diffusion model with entropy-gated Langevin sampling achieves GenPPL 59.76 on LM1B and 27.06 on OWT, closing the gap to autoregressive models at matched entropy with 256 NFEs.
Absorbing discrete diffusion models the conditional distributions of clean data; reparameterizing yields a time-independent RADD that unifies with AO-ARMs and reaches SOTA perplexity among diffusion models on zero-shot language benchmarks.
Temperature scaling of density-matrix eigenvalues from LLM semantic embeddings optimizes proper-score calibration and corrects systematic overconfidence so entropy equals risk.
Entropy-adaptive per-class budgets let clients generate far fewer synthetic samples yet still close most of the accuracy gap caused by label skew in federated learning.
Audex unifies audio understanding and generation on a strong text MoE backbone with multi-stage SFT plus text-only Cascade RL, matching open SOTA audio scores while mostly retaining text capability.
RDM trains one-step generators via MMD on large batches and multi-encoder representations, achieving SOTA SW_r14 of 1.30 on ImageNet and distilling FLUX.2 to one-step with gains on GenEval and PickScore.
The generalization advantage of SGD over random sampling diminishes with growing training set size in binary networks, as measured by joint density of states over train and test accuracy.
OM-Path frames DGP inference as learning a deterministic transport map from reference to posterior inducing variables using probability-flow ODE on Doob-bridged SDE with Onsager-Machlup path regularization, yielding statistically significant gains over DBVI on power and protein UCI datasets.
DDE introduces a compact coordinator network that combines denoised outputs from pre-trained diffusion models to enable generation in larger domains and complex conditioning settings.
Higher-order Langevin dynamics reduce memorization in diffusion models by governing data trajectories with a low-pass-filtered score whose smoothness increases with order.
SSOPD converts intra-group correct-wrong contrast into process supervision by distilling a teacher distribution from the shortest correct completion into prefixes of the longest wrong completion, improving GRPO on AIME and HMMT benchmarks.
DRIFT enables stable offline-to-online fine-tuning of CTMC policies in discrete RL via advantage-weighted discrete flow matching, path-space regularization, and candidate-set approximation.
Coupled initial noises in diffusion models, with designed dependence but unchanged marginal Gaussians, improve generated image diversity on Stable Diffusion variants while preserving quality and alignment.
citing papers explorer
-
Diffusion-Based Posterior Sampling: A Feynman-Kac Analysis of Bias and Stability
Diffusion posterior samplers produce biased outputs that can be expressed as an Ornstein-Uhlenbeck path expectation via a surrogate Gaussian path and Feynman-Kac representation, with STSL flattening the spatially varying bias term.
-
Quotient-Space Diffusion Models
Quotient-space diffusion models generate correct symmetric distributions by removing redundancy on the quotient space, simplifying learning and improving results on small molecules and proteins under SE(3) symmetry.
-
Building Normalizing Flows with Stochastic Interpolants
Normalizing flows are constructed by learning the velocity of a stochastic interpolant via a quadratic loss derived from its probability current, yielding an efficient ODE-based alternative to diffusion models.
-
Dimension-Free Convergence of Discrete Diffusion Models: Adjoint Equations Induce the Right Space
Using adjoint equations, the paper derives S-independent IPM convergence bounds for masked and uniform discrete diffusion, controlled by score-matching losses.
-
Where Should Diffusion Enter a Language Model? Geometry-Guided Hidden-State Replacement
DiHAL uses geometry proxies to pick where to replace the lower layers of a pretrained transformer with a diffusion bridge for hidden-state reconstruction, improving over token-level diffusion baselines on 8B models.
-
Sobolev Regularized MMD Gradient Flow
Sobolev regularization on the witness function enables global convergence of MMD gradient flows for both sampling and generative modeling without isoperimetric assumptions.
-
Kernel-Gradient Drifting Models
Kernel-gradient drifting reformulates drifting models via kernel gradients to yield identifiable one-step generation with smoothed score matching and KL descent on Euclidean, Riemannian, and discrete spaces.
-
Metropolis-Adjusted Diffusion Models
Metropolis-adjusted Langevin correctors using score-based acceptance probabilities, including an exact Bernoulli factory method and a Simpson's rule approximation, reduce sampling bias in diffusion models and improve FID scores.
-
Sinkhorn Treatment Effects: A Causal Optimal Transport Measure
The Sinkhorn treatment effect is a new entropic optimal transport measure of divergence between counterfactual distributions that admits first- and second-order pathwise differentiability, debiased estimators, and asymptotically valid tests for distributional treatment effects.
-
Stochastic Transition-Map Distillation for Fast Probabilistic Inference
STMD distills the full transition map of diffusion sampling SDEs into a conditional Mean Flow model to enable fast one- or few-step stochastic sampling without teacher models or bi-level optimization.
-
TRACE: Transport Alignment Conformal Prediction via Diffusion and Flow Matching Models
TRACE creates valid conformal prediction sets for complex generative models by scoring outputs via averaged denoising or velocity errors along stochastic transport paths instead of likelihoods.
-
Towards Closing the Autoregressive Gap in Language Modeling via Entropy-Gated Continuous Bitstream Diffusion
A 130M-parameter continuous bitstream diffusion model with entropy-gated Langevin sampling achieves GenPPL 59.76 on LM1B and 27.06 on OWT, closing the gap to autoregressive models at matched entropy with 256 NFEs.
-
Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data
Absorbing discrete diffusion models the conditional distributions of clean data; reparameterizing yields a time-independent RADD that unifies with AO-ARMs and reaches SOTA perplexity among diffusion models on zero-shot language benchmarks.
-
Eigenvalue Calibration for Semantic Embeddings of Large Language Models
Temperature scaling of density-matrix eigenvalues from LLM semantic embeddings optimizes proper-score calibration and corrects systematic overconfidence so entropy equals risk.
-
WHERE to Generate Matters: Budget-Aware Synthetic Augmentation for Label Skewed Federated Learning
Entropy-adaptive per-class budgets let clients generate far fewer synthetic samples yet still close most of the accuracy gap caused by label skew in federated learning.
-
Unified Audio Intelligence Without Regressing on Text Intelligence
Audex unifies audio understanding and generation on a strong text MoE backbone with multi-stage SFT plus text-only Cascade RL, matching open SOTA audio scores while mostly retaining text capability.
-
Representation Distribution Matching for One-Step Visual Generation
RDM trains one-step generators via MMD on large batches and multi-encoder representations, achieving SOTA SW_r14 of 1.30 on ImageNet and distilling FLUX.2 to one-step with gains on GenEval and PickScore.
-
Revisiting the Volume Hypothesis
The generalization advantage of SGD over random sampling diminishes with growing training set size in binary networks, as measured by joint density of states over train and test accuracy.
-
Onsager-Machlup Posterior Transport for Deep Gaussian Processes
OM-Path frames DGP inference as learning a deterministic transport map from reference to posterior inducing variables using probability-flow ODE on Doob-bridged SDE with Onsager-Machlup path regularization, yielding statistically significant gains over DBVI on power and protein UCI datasets.
-
Diffusion Domain Expansion: Learning to Coordinate Pre-trained Diffusion Models
DDE introduces a compact coordinator network that combines denoised outputs from pre-trained diffusion models to enable generation in larger domains and complex conditioning settings.
-
Reducing Diffusion Model Memorization with Higher Order Langevin Dynamics
Higher-order Langevin dynamics reduce memorization in diffusion models by governing data trajectories with a low-pass-filtered score whose smoothness increases with order.
-
Self-Supervised On-Policy Distillation for Reasoning Language Models
SSOPD converts intra-group correct-wrong contrast into process supervision by distilling a teacher distribution from the shortest correct completion into prefixes of the longest wrong completion, improving GRPO on AIME and HMMT benchmarks.
-
Discrete Flow Matching for Offline-to-Online Reinforcement Learning
DRIFT enables stable offline-to-online fine-tuning of CTMC policies in discrete RL via advantage-weighted discrete flow matching, path-space regularization, and candidate-set approximation.
-
Couple to Control: Joint Initial Noise Design in Diffusion Models
Coupled initial noises in diffusion models, with designed dependence but unchanged marginal Gaussians, improve generated image diversity on Stable Diffusion variants while preserving quality and alignment.
-
Scaling Pretrained Representations Enables Label-Free Out-of-Distribution Detection Without Fine-Tuning
Scaling pretrained representations improves label-free OOD detection on frozen backbones, causing performance gaps between global and local detectors to vanish across vision and language tasks.
-
Local Intrinsic Dimension Unveils Hallucinations in Diffusion Models
Hallucinations in diffusion models are driven by local intrinsic dimension instabilities on the manifold, which Intrinsic Quenching corrects by deflating it.
-
Stylistic Attribute Control in Latent Diffusion Models
A technique for parametric stylistic control in latent diffusion models learns disentangled directions from synthetic datasets and applies them via guidance composition while preserving semantics.
-
Beyond Continuity: Simulation-free Reconstruction of Discrete Branching Dynamics from Single-cell Snapshots
USB solves the Branching Schrödinger Bridge problem to enable simulation-free inference of stochastic discrete branching dynamics from single-cell omics snapshots.
-
Cold-Start Forecasting of New Product Life-Cycles via Conditional Diffusion Models
CDLF applies conditional diffusion models to produce probabilistic life-cycle forecasts for new products by conditioning on static descriptors and reference trajectories from similar items.
-
LASER: Learning Active Sensing for Continuum Field Reconstruction
LASER couples a latent world model with a GRPO-trained RL policy to adaptively reposition sensors for sparse continuum field reconstruction, outperforming fixed and offline-optimized placement strategies across PDE and real-world benchmarks.
-
Scaling Diffusion Language Models via Adaptation from Autoregressive Models
Adapting autoregressive models via continual pre-training yields diffusion language models from 127M to 7B parameters that outperform prior diffusion models and compete with their autoregressive counterparts on language, reasoning, and commonsense benchmarks.
-
Rectified Flow: A Marginal Preserving Approach to Optimal Transport
A single-objective rectified flow variant uses neural ODEs trained by regression to monotonically decrease a fixed convex transport cost while preserving marginal distributions.
-
GPC: Large-Scale Generative Pretraining for Transferable Motor Control
GPC learns a motion vocabulary via Finite Scalar Quantization and end-to-end RL, then trains an autoregressive transformer for next-token control generation, achieving 99.98% motion reproduction success with emergent robustness.
-
RNA-FM: Flow-Matching Generative Model for Genome-wide RNA-Seq Prediction
RNA-FM is a flow-matching generative model that predicts genome-wide bulk RNA-seq expression from WSIs by learning a conditional velocity field, outperforming deterministic baselines.
-
APEX: Assumption-free Projection-based Embedding eXamination Metric for Image Quality Assessment
APEX is an assumption-free image quality metric using Sliced Wasserstein Distance on CLIP and DINOv2 embeddings that claims superior robustness to degradations and cross-dataset stability.
-
Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation
LHSD estimates local intrinsic dimension in high-D spaces by spectral filtering of the log-density Hessian via SLQ to isolate zero-curvature tangent directions.
-
Consistent Diffusion Language Models
CDLM introduces MPDC training for discrete diffusion models, recovering prior methods as limits and claiming new SOTA text generation performance especially at low sampling budgets.
-
On the Robustness of Distribution Support under Diffusion Guidance
Establishes robustness of distribution support for guided diffusion processes under exact score access across DDIM, DDPM, and exponential integrator discretizations.
- Local-Time Riemannian Score Matching on the Quantum Pure-State Manifold