Diffusion Sampling

Follow this topic Ask about this topic

Diffusion sampling starts with noise and repeatedly denoises it to generate data, using reverse stochastic processes such as DDPM or deterministic flows such as DDIM and probability-flow ODEs. Work improves this process with higher-order or adaptive solvers, learned timestep schedules, and methods that reuse or predict model computations; discrete and masked diffusion models also require choices about which tokens to reveal and when. Guidance, particle methods, and search can steer samples toward rewards, constraints, diversity, or solutions to inverse problems, with applications spanning image and text generation, molecules, and scientific modeling. A central tension is reducing steps and computation without losing fidelity, diversity, or correctness, while theory seeks sharper guarantees for error, stability, and sampling cost.

Latest papers

264 papers49 papers in the last four weeks, up 104% on the four weeks before. 0.3% of all new papers.

NoiseTilt: Noise-Tilted Reverse Kernels for Diffusion Reward Alignment
Jun 16, 2026Jisung Hwang, Yunhong Min, Jaihoon Kim +2Diffusion Model Alignment Diffusion Model Guidance
1KAIST · †This work was done in part while the author was a visiting researcher at The University of Tokyo. · 2The University of Tokyo
Recursive Scaling in Masked Diffusion Models
Jun 16, 2026Alba Carballo-Castro, Julianna Piskorz, Paulius Rauba +2Masked Diffusion Models Diffusion Model Sampling
LTS4, EPFL, Lausanne, Switzerland · University of Cambridge, Cambridge, UK.
Constrained Diffusion Models with Primal-Dual Inference
Jun 15, 2026Samar Hadou, Yigit Berkay Uslu, Alejandro RibeiroPrimal-Dual Optimization Constrained Generative Modeling
Department of Electrical and Systems Engineering, University of Pennsylvania
Diffusion Flow Matching: Dimension-Improved KL Bounds and Wasserstein Guarantees
Jun 15, 2026Marta Gentiloni Silveri, Giovanni Conforti, Alain DurmusDiffusion Sampling Flow-Based Generative Modeling
Ecole Polytechnique, Massy Palaiseau, France · Università degli Studi di Padova, Padua, Italy
teasr: training-efficient any-step diffusion transformer for real-world image super-resolution
Jun 15, 2026Xiang Gao, Chenxin Zhu, Yushun Fang +2Diffusion Transformer Diffusion Model Distillation
Shanghai Jiao Tong University, China
Wasserstein Convergence of ODE-Based Samplers in Decentralized Diffusion Model via Velocity Field Decomposition
Jun 14, 2026Chencheng Tang, Xuanyu Xue, Fangyikang Wang +2Diffusion Sampling Diffusion Models
Peking University · Shanghai Jiao Tong University · MBZUAI +2
Timestep Rescheduling in Diffusion Inversion
Jun 13, 2026Shangquan Sun, Ting Gong, Zhirui Liu +5Diffusion Model Inversion Diffusion Sampling
School of CyberScience and Technology, Sun Yat-sen University, Shenzhen Campus, Shenzhen 518107, China · Tsinghua University · CUHK +2
DiRecT: Safe Diffusion-Based Planning via Receding-Horizon Denoising
Jun 13, 2026Paolo Giaretta, Zeyang Li, Navid AzizanConstrained Motion Planning Diffusion-Based Planning
MIT
Semantic DLM+: Improving Diffusion Language Models through Bias-variance Trade-off in Transition Kernel Design
Jun 13, 2026Keyue Jiang, Yuxiang Wang, Yanan Zhao +7Diffusion Sampling Diffusion Language Models
Alibaba Group · Fudan University · Nanyang Technological University +1
Neither Parallel Nor Sequential: How DiffusionGemma Actually Commits Tokens
Jun 12, 2026Ali Asaria, Tony Salomone, Deep GandhiMasked Diffusion Models Diffusion Sampling
Transformer Lab
Recursively Trained Diffusion Models: Limiting Collapse Distribution and Spectral Characterization
Jun 11, 2026Naïl B. Khelifa, Richard E. Turner, Ramji VenkataramananDiffusion Sampling Model Collapse
University of Cambridge
Enhanced Low-Density Region Exploration in Classifier-Guided Diffusion Models Through Modified Reverse Diffusion Sampling
Jun 11, 2026Jagriti Singh, Shekhar Verma, Muneendra OjhaDiffusion Model Guidance Diffusion Model Sampling
Department of IT IIIT Allahabad
The Geometry of Phase Transitions in Generative Dynamics via Projection Caustics
Jun 11, 2026Ryosuke Sakamoto, Kotaro SakamotoDiffusion Sampling Phase Transitions
Institute for the Advanced Study of Human Biology, Institute for Advanced Study, Kyoto University · Graduate School of Engineering, The University of Tokyo
Divide-and-Denoise: A Game-Theoretic Method for Fairly Composing Diffusion Models
Jun 8, 2026Abhi Gupta, Polina Barabanshchikova, Vikas Garg +2Conditional Image Generation Diffusion Sampling
MIT ORC, USA · MIT CSAIL, USA · ELLIS Institute Finland +3
Tracing the Oracle: Improving Diffusion Timestep Scheduling for 3D CT Reconstruction
Jun 4, 2026Yujia Wu, Zhaoqiang LiuDiffusion Sampling Few-Step Diffusion Sampling
The authors are with the School of Computer Science and Engineering, University of Electronic Science and Technology of China.
DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation
Jun 3, 2026Thanh-Tung Le, Yunhan Zhao, Menglei Chai +5Autoregressive Diffusion Real-Time Interactive Video Generation
University of California, Irvine · Google · Google DeepMind
Inverting the Generation Process of Denoising Diffusion Implicit Models: Empirical Evaluation and a Novel Method
Jun 2, 2026Yan Zeng, Masanori Suganuma, Takayuki OkataniDiffusion Model Inversion Diffusion Sampling
Graduate School of Information Sciences, Tohoku University · RIKEN Center for AIP
BlockGen: Flexible Blockwise Sequence Modeling with Hybrid Samplers
Jun 1, 2026Justin Deschenaux, Caglar GulcehreMasked Diffusion Models Diffusion Sampling
EPFL Lausanne, Switzerland · EPFL, Lausanne, Switzerland · Microsoft AI
Self-Regulating Annealing in Heavy-Tailed Diffusion Models
Jun 1, 2026Keito Wakatsuki, Hideaki ShimazakiStochastic Differential Equations Diffusion Sampling
Graduate School of Informatics Kyoto University, Kyoto, Japan
Uncertainty-Calibrated Diffusion for Reliable 3D Molecular Graph Generation
Jun 1, 2026Fang Wan, Jingxiang Qu, Yi LiuDiffusion Sampling Molecular Graph Generation
State University of New York at Stony Brook
GLENS: Global Search via Learning from Solver Iterates with Diffusion Models
May 29, 2026Anjian Li, Bartolomeo Stellato, Ryne BeesonLearning to Optimize Diffusion Sampling
Department of Electrical and Computer Engineering, Princeton University, Princeton, 08544, NJ, USA. · Department of Operations Research and Financial Engineering, Princeton University, Princeton, 08544, NJ, USA. · Department of Mechanical and Aerospace Engineering, Princeton University, Princeton, 08544, NJ, USA.
Parallel Tempering Initial Sampling in Inference-Time Reward Alignment
May 29, 2026Myeongjun Oh, Gwangho Kim, Sungyoon LeeDiffusion Model Alignment Markov Chain Monte Carlo
Department of Artificial Intelligence 2Department of Computer Science Hanyang University, Seoul, Korea
The Score Hamiltonian: Mapping Diffusion Models to Adiabatic Transport
May 28, 2026Peter Halmos, Boris HaninDiffusion Sampling Score-Based Generative Modeling
Computer Science Department, Princeton University. · ORFE Department, Princeton University.
Colored Noise Diffusion Sampling
May 28, 2026Hadar Davidson, Noam Issachar, Sagie BenaimDiffusion Sampling Spectral Bias
The Hebrew University of Jerusalem
DEMON: Diffusion Engine for Musical Orchestrated Noise
May 27, 2026Ryan FosdickDiffusion Models Controllable Music Generation
Daydream
From Scores to Gibbs Correctors: Accelerating Uniform-Rate Discrete Diffusion Models
May 26, 2026Yuchen Liang, Ness Shroff, Yingbin LiangDiffusion Sampling Diffusion Model Inference Acceleration
The Ohio State University
On the Error-Correcting Effects of Stochasticity in Discrete Diffusion
May 26, 2026William Yuan, Sungwon Jeong, Amirali AghazadehMarkov Models Diffusion Sampling
Georgia Institute of Technology
Beyond Generative Priors: Minority Sampling with JEPA-Guided Diffusion
May 23, 2026Sol Park, Soobin UmDiffusion Model Guidance Diffusion Sampling
Agency for Defense Development, Daejeon, South Korea · Department of Artificial Intelligence, Kookmin University, Seoul, South Korea.
Precise: SDE-Consistent Stochastic Sampling for RL Post-Training of Flow-Matching Models
May 22, 2026Jade Zou, Tao Huang, Weijie Kong +7Diffusion Model Alignment Flow Matching
Peking University · Tencent Hunyuan
Contrastive Distribution Matching for Amortized Sequential Monte Carlo in Discrete Diffusion
May 22, 2026Jaihoon Kim, Taehoon Yoon, Prin Phunyaphibarn +3Markov Chain Monte Carlo Amortized Inference
1KAIST · University of Michigan · 3NVIDIA

Jun 16, 2026J/K move · Enter open · S save

NoiseTilt: Noise-Tilted Reverse Kernels for Diffusion Reward Alignment

Jisung Hwang, Yunhong Min, Jaihoon Kim, I-Chao Shen+1

1KAIST · †This work was done in part while the author was a visiting researcher at The University of Tokyo. · 2The University of Tokyo

Diffusion Model Alignment Diffusion Model Guidance Diffusion Guidance Diffusion Sampling

Open paper AskPDF arXiv

We introduce the Noise-Tilted Reverse Kernel (NTRK), a reward-guided diffusion sampler that injects reward gradients through the noise term, leaving the pretrained reverse kernel unchanged and requiring only a single sample per step. Reward-guided sampling at inference time has greatly expanded the versatility of pretrained diffusion models. Yet existing methods face a trade-off. Gradient-based guidance shifts the reverse mean, steering generation but pushing intermediate states outside the region that the model was trained on and degrading quality. Search-based methods preserve quality but gain no gradient signal. No prior method achieves both. NTRK resolves this by keeping the reverse mean fixed and biasing the noise term toward high reward. This is enabled by a whitening operator, the central mechanism behind NTRK, which converts reward gradients into noise-compatible perturbations without losing their guiding signal. Across various reward alignment tasks, NTRK outperforms recent state-of-the-art baselines without losing sample quality. Remarkably, on aesthetic generation, NTRK surpasses the reward of the best baseline at 500 NFEs using only 25 NFEs, a 20 times reduction in compute.