RoLLMRec: a robust LLM-based recommender system for defending against shilling and prompt injection attacks

Abstract

Large Language Models (LLMs) are increasingly being integrated into recommender systems, offering contextual reasoning, cross-domain adaptability, and natural language interaction. However, their adoption also introduces vulnerabilities such as prompt injection, semantic poisoning, and shilling attacks, which can distort recommendations and erode user trust. Addressing these risks is essential for the safe deployment of LLM-based recommenders. We propose RoLLMRec, a defense oriented architectura

Categories

Framework mappings

Suggested from the entry's categories.

@article{shehmir2026rollmrec,
  title = {{RoLLMRec: a robust LLM-based recommender system for defending against shilling and prompt injection attacks}},
  author = {Sarama Shehmir and Rasha Kashef},
  year = {2026},
  month = mar,
  journal = {Frontiers in Computer Science},
  doi = {10.3389/fcomp.2026.1735253},
  url = {https://doi.org/10.3389/fcomp.2026.1735253}
}