Towards Personalized Bundle Creative Generation with Contrastive Non-Autoregressive Decoding

TL;DR

This paper introduces a contrastive non-autoregressive model for personalized bundle creative generation, effectively producing diverse and user-specific promotional content with improved quality and efficiency.

Contribution

It presents a novel contrastive non-autoregressive approach tailored for generating personalized bundle creatives, addressing both quality and speed in real-world applications.

Findings

Significant improvements in creative quality over baseline models.

Faster generation speed compared to autoregressive methods.

Effective capture of user preferences for personalized content.

Abstract

Current bundle generation studies focus on generating a combination of items to improve user experience. In real-world applications, there is also a great need to produce bundle creatives that consist of mixture types of objects (e.g., items, slogans and templates) for achieving better promotion effect. We study a new problem named bundle creative generation: for given users, the goal is to generate personalized bundle creatives that the users will be interested in. To take both quality and efficiency into account, we propose a contrastive non-autoregressive model that captures user preferences with ingenious decoding objective. Experiments on large-scale real-world datasets verify that our proposed model shows significant advantages in terms of creative quality and generation speed.