discriminative features
- Fuse2Match: Training-Free Fusion of Flow, Diffusion, and Contrastive Models for Zero-Shot Semantic Matching
- Open-Vocabulary Part Segmentation via Progressive and Boundary-Aware Strategy
- Seeing What Matters: Generalizable AI-generated Video Detection with Forensic-Oriented Augmentation
- Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs
- When Kernels Multiply, Clusters Unify: Fusing Embeddings with the Kronecker Product