Metraipolis › archive › dc_ce16f2b76cac

Local Entropy Fluctuations for Black-Box LLM-Generated Text Detection — arXiv:2610.33767

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.33767
License (as archived)
CC0-1.0
Archived
2026-10-27T07:57:20.660Z
SHA-256 of the archived text
cf948e1c0a24ca65ae61e754745753f8655ed5d4c8bad6bec9d12bd3ce28b3c3
Stable URL of this page
https://metraipolis.com/documento/dc_ce16f2b76cac

How to cite

Local Entropy Fluctuations for Black-Box LLM-Generated Text Detection — arXiv:2610.33767 — arXiv (en), https://arxiv.org/abs/2610.33767 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ce16f2b76cac (Metraipolis archive dc_ce16f2b76cac, sha256 cf948e1c0a24ca65ae61e754745753f8655ed5d4c8bad6bec9d12bd3ce28b3c3)
BibTeX
@misc{dcce16f2b76cac,
  title = {Local Entropy Fluctuations for Black-Box LLM-Generated Text Detection — arXiv:2610.33767},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33767},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ce16f2b76cac; sha256 cf948e1c0a24ca65ae61e754745753f8655ed5d4c8bad6bec9d12bd3ce28b3c3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33767. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Zero-shot detection of machine-generated text is vital for mitigating the risks of large language models (LLMs) misuse. While the state-of-the-art dual-model detectors and resampling-based detectors outperform single-model statistical detectors on unprompted continuations, they suffer a catastrophic breakdown on instruction-conditioned text in the black-box setting, where the source model is unknown. We analyze this failure and trace it to poor prompt conditioning for resampling-based methods and diminished cross-model discrepancy for dual-model detectors. Motivated by recent studies that human writing maintains a steady, uniform information flow, whereas machine-generated text exhibits sharp, localized entropy fluctuations, we propose a new metric, the Relative Peak Prominence Rate (RelPPR) to capture this structural contrast along the entropy trajectory and further leverage it as a signal for LLM-generated text detection. Extensive evaluations across diverse model families and datasets demonstrate that RelPPR achieves state-of-the-art black-box detection performance and remains the most reliable detector under various practical situations. Our code is available at this https URL .

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/2610.33767 — 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 2026-10-27T07:55:30.665Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).