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

June 3, 2024

Differentially Private Fine-Tuning of Diffusion Models
Yu-Lin Tsai, Yizhe Li, Zekai Chen, Po-Yu Chen, Chia-Mu Yu, Xuebin Ren, Francois Buet-Golfouse
Diffusion Model Differential Privacy Differential Privacy SGD Privacy Utility Trade Private Fine Tuning
$\Delta$-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers
Pengtao Chen, Mingzhu Shen, Peng Ye, Jianjian Cao, Chongjun Tu, Christos-Savvas Bouganis, Yiren Zhao, Tao Chen
Diffusion Model Image Generation Diffusion Transformer Acceleration Method
Layout Agnostic Scene Text Image Synthesis with Diffusion Models
Qilong Zhangli, Jindong Jiang, Di Liu, Licheng Yu, Xiaoliang Dai, Ankit Ramchandani, Guan Pang, Dimitris N. Metaxas, Praveen Krishnan
Diffusion Model Image Generation Character Image Feature Encoder Scene Text Generation
Constraint-Aware Diffusion Models for Trajectory Optimization
Anjian Li, Zihan Ding, Adji Bousso Dieng, Ryne Beeson
Diffusion Model Trajectory Optimization Trajectory Optimization Problem Hybrid Loss
Faster Diffusion Sampling with Randomized Midpoints: Sequential and Parallel
Shivam Gupta, Linda Cai, Sitan Chen
Diffusion Model Sequential Information Diffusion Sampling Log Concave Sampling Parallel Sampling Discretization Method Randomized Midpoint

June 2, 2024

DistilDIRE: A Small, Fast, Cheap and Lightweight Diffusion Synthesized Deepfake Detection
Yewon Lim, Changyeon Lee, Aerin Kim, Oren Etzioni
Diffusion Model Deepfake Detector Diffusion Generated Image Diffusion Based Deepfakes
Invisible Backdoor Attacks on Diffusion Models
Sen Li, Junchi Ma, Minhao Cheng
Diffusion Model Backdoor Attack Model Watermarking Invisible Backdoor Attack
Covariance-Adaptive Sequential Black-box Optimization for Diffusion Targeted Generation
Yueming Lyu, Kim Yong Tan, Yew Soon Ong, Ivor W. Tsang
Diffusion Model Pre Trained Diffusion Model Stochastic Differential Equation Sequential Optimization Diffusion Generation
Diffusion Features to Bridge Domain Gap for Semantic Segmentation
Yuxiang Ji, Boyong He, Chenyuan Qu, Zhuoyue Tan, Chuan Qin, Liaoni Wu
Diffusion Model Semantic Segmentation Domain Gap Cross Domain Semantic Segmentation Diffusion Feature Semantic Robustness
Diffusion Tuning: Transferring Diffusion Models via Chain of Forgetting
Jincheng Zhong, Xingzhuo Guo, Jiaxiang Dong, Mingsheng Long
Diffusion Model Generative Modeling Anti Forgetting Pre Trained Diffusion Model Side Chain Fine Tuned Model Parameter Efficient Transfer Learning
Deciphering Oracle Bone Language with Diffusion Models
Haisu Guan, Huanxin Yang, Xinyu Wang, Shengwei Han, Yongge Liu, Lianwen Jin, Xiang Bai, Yuliang Liu
Diffusion Model Oracle Bone

May 31, 2024

Sifting through the Noise: A Survey of Diffusion Probabilistic Models and Their Applications to Biomolecules
Trevor Norton, Debswapna Bhattacharya
Diffusion Model Timely Survey Financial Application Diffusion Explainer Industrial Disturbing Noise Probabilistic Model Diffusion Probabilistic Model Biomolecular Structure
Spectrum-Aware Parameter Efficient Fine-Tuning for Diffusion Models
Xinxi Zhang, Song Wen, Ligong Han, Felix Juefei-Xu, Akash Srivastava, Junzhou Huang, Hao Wang, Molei Tao, Dimitris N. Metaxas
Diffusion Model Parameter Efficient Fine Tuning Low Rank Adaptation Pre Trained Generative Model Parameter Efficient Adaptation
Kaleido Diffusion: Improving Conditional Diffusion Models with Autoregressive Latent Modeling
Jiatao Gu, Ying Shen, Shuangfei Zhai, Yizhe Zhang, Navdeep Jaitly, Joshua M. Susskind
Diffusion Model Latent Variable Conditional Diffusion Model Autoregressive Language Model Fair Representation Autoregressive Generative Model Discrete Latent Representation
Amortizing intractable inference in diffusion models for vision, language, and control
Siddarth Venkatraman, Moksh Jain, Luca Scimeca, Minsu Kim, Marcin Sendera, Mohsin Hasan, Luke Rowe, Sarthak Mittal, Pablo Lemos, Emmanuel Bengio, Alexandre Adam, Jarrid Rector-Brooks, Yoshua Bengio, Glen Berseth, Nikolay Malkin
Diffusion Model Generative Modeling Human Language Vision Paper Diffusion Prior Discrete Diffusion Intractable Likelihood
Flow matching achieves minimax optimal convergence
Kenji Fukumizu, Taiji Suzuki, Noboru Isobe, Kazusato Oko, Masanori Koyama
Diffusion Model Convergence Rate Wasserstein Distance Flow Matching

Diffusion Model

Papers

Differentially Private Fine-Tuning of Diffusion Models

$\Delta$-DiT: A Training-Free Acceleration Method Tailored for Diffusion Transformers

Layout Agnostic Scene Text Image Synthesis with Diffusion Models

Constraint-Aware Diffusion Models for Trajectory Optimization

Faster Diffusion Sampling with Randomized Midpoints: Sequential and Parallel

DistilDIRE: A Small, Fast, Cheap and Lightweight Diffusion Synthesized Deepfake Detection

Invisible Backdoor Attacks on Diffusion Models

Covariance-Adaptive Sequential Black-box Optimization for Diffusion Targeted Generation

Diffusion Features to Bridge Domain Gap for Semantic Segmentation

Diffusion Tuning: Transferring Diffusion Models via Chain of Forgetting

Deciphering Oracle Bone Language with Diffusion Models

Learning to Approximate Particle Smoothing Trajectories via Diffusion Generative Models

Memorized Images in Diffusion Models share a Subspace that can be Located and Deleted

RecDiff: Diffusion Model for Social Recommendation

GenPalm: Contactless Palmprint Generation with Diffusion Models

Sifting through the Noise: A Survey of Diffusion Probabilistic Models and Their Applications to Biomolecules

Spectrum-Aware Parameter Efficient Fine-Tuning for Diffusion Models

Kaleido Diffusion: Improving Conditional Diffusion Models with Autoregressive Latent Modeling

Amortizing intractable inference in diffusion models for vision, language, and control

Flow matching achieves minimax optimal convergence