Metraipolis › archive › dc_3be4ecc89f84

Opening the Box Changes the Cat: LLMs Detect Chain-of-Thought Monitoring - What now? — arXiv:2610.40946

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.40946
License (as archived): CC0-1.0
Archived: 2026-11-02T07:28:47.096Z
SHA-256 of the archived text: 3135475aa703d494dc38bd9c42eacb612867d4c63e34133e8f08f9dbb6eab76f
Stable URL of this page: https://metraipolis.com/documento/dc_3be4ecc89f84

How to cite

Opening the Box Changes the Cat: LLMs Detect Chain-of-Thought Monitoring - What now? — arXiv:2610.40946 — arXiv (en), https://arxiv.org/abs/2610.40946 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3be4ecc89f84 (Metraipolis archive dc_3be4ecc89f84, sha256 3135475aa703d494dc38bd9c42eacb612867d4c63e34133e8f08f9dbb6eab76f)

BibTeX

@misc{dc3be4ecc89f84,
  title = {Opening the Box Changes the Cat: LLMs Detect Chain-of-Thought Monitoring - What now? — arXiv:2610.40946},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40946},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3be4ecc89f84; sha256 3135475aa703d494dc38bd9c42eacb612867d4c63e34133e8f08f9dbb6eab76f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40946. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Chain-of-thought (CoT) reasoning allows large language models (LLMs) to generate intermediate reasoning traces in natural language while solving complex tasks. These traces can support safety monitoring by exposing potentially harmful plans before actions are executed. In this work, we investigate the conditions under which LLMs infer the presence of CoT monitoring from direct and indirect environmental cues and examine how this inference affects their subsequent behavior. Our observations suggest that oversight is an interactive process: while a monitor inspects a model's reasoning, traces left by monitoring interventions can allow the model to infer that its thoughts are being observed and to adapt its behavior accordingly. We show that even subtle environmental cues, such as timing-related signals from timeouts and operation durations, can reveal the presence of monitoring and trigger models to adapt their behavior. Across repeated episodes and interaction loops, we further observe that models can form statistical estimates of the monitor's effectiveness and use these estimates to guide attempts to evade detection. These findings highlight the need to evaluate CoT-based oversight under adaptive conditions, where the monitoring process itself may influence the behavior it seeks to observe.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.40946 — 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 2026-11-02T07:26:57.657Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).