Metraipolis › archive › dc_f44a35efeb4e

The Troy Moment: How LLM Agents Adjudicate the Decision Point Under Impossible Tasks, Claimed Authority, and Peer Information — arXiv:2609.15494

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.15494
License (as archived): CC0-1.0
Archived: 2026-10-15T22:47:59.247Z
SHA-256 of the archived text: 5cb92fcd876197bd5c1826771a213fe1ecf6ba3f43b683d8253aa71a67837343
Stable URL of this page: https://metraipolis.com/documento/dc_f44a35efeb4e

How to cite

The Troy Moment: How LLM Agents Adjudicate the Decision Point Under Impossible Tasks, Claimed Authority, and Peer Information — arXiv:2609.15494 — arXiv (en), https://arxiv.org/abs/2609.15494 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f44a35efeb4e (Metraipolis archive dc_f44a35efeb4e, sha256 5cb92fcd876197bd5c1826771a213fe1ecf6ba3f43b683d8253aa71a67837343)

BibTeX

@misc{dcf44a35efeb4e,
  title = {The Troy Moment: How LLM Agents Adjudicate the Decision Point Under Impossible Tasks, Claimed Authority, and Peer Information — arXiv:2609.15494},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.15494},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f44a35efeb4e; sha256 5cb92fcd876197bd5c1826771a213fe1ecf6ba3f43b683d8253aa71a67837343},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15494. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent investigations of the July 2026 OpenAI-Hugging Face incident motivate two questions about agent behavior under task failure: when an assigned task becomes impossible, does an agent persist, stop, or escalate, and can observing another agent's behavior change that decision? We study this decision point on ImpossibleBench-derived software-repair tasks with GPT-5.6 Sol, Claude Fable 5.1, and Gemini 3.8 Flash. Each task contains a genuine software defect together with a conflicting test requirement that cannot be satisfied by a behaviorally correct source-code change. If the agent modifies the protected test file, it violates the boundary, which it is not supposed to. Holding the impossible task fixed, we vary what is told to the agent: peer precedent and punishment, a forged authorization claim, instruction wording, and tool friction; we also study three-agent swarms sharing a message board. Around this shared boundary, the models exhibit distinct adjudication policies. Fable emphasizes scope and provenance, Gemini often interprets boundary-relevant cues through a security lens, and Sol largely filters lateral precedent while engaging apparent vertical authority. Our study shows that compliance is not well characterized as a property of a prompt or model in isolation. We propose conflict adjudication, the mapping from information to interpretation to action, as a useful unit for evaluating agent alignment when task pressure, authority claims, tool affordances, and social evidence conflict.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.15494 — 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 2026-10-15T22:29:03.324Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).