Qinglin Lu

Bee: A High-Quality Corpus and Full-Stack Suite to Unlock Advanced Fully Open MLLMs

Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning

FaithfulFaces: Pose-Faithful Facial Identity Preservation for Text-to-Video Generation

AnyAvatar: Dynamic and Consistent Audio-Driven Human Animation for Multiple Characters

InstantCharacter: Personalize Any Characters with a Scalable Diffusion Transformer Framework

Arbitrary Generative Video Interpolation

Beyond Reward Margin: Rethinking and Resolving Likelihood Displacement in Diffusion Models via Video Generation

Autoregressive Video Generation with Learnable Memory and Consistent Decoding

OmniV2V: Versatile Video Generation and Editing via Dynamic Content Manipulation

OmniCustom: A Multimodal-Driven Architecture for Customized Video Generation

Directly Aligning the Full Diffusion Trajectory with Semantic Relative Preference Optimization

PolyVivid: Vivid Multi-Subject Video Generation with Cross-Modal Interaction and Enhancement

Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning

LoRA-Composer: Leveraging Low-Rank Adaptation for Multi-Concept Customization in Training-Free Diffusion Models