Advancing Compositional Awareness in CLIP with Efficient Fine-Tuning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Peleg, Amit, Singh, Naman Deep, Hein, Matthias
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909872439689216
author Peleg, Amit
Singh, Naman Deep
Hein, Matthias
author_facet Peleg, Amit
Singh, Naman Deep
Hein, Matthias
contents Vision-language models like CLIP have demonstrated remarkable zero-shot capabilities in classification and retrieval. However, these models often struggle with compositional reasoning - the ability to understand the relationships between concepts. A recent benchmark, SugarCrepe++, reveals that previous works on improving compositionality have mainly improved lexical sensitivity but neglected semantic understanding. In addition, downstream retrieval performance often deteriorates, although one would expect that improving compositionality should enhance retrieval. In this work, we introduce CLIC (Compositionally-aware Learning in CLIP), a fine-tuning method based on a novel training technique combining multiple images and their associated captions. CLIC improves compositionality across architectures as well as differently pre-trained CLIP models, both in terms of lexical and semantic understanding, and achieves consistent gains in retrieval performance. This even applies to the recent CLIPS, which achieves SOTA retrieval performance. Nevertheless, the short fine-tuning with CLIC leads to an improvement in retrieval and to the best compositional CLIP model on SugarCrepe++. All our models and code are available at https://clic-compositional-clip.github.io
format Preprint
id arxiv_https___arxiv_org_abs_2505_24424
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Advancing Compositional Awareness in CLIP with Efficient Fine-Tuning
Peleg, Amit
Singh, Naman Deep
Hein, Matthias
Machine Learning
Computer Vision and Pattern Recognition
Vision-language models like CLIP have demonstrated remarkable zero-shot capabilities in classification and retrieval. However, these models often struggle with compositional reasoning - the ability to understand the relationships between concepts. A recent benchmark, SugarCrepe++, reveals that previous works on improving compositionality have mainly improved lexical sensitivity but neglected semantic understanding. In addition, downstream retrieval performance often deteriorates, although one would expect that improving compositionality should enhance retrieval. In this work, we introduce CLIC (Compositionally-aware Learning in CLIP), a fine-tuning method based on a novel training technique combining multiple images and their associated captions. CLIC improves compositionality across architectures as well as differently pre-trained CLIP models, both in terms of lexical and semantic understanding, and achieves consistent gains in retrieval performance. This even applies to the recent CLIPS, which achieves SOTA retrieval performance. Nevertheless, the short fine-tuning with CLIC leads to an improvement in retrieval and to the best compositional CLIP model on SugarCrepe++. All our models and code are available at https://clic-compositional-clip.github.io
title Advancing Compositional Awareness in CLIP with Efficient Fine-Tuning
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.24424