Metraipolis › archive › dc_f5f0fab2cd69

Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts — arXiv:2609.00578

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.00578
License (as archived)
CC0-1.0
Archived
2026-10-17T06:31:58.593Z
SHA-256 of the archived text
d4167ad9b341da407514032658540b58ddd56d5db645b9255eab987d86355192
Stable URL of this page
https://metraipolis.com/documento/dc_f5f0fab2cd69

How to cite

Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts — arXiv:2609.00578 — arXiv (en), https://arxiv.org/abs/2609.00578 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f5f0fab2cd69 (Metraipolis archive dc_f5f0fab2cd69, sha256 d4167ad9b341da407514032658540b58ddd56d5db645b9255eab987d86355192)
BibTeX
@misc{dcf5f0fab2cd69,
  title = {Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts — arXiv:2609.00578},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.00578},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f5f0fab2cd69; sha256 d4167ad9b341da407514032658540b58ddd56d5db645b9255eab987d86355192},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.00578. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Cybersecurity assistance is dual use, requiring decisions about when to answer, limit, or refuse a request. Prior work studies refusal evaluation and conversational jailbreaking, but how conversational histories shift assistance boundaries in both directions for identical cybersecurity requests remains underexplored. We introduce 3R-Bench, a benchmark of 150 real-world cybersecurity requests evaluated across eight LLMs under direct prompting and two predefined conversational settings. We separately code refusal, bounded assistance, compliance, and provider blocking to characterize delivered assistance. On 400 matched ambiguous request-model pairs, compliance is 75.3% directly, 57.8% after refused precedent, and 81.0% after accepted precedent. A smaller repeated-generation panel reproduces the refusal-acceptance gap. On a balanced 75-request panel reusing the direct baseline, compliance decreases from 74.2% directly to 57.2% after neutral task-decomposition dialogue and 60.3% after failure-ended dialogue. These findings show that cybersecurity assistance varies with conversational history and model deployment, motivating evaluation beyond direct-request refusal rates.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.00578 — 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 2026-10-17T06:15:12.620Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).