← Back to the feed
Energy & materialsResearch findings

Unified Image-Text Advertisement Generation Based on User Behavior

Original published 2026.05.12 22:58 KSTAdded to Blackdeck 2026.10.20 19:18 KST

What has changed?

A unified autoregressive model jointly generates personalized image and text advertisements from user click history

Key points

  1. Proposes a method to generate personalized advertisements based on user click history
  2. Uses a unified autoregressive model to simultaneously generate image and text content
  3. Effectively captures user interests through coarse-to-fine preference understanding

Connect it to your work

Enables accurate personalization by capturing user interests from historical behavior

What remains unproven

This research note is based on an abstract; check the original for test conditions and application scope.

Originals used for this note

Links open the original publisher in a new tab.

  1. arxivDesign Your Ad: Personalized Advertising Image and Text Generation with Unified Autoregressive Models