AutoPP: Towards Automated Product Poster Generation and Optimization

2025-12-26

AI Summary

To address the high labor cost of manual product poster creation and low efficiency of online optimization, this paper proposes an end-to-end CTR-driven framework for automatic poster generation and optimization. Methodologically: (1) we introduce IDPO (Isolated Design-element Preference Optimization), an element-level isolation mechanism enabling fine-grained CTR attribution and user preference alignment; (2) we construct AutoPP1M, the first million-scale poster generation–feedback dataset; (3) we integrate a unified design module, conditional token encoding, and online A/B-style element replacement. The framework achieves state-of-the-art performance in both offline evaluations and large-scale online A/B tests over tens of millions of real traffic impressions, yielding statistically significant CTR improvements. All code and the AutoPP1M dataset are publicly released.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Learning Preferences or Rankings
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Product posters blend striking visuals with informative text to highlight the product and capture customer attention. However, crafting appealing posters and manually optimizing them based on online performance is laborious and resource-consuming. To address this, we introduce AutoPP, an automated pipeline for product poster generation and optimization that eliminates the need for human intervention. Specifically, the generator, relying solely on basic product information, first uses a unified design module to integrate the three key elements of a poster (background, text, and layout) into a cohesive output. Then, an element rendering module encodes these elements into condition tokens, efficiently and controllably generating the product poster. Based on the generated poster, the optimizer enhances its Click-Through Rate (CTR) by leveraging online feedback. It systematically replaces elements to gather fine-grained CTR comparisons and utilizes Isolated Direct Preference Optimization (IDPO) to attribute CTR gains to isolated elements. Our work is supported by AutoPP1M, the largest dataset specifically designed for product poster generation and optimization, which contains one million high-quality posters and feedback collected from over one million users. Experiments demonstrate that AutoPP achieves state-of-the-art results in both offline and online settings. Our code and dataset are publicly available at: https://github.com/JD-GenX/AutoPP

Problem

• Automates product poster creation without human intervention
• Optimizes poster click-through rates using online feedback
• Generates posters from basic product information and unified design

Innovation

• Automated pipeline for product poster generation and optimization
• Element rendering module encodes design elements into condition tokens
• Optimizer uses IDPO to attribute CTR gains to isolated elements