ICLR 2023 accepted papers on Image generation
75 of the 1,575 papers accepted at ICLR 2023 (4.8%) are on Image generation, compared with 4.2% at ICLR 2022. They fall into 7 topics.
See ICLR 2023 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 · 27 papers
- Diffusion Posterior Sampling for General Noisy Inverse Problems (154 citations)
- Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow (84 citations)
- Analog Bits: Generating Discrete Data using Diffusion Models with Self-Conditioning (80 citations)
- Fast Sampling of Diffusion Models with Exponential Integrator (57 citations)
- Diffusion Adversarial Representation Learning for Self-supervised Vessel Segmentation (30 citations)
- gDDIM: Generalized denoising diffusion implicit models (27 citations)
- Truncated Diffusion Probabilistic Models and Diffusion-based Adversarial Auto-Encoders (19 citations)
- Sampling is as easy as learning the score: theory for diffusion models with minimal data assumptions (14 citations)
- Understanding DDPM Latent Codes Through Optimal Transport (12 citations)
- Blurring Diffusion Models (11 citations)
- STaSy: Score-based Tabular data Synthesis (9 citations)
- Score-based Continuous-time Discrete Diffusion Models (6 citations)
and 15 more in the interactive map.
Autoregressive and latent generation · 18 papers
- InCoder: A Generative Model for Code Infilling and Synthesis (133 citations)
- DiffuSeq: Sequence to Sequence Text Generation with Diffusion Models (92 citations)
- 3D Equivariant Diffusion for Target-Aware Molecule Generation and Affinity Prediction (81 citations)
- AudioGen: Textually Guided Audio Generation (52 citations)
- The Role of ImageNet Classes in Fréchet Inception Distance (44 citations)
- Distilling Model Failures as Directions in Latent Space (11 citations)
- Generative Modelling with Inverse Heat Dissipation (11 citations)
- Discrete Contrastive Diffusion for Cross-Modal Music and Image Generation (9 citations)
- Rarity Score : A New Metric to Evaluate the Uncommonness of Synthesized Images (6 citations)
- Finding the Global Semantic Representation in GAN through Fréchet Mean (2 citations)
- FunkNN: Neural Interpolation for Functional Generation (2 citations)
- Edge Guided GANs with Contrastive Learning for Semantic Image Synthesis (1 citation)
and 6 more in the interactive map.
Editing and style transfer · 9 papers
- DiffEdit: Diffusion-based semantic image editing with mask guidance (100 citations)
- Diffusion Models Already Have A Semantic Latent Space (51 citations)
- Diffusion-based Image Translation using disentangled style and content representation (46 citations)
- Dual Diffusion Implicit Bridges for Image-to-Image Translation (13 citations)
- Interactive Portrait Harmonization (7 citations)
- Arbitrary Virtual Try-on Network: Characteristics Representation and Trade-off between Body and Clothing (3 citations)
- Learning Input-agnostic Manipulation Directions in StyleGAN with Text Guidance
- DiffusER: Diffusion via Edit-based Reconstruction
- Multi-domain image generation and translation with identifiability guarantees
Image restoration · 8 papers
- Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model (90 citations)
- Embedding Fourier for Ultra-High-Definition Low-Light Image Enhancement (44 citations)
- Scaling Laws For Deep Learning Based Image Reconstruction (22 citations)
- DDM$^2$: Self-Supervised Diffusion MRI Denoising with Generative Diffusion Models (16 citations)
- RGI: robust GAN-inversion for mask-free image inpainting and unsupervised pixel-wise anomaly detection (8 citations)
- NERDS: A General Framework to Train Camera Denoisers from Raw-RGB Noisy Image Pairs
- Pseudoinverse-Guided Diffusion Models for Inverse Problems
- Selective Frequency Network for Image Restoration
Text-to-image diffusion · 7 papers
- An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion (471 citations)
- DreamFusion: Text-to-3D using 2D Diffusion (469 citations)
- Prompt-to-Prompt Image Editing with Cross-Attention Control (362 citations)
- Training-Free Structured Diffusion Guidance for Compositional Text-to-Image Synthesis (69 citations)
- IS SYNTHETIC DATA FROM GENERATIVE MODELS READY FOR IMAGE RECOGNITION? (54 citations)
- kNN-Diffusion: Image Generation via Large-Scale Retrieval (44 citations)
- Re-Imagen: Retrieval-Augmented Text-to-Image Generator (43 citations)
GANs · 5 papers
- Diffusion-GAN: Training GANs with Diffusion (66 citations)
- BigVGAN: A Universal Neural Vocoder with Large-Scale Training (46 citations)
- PandA: Unsupervised Learning of Parts and Appearances in the Feature Maps of GANs (10 citations)
- Forward Super-Resolution: How Can GANs Learn Hierarchical Generative Models for Real-World Distributions (3 citations)
- Few-shot Cross-domain Image Generation via Inference-time Latent-code Learning