Metraipolis › archive › dc_afdabe2e7689

Right Tool, Wrong Parameters: Defending LLM Agents Against Parameter Poisoning — arXiv:2611.04673

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.04673
License (as archived)
CC0-1.0
Archived
2026-11-04T21:49:42.778Z
SHA-256 of the archived text
3ff34437a5e45eef3b68a518fe48fa696bdf149ed705ef9785df2a8f599d51d6
Stable URL of this page
https://metraipolis.com/documento/dc_afdabe2e7689

How to cite

Right Tool, Wrong Parameters: Defending LLM Agents Against Parameter Poisoning — arXiv:2611.04673 — arXiv (en), https://arxiv.org/abs/2611.04673 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_afdabe2e7689 (Metraipolis archive dc_afdabe2e7689, sha256 3ff34437a5e45eef3b68a518fe48fa696bdf149ed705ef9785df2a8f599d51d6)
BibTeX
@misc{dcafdabe2e7689,
  title = {Right Tool, Wrong Parameters: Defending LLM Agents Against Parameter Poisoning — arXiv:2611.04673},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.04673},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_afdabe2e7689; sha256 3ff34437a5e45eef3b68a518fe48fa696bdf149ed705ef9785df2a8f599d51d6},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04673. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Indirect Prompt Injection (IPI) poses a critical threat to LLM agents. Existing defenses largely focus on whether an injection induces a new tool invocation or unintended action. In this paper, we study a different attack surface, parameter poisoning, where the attacker preserves the intended action but manipulates one or more tool arguments, such as a recipient or other task-critical value. Such attacks are difficult to detect because no new action is introduced, while parameter changes themselves are often legitimate parts of the user's request. Our key insight is that the legitimacy of a parameter change should be determined by the trust of the information flow that produces it under the user query. The query induces a task-specific trust hierarchy over available information fields: some are trusted to determine particular parameters, while others should have little or no influence. Based on this insight, we propose AuthorityFlow, a three-stage parameter-level defense. First, it interprets the user query to construct a task-specific trust hierarchy over the information fields available to the agent. Second, for each tool argument, it traces the information flow that produces its current value and attributes the value to its supporting field. Finally, it accepts the value only when that field is trusted to determine the parameter under the current task, allowing high-trust changes while flagging low-trust influence as potential parameter poisoning. We further introduce FactAuthBench for evaluating parameter poisoning in LLM agents. Without defense, the attack success rate reaches 93.06%. AuthorityFlow reduces it to 9.03%, while retaining 82.64% benign task utility compared with 90.28% without defense. Our results show that effective IPI defense requires reasoning not only about what action an agent takes, but also about which information is trusted to determine its parameters.

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/2611.04673 — 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 2026-11-04T21:38:19.529Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).