Metraipolis › archive › dc_a895b91ec773

The Devil is in the Context: Nine Dimensions of Prompt Injection Attacks — arXiv:2611.04672

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04672
License (as archived)
CC0-1.0
Archived
2026-11-04T21:49:42.349Z
SHA-256 of the archived text
30cd2b123ac49b3e2c382bb2ee9aa3955111d08c7ee839cee3ce00f4e9ea00cc
Stable URL of this page
https://metraipolis.com/documento/dc_a895b91ec773

How to cite

The Devil is in the Context: Nine Dimensions of Prompt Injection Attacks — arXiv:2611.04672 — arXiv (en), https://arxiv.org/abs/2611.04672 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a895b91ec773 (Metraipolis archive dc_a895b91ec773, sha256 30cd2b123ac49b3e2c382bb2ee9aa3955111d08c7ee839cee3ce00f4e9ea00cc)
BibTeX
@misc{dca895b91ec773,
  title = {The Devil is in the Context: Nine Dimensions of Prompt Injection Attacks — arXiv:2611.04672},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04672},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a895b91ec773; sha256 30cd2b123ac49b3e2c382bb2ee9aa3955111d08c7ee839cee3ce00f4e9ea00cc},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04672. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Prompt injection is usually discussed and evaluated as a single, homogeneous type of attack. However, vulnerabilities reported in real-world systems show that prompt injections vary widely. Moreover, the same instruction can be a prompt injection in one context and a legitimate request in another, depending on the user's intent and the system's policies. Ignoring this variation leaves gaps in how defenses are evaluated, and can give a false sense of security against real-world attacks. We identify nine dimensions along which prompt injection attacks differ. Together, they form a taxonomy that describes the injected instruction itself, its relation to the user's task, and how it manifests in the target system. Each dimension affects which defenses can detect or prevent an attack. We first use the taxonomy to characterize six real-world prompt injection vulnerabilities, showing how much these attacks differ. We then apply it to nine prompt injection benchmarks and find that most cover only a small region of this space. Combining benchmarks helps, but 13% of pairwise category combinations remain untested, mostly among the less commonly represented categories. Finally, we show that the taxonomy supports automated annotation of benchmarks and targeted generation of new test cases to fill these gaps. The taxonomy is also useful beyond benchmarks. Defense authors can use it to state which types of prompt injection they evaluated against, and system operators can use it to identify which attacks apply to their system and choose defenses to match.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2611.04672 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-04T21:38:19.528Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).