Visual Inception: Compromising Long-term Planning in Agentic Recommenders via Multimodal Memory Poisoning

Abstract

The evolution from static ranking models to Agentic Recommender Systems (Agentic RecSys) empowers AI agents to maintain long-term user profiles and autonomously plan service tasks. While this paradigm shift enhances personalization, it introduces a vulnerability: reliance on Long-term Memory (LTM). In this paper, we uncover a threat termed "Visual Inception." Unlike traditional adversarial attacks that seek immediate misclassification, Visual Inception injects triggers into user-uploaded images

Categories

Framework mappings

Suggested from the entry's categories.

@misc{qian2026visual,
  title = {{Visual Inception: Compromising Long-term Planning in Agentic Recommenders via Multimodal Memory Poisoning}},
  author = {Jiachen Qian},
  year = {2026},
  month = apr,
  eprint = {2604.16966},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.16966}
}