Tree-based Text-Vision BERT for Video Search in Baidu Video Advertising

TL;DR

This paper introduces a tree-based combo-attention network (TCAN) for large-scale query-to-video search in Baidu's advertising platform, improving click-through and conversion rates by efficiently handling cross-modal attention.

Contribution

The paper proposes a practical, scalable tree-based attention model for large-scale video search, addressing the computational challenges of traditional cross-modal attention methods.

Findings

Click-through rate increased by 2.29%.

Conversion rate increased by 2.63%.

Effective deployment in Baidu's video advertising platform.

Abstract

The advancement of the communication technology and the popularity of the smart phones foster the booming of video ads. Baidu, as one of the leading search engine companies in the world, receives billions of search queries per day. How to pair the video ads with the user search is the core task of Baidu video advertising. Due to the modality gap, the query-to-video retrieval is much more challenging than traditional query-to-document retrieval and image-to-image search. Traditionally, the query-to-video retrieval is tackled by the query-to-title retrieval, which is not reliable when the quality of tiles are not high. With the rapid progress achieved in computer vision and natural language processing in recent years, content-based search methods becomes promising for the query-to-video retrieval. Benefited from pretraining on large-scale datasets, some visionBERT methods based on cross-modal attention have achieved excellent performance in many vision-language tasks not only in academia but also in industry. Nevertheless, the expensive computation cost of cross-modal attention makes it impractical for large-scale search in industrial applications. In this work, we present a tree-based combo-attention network (TCAN) which has been recently launched in Baidu's dynamic video advertising platform. It provides a practical solution to deploy the heavy cross-modal attention for the large-scale query-to-video search. After launching tree-based combo-attention network, click-through rate gets improved by 2.29\% and conversion rate get improved by 2.63\%.