Stanford Online Products

anchor

Findings

• IC-280CLIP, OpenCLIP, and SigLIP exhibit intra-modal misalignment: intra-modal similarity comparisons are suboptimal for image-to-image and text-to-text retrieval [eval]
• IC-281SLIP's intra-modal self-supervised loss reduces intra-modal misalignment, making inter-modal inversion unnecessary for image retrieval [eval]
• IC-357Off-the-shelf foundation models (DINO, CLIP, DINOv2, ViT) exhibit higher variance in their cosine similarity distributions than dataset-specific models, reducing the discriminative power of cosine similarity retrieval [eval]