Lei Zhu

LucidFlux: Caption-Free Universal Image Restoration via a Large-Scale Diffusion Transformer

PosterCraft: Rethinking High-Quality Aesthetic Poster Generation in a Unified Framework

Learning Label-Efficient Interpretable Medical Image Diagnosis via Semi-supervised Hypergraph Concept Bottleneck Model

PhotoArtAgent: Intelligent Photo Retouching with Language Model-based Artist Agent

Federated Residual Low-Rank Adaptation of Large Language Models

VQ-Seg: Vector-Quantized Token Perturbation for Semi-Supervised Medical Image Segmentation

On the Importance of Language-driven Representation Learning for Heterogeneous Federated Learning

Meissonic: Revitalizing Masked Generative Transformers for Efficient High-Resolution Text-to-Image Synthesis

Online Agglomerative Pooling for Scalable Self-Supervised Universal Segmentation

Farther Than Mirror: Explore Pattern-Compensated Depth of Mirror with Temporal Changes for Video Mirror Detection