Diffusion Model

Diffusion models are generative models that create data by reversing a noise-diffusion process, aiming to generate high-quality samples from complex distributions. Current research focuses on improving efficiency through techniques like stochastic Runge-Kutta methods and dynamic model architectures (e.g., Dynamic Diffusion Transformer), as well as enhancing controllability and safety via methods such as classifier-free guidance and reinforcement learning from human feedback. These advancements are significantly impacting various fields, including medical imaging, robotics, and artistic creation, by enabling novel applications in image generation, inverse problem solving, and multi-modal data synthesis.

Papers

May 31, 2024

Diffusion Models Are Innate One-Step Generators
Bowen Zheng, Tianming Yang
Diffusion Model Convolutional Layer Synchronous Generator Innate Document Segment Structure Distribution Matching Distillation
Trajectory Forecasting through Low-Rank Adaptation of Discrete Latent Codes
Riccardo Benaglia, Angelo Porrello, Pietro Buzzega, Simone Calderara, Rita Cucchiara
Diffusion Model Adaptation Concern Trajectory Forecasting Vector Quantized Variational
Unleashing the Potential of Diffusion Models for Incomplete Data Imputation
Hengrui Zhang, Liancheng Fang, Philip S. Yu
Diffusion Model Full Potential Expectation Maximization Imputation Algorithm Data Imputation Imputation Task Missing Data Imputation Various Imputation
Stochastic Optimal Control for Diffusion Bridges in Function Spaces
Byoungwoo Park, Jungwon Choi, Sungbin Lim, Juho Lee
Diffusion Model Diffusion Based Generative Model Function Space Diffusion Bridge Stochastic Optimal Control Time Series Diffusion

May 30, 2024

Slight Corruption in Pre-training Data Makes Better Diffusion Models
Hao Chen, Yujin Han, Diganta Misra, Xiang Li, Kai Hu, Difan Zou, Masashi Sugiyama, Jindong Wang, Bhiksha Raj
Diffusion Model Raw Data Pre Training High Resolution Corruption Level
Gradient Inversion of Federated Diffusion Models
Jiyue Huang, Chi Hong, Lydia Y. Chen, Stefanie Roos
Diffusion Model Gradient Inversion Attack Gradient Inversion Inversion Attack Privacy Preserving Training
Don't drop your samples! Coherence-aware training benefits Conditional diffusion
Nicolas Dufour, Victor Besnier, Vicky Kalogeiton, David Picard
Diffusion Model Generative Model Conditional Diffusion Model External Sample Discourse Coherence Conditional Diffusion Diffusion Generated Image Coherence Aware
Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks
Xiaoyu Wu, Jiaru Zhang, Yang Hua, Bohan Lyu, Hao Wang, Tao Song, Haibing Guan
Diffusion Model Fine Tuning Variational Inference Bayesian Neural Network Faulty Negative Mitigation Corruption Level Diffusion Loss Shot Fine Tuning Shot Tuning
Learning from Random Demonstrations: Offline Reinforcement Learning with Importance-Sampled Diffusion Models
Zeyu Fang, Tian Lan
LeArning Abstract Diffusion Model Offline Reinforcement Learning Policy Evaluation Offline Policy Evaluation Predictive World Model
Streaming Video Diffusion: Online Video Editing with Diffusion Models
Feng Chen, Zhen Yang, Bohan Zhuang, Qi Wu
Diffusion Model Zero Shot Stable Diffusion Video Editing Video Diffusion
Bridging Model-Based Optimization and Generative Modeling via Conservative Fine-Tuning of Diffusion Models
Masatoshi Uehara, Yulai Zhao, Ehsan Hajiramezanali, Gabriele Scalia, Gökcen Eraslan, Avantika Lal, Sergey Levine, Tommaso Biancalani
Diffusion Model Generative Modeling Reward Model Bridging Text Model Based Optimization Conservative Reward

May 29, 2024

MemControl: Mitigating Memorization in Diffusion Models via Automated Parameter Selection
Raman Dutt, Ondrej Bohdal, Pedro Sanchez, Sotirios A. Tsaftaris, Timothy Hospedales
Diffusion Model Parameter Efficient Fine Tuning Deep Model Limited Memorization Data Memorization Verbatim Memorization Memorization Capability
ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning
Ruchika Chavhan, Da Li, Timothy Hospedales
Diffusion Model Text to Image Diffusion Model Text to Image Jailbreak Attack Concept Erasure Neuron Pruning
Long-Horizon Rollout via Dynamics Diffusion for Offline Reinforcement Learning
Hanye Zhao, Xiaoshen Han, Zhengbang Zhu, Minghuan Liu, Yong Yu, Weinan Zhang
Diffusion Model Offline Reinforcement Learning Dynamic Model Policy Alignment Model Rollouts Diffusion Dynamic
Inference-Time Alignment of Diffusion Models with Direct Noise Optimization
Zhiwei Tang, Jiangweizhi Peng, Jiasheng Tang, Mingyi Hong, Fan Wang, Tsung-Hui Chang
Diffusion Model Reward Function Alignment Problem Reward Hacking Differentiable Reward Noise Optimization
Principled Probabilistic Imaging using Diffusion Models as Plug-and-Play Priors
Zihui Wu, Yu Sun, Yifan Chen, Bingliang Zhang, Yisong Yue, Katherine L. Bouman
Diffusion Model Posterior Inference Plug and Play Bayesian Inverse Problem Bayesian Imaging INverse Problem Inverse Imaging

May 28, 2024

DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention
Lianghui Zhu, Zilong Huang, Bencheng Liao, Jun Hao Liew, Hanshu Yan, Jiashi Feng, Xinggang Wang
Diffusion Model Diffusion Transformer Linear Attention Linear Attention Transformer FlashAttention 2
Phased Consistency Model
Fu-Yun Wang, Zhaoyang Huang, Alexander William Bergman, Dazhong Shen, Peng Gao, Michael Lingelbach, Keqiang Sun, Weikang Bian, Guanglu Song, Yu Liu, Hongsheng Li, Xiaogang Wang
Diffusion Model Video Generation Text to Video Generation Consistency Model Text Conditioned Image Generation
Are Images Indistinguishable to Humans Also Indistinguishable to Classifiers?
Zebin You, Xinyu Zhang, Hanzhong Guo, Jingdong Wang, Chongxuan Li
Diffusion Model Generative Model Image Generation Real Human Simple Classifier Distribution Learning Image Distribution
MAVIN: Multi-Action Video Generation with Diffusion Models via Transition Video Infilling
Bowen Zhang, Xiaofei Xie, Haotian Lu, Na Ma, Tianlin Li, Qing Guo
Diffusion Model Video Generation Video Transition

Diffusion Model

Papers

Diffusion Models Are Innate One-Step Generators

Trajectory Forecasting through Low-Rank Adaptation of Discrete Latent Codes

Unleashing the Potential of Diffusion Models for Incomplete Data Imputation

Stochastic Optimal Control for Diffusion Bridges in Function Spaces

Slight Corruption in Pre-training Data Makes Better Diffusion Models

Gradient Inversion of Federated Diffusion Models

Don't drop your samples! Coherence-aware training benefits Conditional diffusion

Exploring Diffusion Models' Corruption Stage in Few-Shot Fine-tuning and Mitigating with Bayesian Neural Networks

Learning from Random Demonstrations: Offline Reinforcement Learning with Importance-Sampled Diffusion Models

Streaming Video Diffusion: Online Video Editing with Diffusion Models

Bridging Model-Based Optimization and Generative Modeling via Conservative Fine-Tuning of Diffusion Models

MemControl: Mitigating Memorization in Diffusion Models via Automated Parameter Selection

ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning

Long-Horizon Rollout via Dynamics Diffusion for Offline Reinforcement Learning

Inference-Time Alignment of Diffusion Models with Direct Noise Optimization

Principled Probabilistic Imaging using Diffusion Models as Plug-and-Play Priors

DiG: Scalable and Efficient Diffusion Models with Gated Linear Attention

Phased Consistency Model

Are Images Indistinguishable to Humans Also Indistinguishable to Classifiers?

MAVIN: Multi-Action Video Generation with Diffusion Models via Transition Video Infilling