ICLR 2025 accepted papers on Image generation
343 of the 3,703 papers accepted at ICLR 2025 (9.3%) are on Image generation, compared with 8.2% at ICLR 2024. They fall into 7 topics.
See ICLR 2025 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Fast diffusion sampling · 151 papers
- Physics-Informed Diffusion Models (8 citations)
- Accelerating Diffusion Transformers with Token-wise Feature Caching (4 citations)
- LeFusion: Controllable Pathology Synthesis via Lesion-Focused Diffusion Models (4 citations)
- SafeDiffuser: Safe Planning with Diffusion Probabilistic Models (4 citations)
- Adjoint Matching: Fine-tuning Flow and Diffusion Generative Models with Memoryless Stochastic Optimal Control (3 citations)
- Diffusion Models are Evolutionary Algorithms (3 citations)
- Simple Guidance Mechanisms for Discrete Diffusion Models (3 citations)
- TabDiff: a Mixed-type Diffusion Model for Tabular Data Generation (3 citations)
- Variational Diffusion Posterior Sampling with Midpoint Guidance (3 citations)
- A Periodic Bayesian Flow for Material Generation (2 citations)
- BrainACTIV: Identifying visuo-semantic properties driving cortical selectivity using diffusion-based image manipulation (2 citations)
- Dynamic Negative Guidance of Diffusion Models (2 citations)
and 139 more in the interactive map.
Text-to-image diffusion · 83 papers
- HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models (4 citations)
- ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruning (2 citations)
- Lotus: Diffusion-based Visual Foundation Model for High-quality Dense Prediction (2 citations)
- Progressive Compositionality in Text-to-Image Generative Models (2 citations)
- Build-A-Scene: Interactive 3D Layout Control for Diffusion-Based Image Generation (1 citation)
- ClassDiffusion: More Aligned Personalization Tuning with Explicit Class Guidance (1 citation)
- Fluid: Scaling Autoregressive Text-to-image Generative Models with Continuous Tokens (1 citation)
- Meissonic: Revitalizing Masked Generative Transformers for Efficient High-Resolution Text-to-Image Synthesis (1 citation)
- MS-Diffusion: Multi-subject Zero-shot Image Personalization with Layout Guidance (1 citation)
- One-Prompt-One-Story: Free-Lunch Consistent Text-to-Image Generation Using a Single Prompt (1 citation)
- Rethinking Artistic Copyright Infringements In the Era Of Text-to-Image Generative Models (1 citation)
- SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generation (1 citation)
and 71 more in the interactive map.
Autoregressive and latent generation · 57 papers
- DPLM-2: A Multimodal Diffusion Protein Language Model (7 citations)
- SymmCD: Symmetry-Preserving Crystal Generation with Diffusion Models (7 citations)
- A Sanity Check for AI-generated Image Detection (4 citations)
- Proteina: Scaling Flow-based Protein Structure Generative Models (4 citations)
- Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think (4 citations)
- MorphoDiff: Cellular Morphology Painting with Diffusion Models (3 citations)
- Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models (2 citations)
- Diffusion-Based Planning for Autonomous Driving with Flexible Guidance (2 citations)
- EC-Diffuser: Multi-Object Manipulation via Entity-Centric Behavior Generation (2 citations)
- NExT-Mol: 3D Diffusion Meets 1D Language Modeling for 3D Molecule Generation (2 citations)
- Adaptive Length Image Tokenization via Recurrent Allocation (1 citation)
- Beyond the convexity assumption: Realistic tabular data generation under quantifier-free real linear constraints (1 citation)
and 45 more in the interactive map.
Image restoration · 26 papers
- Deconstructing Denoising Diffusion Models for Self-Supervised Learning (17 citations)
- Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Model (10 citations)
- AdaIR: Adaptive All-in-One Image Restoration via Frequency Mining and Modulation (9 citations)
- Exposure Bracketing Is All You Need For A High-Quality Image (1 citation)
- Watermark Anything With Localized Messages (1 citation)
- $InterLCM$: Low-Quality Images as Intermediate States of Latent Consistency Models for Effective Blind Face Restoration
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios
- Moner: Motion Correction in Undersampled Radial MRI with Unsupervised Neural Representation
- On Disentangled Training for Nonlinear Transform in Learned Image Compression
- PostCast: Generalizable Postprocessing for Precipitation Nowcasting via Unsupervised Blurriness Modeling
- PostEdit: Posterior Sampling for Efficient Zero-Shot Image Editing
- Posterior-Mean Rectified Flow: Towards Minimum MSE Photo-Realistic Image Restoration
and 14 more in the interactive map.
Editing and style transfer · 20 papers
- GoodDrag: Towards Good Practices for Drag Editing with Diffusion Models (1 citation)
- HQ-Edit: A High-Quality Dataset for Instruction-based Image Editing (1 citation)
- Lightning-Fast Image Inversion and Editing for Text-to-Image Diffusion Models (1 citation)
- Overcoming False Illusions in Real-World Face Restoration with Multi-Modal Guided Diffusion Model (1 citation)
- Semantic Image Inversion and Editing using Rectified Stochastic Differential Equations (1 citation)
- Add-it: Training-Free Object Insertion in Images With Pretrained Diffusion Models
- DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing
- Incorporating Visual Correspondence into Diffusion Model for Virtual Try-On
- Multi-Reward as Condition for Instruction-based Image Editing
- OmniEdit: Building Image Editing Generalist Models Through Specialist Supervision
- ProtoSnap: Prototype Alignment For Cuneiform Signs
- Semantix: An Energy-guided Sampler for Semantic Style Transfer
and 8 more in the interactive map.