Newest papers

Prompt Injection Through Forensic Evidence: A Taxonomy and Proof of Concept of Anti-Forensic Attacks on LLM-Assisted Digital Investigations

Frank Breitinger, Jan Gruber, Christopher Hargreaves, Jan-Niclas Hilgert, Gaëtan Michelet, Mark Scanlon · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.29747 · cited by 0 works (OpenAlex) · open-access PDF

Russian

Большие языковые модели (LLM) входят в цифровую криминалистику в качестве экспертных ассистентов и автономных агентов. Цифровая криминалистика переворачивает привычное положение для косвенного инъекционного устройства: доказательства, подлежащие анализу, контролируются нападающим по определению, и эксперт обязан внимательно прочитать его. Это делает слой рассуждений новой антикриминалистической мишенью, рядом с классическими данными и инструментами. Классические антикриминалистические таксономии не охватывают его, а защитные средства быстрого впрыска не предсказывают его. Мы расширяем антикриминалистическую модель, чтобы отразить манипуляцию рассуждениями, и представляем таксономию атак, осуществляемых через противокриминалистические доказательства:

Machine translation (Meta M2M100); the English original follows.

Abstract

Large language models (LLMs) are entering digital forensics as examination assistants and autonomous agents. Digital forensics inverts the usual setting for indirect prompt injection: the evidence under analysis is attacker-controlled by definition, and the examiner is obliged to read it closely. This makes the reasoning layer a new anti-forensic target, next to the classical ones of data and tools. Classical anti-forensic taxonomies do not cover it, and prompt injection defences do not anticipate it. We extend the anti-forensic model to reflect the manipulation of reasoning, and present a taxonomy of attacks carried through adversarial evidence: an attacker selects an objective, a technique, a payload location, and a concealment method. We observe a perception gap, i.e., the fact that deterministic tools, the human examiner, and the model perceive the same evidence differently, which is what lets an injection reach the model unseen by the examiner. We demonstrate three attacks against an LLM-assisted analysis pipeline, changing a reported conclusion, writing attacker-chosen content to the analyst's workstation, and causing outbound network traffic, each driven only by text planted on the seized device. Mitigations to these attacks are also proposed.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Adversarial system, Taxonomy (biology), Computer security, Digital evidence, Digital forensics

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.29747, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange