PinCLIP: Large-scale Foundational Multimodal Representation at Pinterest

TL;DR

PinCLIP is a large-scale multimodal visual language model designed for Pinterest that improves retrieval, ranking, and cold-start content recommendation by leveraging a novel hybrid architecture and cross-modal alignment objectives, leading to significant business gains.

Contribution

Introduces PinCLIP, a hybrid Vision Transformer architecture with a neighbor alignment objective, enhancing multimodal content understanding and recommendation at Pinterest.

Findings

Outperforms state-of-the-art baselines by 20% in retrieval tasks

Achieves 15% increase in organic Repins for new content

Results in 8.7% higher click-through rate for new ads

Abstract

While multi-modal Visual Language Models (VLMs) have demonstrated significant success across various domains, the integration of VLMs into recommendation and retrieval systems remains a challenge, due to issues like training objective discrepancies and serving efficiency bottlenecks. This paper introduces PinCLIP, a large-scale visual representation learning approach developed to enhance retrieval and ranking models at Pinterest by leveraging VLMs to learn image-text alignment. We propose a novel hybrid Vision Transformer architecture that utilizes a VLM backbone and a hybrid fusion mechanism to capture multi-modality content representation at varying granularities. Beyond standard image-to-text alignment objectives, we introduce a neighbor alignment objective to model the cross-fusion of multi-modal representations within the Pinterest Pin-Board graph. Offline evaluations show that PinCLIP outperforms state-of-the-art baselines, such as Qwen, by 20% in multi-modal retrieval tasks. Online A/B testing demonstrates significant business impact, including substantial engagement gains across all major surfaces in Pinterest. Notably, PinCLIP significantly addresses the "cold-start" problem, enhancing fresh content distribution with a 15% Repin increase in organic content and 8.7% higher click for new Ads.