ADAM: A Systematic Data Extraction Attack on Agent Memory via Adaptive Querying

Xingyu Lyu, Jianfeng He, Ning Wang, Yidan Hu, Tao Li, Danjue Chen, Shixiong Li, Yimin Chen

Abstract

Large Language Model (LLM) agents have achieved rapid adoption and demonstrated remarkable capabilities across a wide range of applications. To improve reasoning and task execution, modern LLM agents would incorporate memory modules or retrieval-augmented generation (RAG) mechanisms, enabling them to further leverage prior interactions or external knowledge. However, such a design also introduces a group of critical privacy vulnerabilities: sensitive information stored in memory can be leaked through query-based attacks. Although feasible, existing attacks often achieve only limited performance, with low attack success rates (ASR). In this paper, we propose ADAM, a novel privacy attack that features data distribution estimation of a victim agent's memory and employs an entropy-guided query strategy for maximizing privacy leakage. Extensive experiments demonstrate that our attack substantially outperforms state-of-the-art ones, achieving up to 100% ASRs. These results thus underscore the urgent need for robust privacy-preserving methods for current LLM agents.
BibTeX
@misc{lyu2026adam,
  author = {Xingyu Lyu and Jianfeng He and Ning Wang and Yidan Hu and Tao Li and Danjue Chen and Shixiong Li and Yimin Chen},
  title = {{ADAM: A Systematic Data Extraction Attack on Agent Memory via Adaptive Querying}},
  year = {2026},
  eprint = {2604.09747},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2604.09747},
}

Loading related content...

Figure