ProGRank: Probe-Gradient Reranking to Defend Dense-Retriever RAG from Corpus Poisoning

Abstract

Retrieval-Augmented Generation (RAG) improves the reliability of large language model applications by grounding generation in retrieved evidence, but it also introduces a new attack surface: corpus poisoning. In this setting, an adversary injects or edits passages so that they are ranked into the Top-$K$ results for target queries and then affect downstream generation. Existing defences against corpus poisoning often rely on content filtering, auxiliary models, or generator-side reasoning, which

Categories

@misc{yin2026progrank,
  title = {{ProGRank: Probe-Gradient Reranking to Defend Dense-Retriever RAG from Corpus Poisoning}},
  author = {Xiangyu Yin and Yi Qi and Chih-Hong Cheng},
  year = {2026},
  month = mar,
  eprint = {2603.22934},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2603.22934}
}