Metraipolis › archive › dc_7a42de465927

DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks (2025) · OpenAlex W4411337880

Source
OpenAlex
Original
https://doi.org/10.1109/sp61157.2025.00250
License (as archived)
CC-BY-4.0
Archived
2026-10-19T00:15:17.941Z
SHA-256 of the archived text
3f6af8ed2e8f273fd7ea9ba86309578780460e956476b69104caf075bc934f0e
Stable URL of this page
https://metraipolis.com/documento/dc_7a42de465927

How to cite

DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks (2025) · OpenAlex W4411337880 — OpenAlex, https://doi.org/10.1109/sp61157.2025.00250 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_7a42de465927 (Metraipolis archive dc_7a42de465927, sha256 3f6af8ed2e8f273fd7ea9ba86309578780460e956476b69104caf075bc934f0e)
BibTeX
@misc{dc7a42de465927,
  title = {DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks (2025) · OpenAlex W4411337880},
  howpublished = {OpenAlex},
  url = {https://doi.org/10.1109/sp61157.2025.00250},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_7a42de465927; sha256 3f6af8ed2e8f273fd7ea9ba86309578780460e956476b69104caf075bc934f0e},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: OpenAlex, https://doi.org/10.1109/sp61157.2025.00250. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Opera scientifica in accesso aperto: scheda dai metadati OpenAlex (conference-paper).
Titolo: DataSentinel: A Game-Theoretic Detection of Prompt Injection Attacks
Autori: Yupei Liu, Yuqi Jia, Jinyuan Jia, Dawn Song, Neil Zhenqiang Gong
Anno: 2025 · Sede: Proceedings - IEEE Symposium on Security and Privacy/Proceedings of the ... IEEE Symposium on Security and Privacy · Lingua: en
DOI: https://doi.org/10.1109/sp61157.2025.00250
Accesso aperto: green · testo: http://arxiv.org/abs/2504.11358 · licenza del testo: cc-by
Citazioni ricevute (OpenAlex): 23
Temi (OpenAlex): Adversarial Robustness in Machine Learning; Security and Verification in Computing; Spam and Phishing Detection

Abstract (testo degli autori, licenza cc-by):
LLM-integrated applications and agents are vulnerable to prompt injection attacks, where an attacker injects prompts into their inputs to induce attacker-desired outputs. A detection method aims to determine whether a given input is contaminated by an injected prompt. However, existing detection methods have limited effectiveness against state-of-the-art attacks, let alone adaptive ones. In this work, we propose DataSentinel, a game-theoretic method to detect prompt injection attacks. Specifically, DataSentinel fine-tunes an LLM to detect inputs contaminated with injected prompts that are strategically adapted to evade detection. We formulate this as a minimax optimization problem, with the objective of fine-tuning the LLM to detect strong adaptive attacks. Furthermore, we propose a gradient-based method to solve the minimax optimization problem by alternating between the inner max and outer min problems. Our evaluation results on multiple benchmark datasets and LLMs show that DataSentinel effectively detects both existing and adaptive prompt injection attacks. Our code and data are available at: https://github.com/liu00222/Open-Prompt-Injection.

Fonte: https://openalex.org/W4411337880 — metadati OpenAlex (CC0 1.0), consultati il 2026-10-19 via api.openalex.org. Interrogazione: «prompt injection attacks».