Saved in:
Bibliographic Details
Main Author: Lu, Wenbo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.03019
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909887423840256
author Lu, Wenbo
author_facet Lu, Wenbo
contents Vision-Language Pretraining (VLP) has achieved remarkable success across various downstream tasks, but such gains are largely driven by scaling up on training data. Yet, literature methods treat image-text pairs as isolated training examples; this neglects the rich relational structure naturally present in many domains, such as e-commerce product co-purchase graphs and social recommendation networks. Inspired by neuroscientific evidence that human encodes knowledge as relationship cognitive maps, we introduce Structure-aware Language-Image Pretraining (SLIP). SLIP integrates a structural contrastive loss to align modalities while also modeling relationships between neighboring entities in a structured graph. To support this paradigm, we construct a large-scale Amazon Product Co-purchase Multimodal Graph Dataset, enabling structured cross-modality supervision at scale. Experiment results show that SLIP consistently outperforms CLIP on cross-modal retrieval and classification tasks in both zero-shot and few-shot settings, showing the value of relational supervision for cross-modal alignment.
format Preprint
id arxiv_https___arxiv_org_abs_2511_03019
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SLIP: Structural-aware Language-Image Pretraining for Vision-Language Alignment
Lu, Wenbo
Computer Vision and Pattern Recognition
Artificial Intelligence
Vision-Language Pretraining (VLP) has achieved remarkable success across various downstream tasks, but such gains are largely driven by scaling up on training data. Yet, literature methods treat image-text pairs as isolated training examples; this neglects the rich relational structure naturally present in many domains, such as e-commerce product co-purchase graphs and social recommendation networks. Inspired by neuroscientific evidence that human encodes knowledge as relationship cognitive maps, we introduce Structure-aware Language-Image Pretraining (SLIP). SLIP integrates a structural contrastive loss to align modalities while also modeling relationships between neighboring entities in a structured graph. To support this paradigm, we construct a large-scale Amazon Product Co-purchase Multimodal Graph Dataset, enabling structured cross-modality supervision at scale. Experiment results show that SLIP consistently outperforms CLIP on cross-modal retrieval and classification tasks in both zero-shot and few-shot settings, showing the value of relational supervision for cross-modal alignment.
title SLIP: Structural-aware Language-Image Pretraining for Vision-Language Alignment
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2511.03019