Pith. sign in

SoftCLIP: Softer Cross-modal Alignment Makes CLIP Stronger

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

During the preceding biennium, vision-language pre-training has achieved noteworthy success on several downstream tasks. Nevertheless, acquiring high-quality image-text pairs, where the pairs are entirely exclusive of each other, remains a challenging task, and noise exists in the commonly used datasets. To address this issue, we propose SoftCLIP, a novel approach that relaxes the strict one-to-one constraint and achieves a soft cross-modal alignment by introducing a softened target, which is generated from the fine-grained intra-modal self-similarity. The intra-modal guidance is indicative to enable two pairs have some local similarities and model many-to-many relationships between the two modalities. Besides, since the positive still dominates in the softened target distribution, we disentangle the negatives in the distribution to further boost the relation alignment with the negatives in the cross-modal learning. Extensive experiments demonstrate the effectiveness of SoftCLIP. In particular, on ImageNet zero-shot classification task, using CC3M/CC12M as pre-training dataset, SoftCLIP brings a top-1 accuracy improvement of 6.8%/7.2% over the CLIP baseline.

citation-role summary

background 1

citation-polarity summary

fields

cs.CV 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

unclear 1

representative citing papers

citing papers explorer

Showing 1 of 1 citing paper.

  • UniCoRN: Unified Commented Retrieval Network with LMMs cs.CV · 2025-02-12 · conditional · none · ref 26 · internal anchor

    A frozen multimodal LLM is extended with a retrieval adapter and an entity adapter to retrieve a relevant image and generate a supportive textual comment.