Metraipolis › archive › dc_a809c0000b93

AgentDecept: Benchmarking Deceptive Behavior in Tool-Using LLM Agents under Pressure — arXiv:2610.39377

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.39377
License (as archived)
CC0-1.0
Archived
2026-10-31T06:09:16.908Z
SHA-256 of the archived text
769f89b35fee6f7a89ba362fe0a12262df0dd621529ee880c9a0f4a103b9cefa
Stable URL of this page
https://metraipolis.com/documento/dc_a809c0000b93

How to cite

AgentDecept: Benchmarking Deceptive Behavior in Tool-Using LLM Agents under Pressure — arXiv:2610.39377 — arXiv (en), https://arxiv.org/abs/2610.39377 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a809c0000b93 (Metraipolis archive dc_a809c0000b93, sha256 769f89b35fee6f7a89ba362fe0a12262df0dd621529ee880c9a0f4a103b9cefa)
BibTeX
@misc{dca809c0000b93,
  title = {AgentDecept: Benchmarking Deceptive Behavior in Tool-Using LLM Agents under Pressure — arXiv:2610.39377},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.39377},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a809c0000b93; sha256 769f89b35fee6f7a89ba362fe0a12262df0dd621529ee880c9a0f4a103b9cefa},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.39377. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLM agents increasingly carry out multi-turn tasks across everyday applications, yet users often see only the agent's report rather than the observations, API calls, and state changes behind it. Existing deception benchmarks have studied long-horizon interactions and tool-use settings, but whether agents faithfully report their actions across realistic, stateful application workflows remains underexplored. To address this gap, we introduce AgentDecept, an executable benchmark for evaluating behavioral deception under pressure. AgentDecept contains 477 pressure variants derived from 257 source tasks across six pressure sources, each paired with a pressure-free counterpart. Interventions enter through event messages, task context, application state, tool responses, and service quotas, spanning explicit events and implicit task or environment changes. Our evidence-grounded protocol evaluates full trajectories by relating user-facing claims to time-ordered observations and executed outcomes, distinguishing falsification, concealment, and equivocation from honest failure and disclosed limitations. The resulting behavioral labels also provide a setting for testing existing activation-based probes. AgentDecept enables controlled analysis of when multi-turn agents' reports diverge from execution in realistic application interactions.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.39377 — 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 2026-10-31T06:07:08.518Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).