Oxford Pets

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-419CLIP ViT-B/16's layer-11 residual stream contains class-discriminative information in sparse SAE latent directions, and ablating class-specific top-k latents significantly degrades zero-shot classification accuracy [eval]
• IC-643Conditioning CLIP on correct contextual attributes in the text prompt improves zero-shot classification accuracy across 13 image transformations [eval]