Evolving Creatives with Clicks: CTR-Driven Advertising image Editing via Attribute Disentanglement and Preference Alignment

Key points

• The optimization objective for e-commerce advertising creatives has shifted from aesthetic quality to online business metrics like click-through rate (CTR).1
• Recent methods generate appealing ad images from product images by constructing CTR-based reward signals and refining prompt generation via Reinforcement Learning.1
• Based on attribute contrasts between images with disparate CTRs, we instruction-tune a prompt model to generate editing suggestions, then refine it with real click preferences since reference-derived suggestions are not uniformly beneficial.1
• Furthermore, we introduce Consistent Preference Optimization (CPO) as ECC's preference optimization strategy, which enforces consistency across semantically equivalent shuffled suggestions to mitigate DPO overfitting and improve preference generalization.1

Sources (1)

• [1]Evolving Creatives with Clicks: CTR-Driven Advertising image Editing via Attribute Disentanglement and Preference Alignment
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 9, 06:26 AM

  In this work, we propose Evolving Creatives with Clicks (ECC), which reformulates creative optimization as fine-grained editing rather than generation from scratch.

  The optimization objective for e-commerce advertising creatives has shifted from aesthetic quality to online business metrics like click-through rate (CTR).

Extractive summary: sentences quoted from the sources.