Wei Xiong

9

Papers

488

Total Citations

Papers (9)

InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning

IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representation

RRM: Robust Reward Model Training Mitigates Reward Hacking

DIVE: Taming DINO for Subject-Driven Video Editing

Refine-by-Align: Reference-Guided Artifacts Refinement through Semantic Alignment

Iterative Preference Learning from Human Feedback: Bridging Theory and Practice for RLHF under KL-constraint

Earthfarsser: Versatile Spatio-Temporal Dynamical Systems Modeling in One Model

Relightful Harmonization: Lighting-aware Portrait Background Replacement

MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesis