Richang Hong

Streaming Drag-Oriented Interactive Video Manipulation: Drag Anything, Anytime!

ProSafePrune: Projected Safety Pruning for Mitigating Over-Refusal in LLMs

MoSE: Decoupled Tuning for Forgetting-Resilient Multi-task Fine-tuning of LLMs

Text2Lip: Progressive Lip-Synced Talking Face Generation from Text via Viseme-Guided Rendering

SignAligner: Harmonizing Complementary Pose Modalities for Coherent Sign Language Generation

Generative Distribution Distillation

Visual Token Compression Enhances Model Robustness of VLMs

Rebalancing Contrastive Alignment with Bottlenecked Semantic Increments in Text-Video Retrieval

Learning with Analogical Reasoning for Robust Few-Shot Learning

GT-Mean Loss: A Simple Yet Effective Solution for Brightness Mismatch in Low-Light Image Enhancement

UNLEARNING IS BETTER THAN UNSEEN: UNLEARNING SCORE-BASED GENERATIVE MODEL