Metraipolis › archive › dc_831abc402d03
Pragmatic Attack Surface: Vulnerabilities of Implicit Context in Large Language Models — arXiv:2608.09551
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.09551
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:50:26.662Z
- SHA-256 of the archived text
7ddba37cb22a8afc5fe54ad6c34794a969f9c1347000eafee5edc80876409b26- Stable URL of this page
https://metraipolis.com/documento/dc_831abc402d03
How to cite
Pragmatic Attack Surface: Vulnerabilities of Implicit Context in Large Language Models — arXiv:2608.09551 — arXiv (en), https://arxiv.org/abs/2608.09551 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_831abc402d03 (Metraipolis archive dc_831abc402d03, sha256 7ddba37cb22a8afc5fe54ad6c34794a969f9c1347000eafee5edc80876409b26)
BibTeX
@misc{dc831abc402d03,
title = {Pragmatic Attack Surface: Vulnerabilities of Implicit Context in Large Language Models — arXiv:2608.09551},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.09551},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_831abc402d03; sha256 7ddba37cb22a8afc5fe54ad6c34794a969f9c1347000eafee5edc80876409b26},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.09551. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In the era of large language models (LLMs), attackers often manipulate natural language to elicit unsafe or harmful outputs, creating a new natural language attack surface unique to LLM-based systems, where attacks directly exploit explicit linguistic cues in user prompts to bypass the safety mechanism of LLMs. However, such attacks can often be mitigated by existing safety alignment algorithms. On the other hand, human language is inherently grounded in pragmatics, necessitating typical context to interpret language, e.g., world knowledge, social norms. However, such contexts are often implicit because they are not directly expressed in human language and are not sufficiently leveraged in safety alignment, creating a fundamental mismatch between human language interpretation and safety alignment approaches. In this paper, we demonstrate that this mismatch exposes vulnerabilities in LLMs. We refer to this vulnerability as the pragmatic attack surface, which can be exploited to achieve high attack success rates. The experimental results demonstrate that our proposed approach outperforms baseline attack methods across various open-source and closed-source models by a substantial margin. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-11 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.09551 — 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 2026-10-19T08:49:41.935Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).