Metraipolis › archive › dc_d253eeea77b5

MPIDetector: Detecting Multi-modal Prompt Injection via LLM-based Reasoning — arXiv:2610.30202

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.30202
License (as archived): CC0-1.0
Archived: 2026-10-25T20:56:05.943Z
SHA-256 of the archived text: f9f8b1fb0f3e90d12ff2256aff26cea4cf1c414696c714c621e44c5efd96d6a7
Stable URL of this page: https://metraipolis.com/documento/dc_d253eeea77b5

How to cite

MPIDetector: Detecting Multi-modal Prompt Injection via LLM-based Reasoning — arXiv:2610.30202 — arXiv (en), https://arxiv.org/abs/2610.30202 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d253eeea77b5 (Metraipolis archive dc_d253eeea77b5, sha256 f9f8b1fb0f3e90d12ff2256aff26cea4cf1c414696c714c621e44c5efd96d6a7)

BibTeX

@misc{dcd253eeea77b5,
  title = {MPIDetector: Detecting Multi-modal Prompt Injection via LLM-based Reasoning — arXiv:2610.30202},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30202},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d253eeea77b5; sha256 f9f8b1fb0f3e90d12ff2256aff26cea4cf1c414696c714c621e44c5efd96d6a7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30202. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

With the rapid integration of Multi-modal Large Language Models (MLLMs) into real-world applications, Prompt Injection (PI) has emerged as a critical security threat. Attackers can conceal malicious instructions within textual narratives or visual inputs to bypass safety alignment. Existing approaches, such as signature- and perplexity-based methods, rely heavily on pattern matching, making them insufficient to detect sophisticated, cross-modal PI payloads. In this paper, we propose a novel detection framework, called MPIDetector, that leverages LLMs' intrinsic reasoning capabilities via few-shot In-Context Learning (ICL). Instead of fine-tuning, we provide the model with specific prompt templates and representative attack examples, enabling it to scrutinize the semantic intent of input across modalities. We constructed a comprehensive evaluation benchmark comprising eight textual and seven visual malicious categories, supplemented with benign datasets to rigorously test for false positives. Our evaluation across multiple leading LLMs, including Gemini-2.0-Flash, Qwen3-VL-32B-Instruct, and Doubao-seed-1.6-flash, demonstrates the efficacy and generalizability of this approach. The proposed method achieves up to 90.91% accuracy and 100% recall on textual tasks. In the visual domain, it achieves up to 99.28% accuracy while maintaining 100% precision across all evaluated models. Crucially, when evaluated against 500 highly obfuscated environmental web-agent injections (WebInject), MPIDetector achieved a detection rate up to 99.80%. Comparisons with two baseline approaches for detecting textual PI payloads, along with two additional methods for visual input detection, demonstrate MPIDetector's superior accuracy. These results show that LLM-driven reasoning provides a robust, cross-modal defense against advanced prompt injections.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.30202 — 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 2026-10-25T20:47:56.967Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).