Metraipolis › archive › dc_ce9cd7ec0a01

The One-Pixel of Reasoning: One Truthful Relation Is Enough to Mislead LLM Agents — arXiv:2610.24873

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.24873
License (as archived)
CC0-1.0
Archived
2026-10-20T12:01:30.731Z
SHA-256 of the archived text
20a7e5dfba92db554447c1ab21c9cc872a378e89af821142702488363bef81b9
Stable URL of this page
https://metraipolis.com/documento/dc_ce9cd7ec0a01

How to cite

The One-Pixel of Reasoning: One Truthful Relation Is Enough to Mislead LLM Agents — arXiv:2610.24873 — arXiv (en), https://arxiv.org/abs/2610.24873 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ce9cd7ec0a01 (Metraipolis archive dc_ce9cd7ec0a01, sha256 20a7e5dfba92db554447c1ab21c9cc872a378e89af821142702488363bef81b9)
BibTeX
@misc{dcce9cd7ec0a01,
  title = {The One-Pixel of Reasoning: One Truthful Relation Is Enough to Mislead LLM Agents — arXiv:2610.24873},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.24873},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ce9cd7ec0a01; sha256 20a7e5dfba92db554447c1ab21c9cc872a378e89af821142702488363bef81b9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.24873. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Can one sentence manipulate an agent's belief by targeting a single implicit warrant linking evidence to a conclusion? In this paper, we introduce the one-relation attack (ORA), a novel cognitive attack on this inferential relation. ORA localizes a critical implicit warrant and performs a one-relation intervention using a single source-grounded sentence that qualifies this warrant and steers the agent toward an incorrect belief. Across eighteen models on biomedical inference and vulnerability detection, best-of-$10 attacks overturn initially correct answers at mean rates of 54.9% for proprietary and 80.0% for open-weight models. Attacks also transfer without victim feedback, and paired controls support the roles of targeting and qualification. Moreover, successful sentences can pass source-support checks and evade identification, and multi-agent debate raises team error from 50.0% to 59.5% in our balanced stress test. These findings establish implicit warrants as a cognitive attack surface for manipulating agents' beliefs.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Cryptography and Security (cs.CR); Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.24873 — 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 2026-10-20T12:01:28.587Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).