Songhua Liu

papers

1,459

total citations

papers (27)

Master: Meta Style Transformer for Controllable Zero-Shot and Few-Shot Artistic Style Transfer

CVPR 2023arXiv

citations

Partial Network Cloning

CVPR 2023arXiv

citations

Mutual-Modality Adversarial Attack with Semantic Perturbation

AAAI 2024arXiv

citations

Teddy: Efficient Large-Scale Dataset Distillation via Taylor-Approximated Matching

ECCV 2024arXiv

citations

Ultra-Resolution Adaptation with Ease

ICML 2025arXiv

citations

StyDeSty: Min-Max Stylization and Destylization for Single Domain Generalization

ICML 2024arXiv

citations

Diffusion Model is Effectively Its Own Teacher

CVPR 2025

citations

Heavy Labels Out! Dataset Distillation with Label Space Lightening

ICCV 2025arXiv

citations

Distilled Datamodel with Reverse Gradient Matching

CVPR 2024arXiv

citations

Image Editing As Programs with Diffusion Models

NEURIPS 2025arXiv

citations

Slimmable Dataset Condensation

CVPR 2023

citations

UniversalBooth: Model-Agnostic Personalized Text-to-Image Generation

ICCV 2025

citations

Few-Shot Dataset Distillation via Translative Pre-Training

ICCV 2023

citations

MGDD: A Meta Generator for Fast Dataset Distillation

NEURIPS 2023

citations

Songhua Liu

papers (27)

AdaAttN: Revisit Attention Mechanism in Arbitrary Neural Style Transfer

OminiControl: Minimal and Universal Control for Diffusion Transformer

Dataset Distillation via Factorization

Deep Model Reassembly

Paint Transformer: Feed Forward Neural Painting With Stroke Prediction

MindBridge: A Cross-Subject Brain Decoding Framework

SG-Former: Self-guided Transformer with Evolving Token Reallocation

Learning with Recoverable Forgetting

DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation

Distribution Shift Inversion for Out-of-Distribution Prediction

POSTA: A Go-to Framework for Customized Artistic Poster Generation

CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up

Hash3D: Training-free Acceleration for 3D Generation

Master: Meta Style Transformer for Controllable Zero-Shot and Few-Shot Artistic Style Transfer

Partial Network Cloning

Mutual-Modality Adversarial Attack with Semantic Perturbation

Teddy: Efficient Large-Scale Dataset Distillation via Taylor-Approximated Matching

Ultra-Resolution Adaptation with Ease

StyDeSty: Min-Max Stylization and Destylization for Single Domain Generalization

Diffusion Model is Effectively Its Own Teacher

Heavy Labels Out! Dataset Distillation with Label Space Lightening

Distilled Datamodel with Reverse Gradient Matching

Image Editing As Programs with Diffusion Models

Slimmable Dataset Condensation

UniversalBooth: Model-Agnostic Personalized Text-to-Image Generation

Few-Shot Dataset Distillation via Translative Pre-Training

MGDD: A Meta Generator for Fast Dataset Distillation

papers (27)

AdaAttN: Revisit Attention Mechanism in Arbitrary Neural Style Transfer

OminiControl: Minimal and Universal Control for Diffusion Transformer

Dataset Distillation via Factorization

Deep Model Reassembly

Paint Transformer: Feed Forward Neural Painting With Stroke Prediction

MindBridge: A Cross-Subject Brain Decoding Framework

SG-Former: Self-guided Transformer with Evolving Token Reallocation

Learning with Recoverable Forgetting

DynaST: Dynamic Sparse Transformer for Exemplar-Guided Image Generation

Distribution Shift Inversion for Out-of-Distribution Prediction

POSTA: A Go-to Framework for Customized Artistic Poster Generation

CLEAR: Conv-Like Linearization Revs Pre-Trained Diffusion Transformers Up

Hash3D: Training-free Acceleration for 3D Generation

Master: Meta Style Transformer for Controllable Zero-Shot and Few-Shot Artistic Style Transfer

Partial Network Cloning

Mutual-Modality Adversarial Attack with Semantic Perturbation

Teddy: Efficient Large-Scale Dataset Distillation via Taylor-Approximated Matching

Ultra-Resolution Adaptation with Ease

StyDeSty: Min-Max Stylization and Destylization for Single Domain Generalization

Diffusion Model is Effectively Its Own Teacher

Heavy Labels Out! Dataset Distillation with Label Space Lightening

Distilled Datamodel with Reverse Gradient Matching

Image Editing As Programs with Diffusion Models

Slimmable Dataset Condensation

UniversalBooth: Model-Agnostic Personalized Text-to-Image Generation

Few-Shot Dataset Distillation via Translative Pre-Training

MGDD: A Meta Generator for Fast Dataset Distillation