Metraipolis › archive › dc_416f60dc7601

NormDetect: Anchored Calibration for Low-False-Positive Detection of Machine-Generated Web Text — arXiv:2610.41459

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.41459
License (as archived)
CC0-1.0
Archived
2026-11-02T04:52:10.506Z
SHA-256 of the archived text
166a27ec2cbfc36a24f3d71179a42d83fe0e54cd0529fd71984977b9797c1961
Stable URL of this page
https://metraipolis.com/documento/dc_416f60dc7601

How to cite

NormDetect: Anchored Calibration for Low-False-Positive Detection of Machine-Generated Web Text — arXiv:2610.41459 — arXiv (en), https://arxiv.org/abs/2610.41459 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_416f60dc7601 (Metraipolis archive dc_416f60dc7601, sha256 166a27ec2cbfc36a24f3d71179a42d83fe0e54cd0529fd71984977b9797c1961)
BibTeX
@misc{dc416f60dc7601,
  title = {NormDetect: Anchored Calibration for Low-False-Positive Detection of Machine-Generated Web Text — arXiv:2610.41459},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.41459},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_416f60dc7601; sha256 166a27ec2cbfc36a24f3d71179a42d83fe0e54cd0529fd71984977b9797c1961},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.41459. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Web platforms screen posts for text written by large language models, and every false flag hits a human writer. Detectors must therefore run at a very low false positive rate (FPR). Zero-shot detectors such as Binoculars score text with frozen language models, but invisible edits such as homoglyphs shatter tokenization and push machine text into the human range. Unicode normalization undoes these edits, yet the normalized score still misses half of the machine text at a strict FPR, under attacks that leave nothing to clean. We find that with the Falcon-7B scoring pair, heads on token statistics recall less than the normalized score at this FPR, and several flag too many human posts in some sections. We propose NormDetect, which anchors a learned per-section head to the normalized zero-shot score. A blend weight chosen on training folds under a human FPR cap combines the two scores, and one threshold is calibrated on human text only. On eight sections of the RAID benchmark with eleven attacks, NormDetect raises recall at a 0.1% calibrated FPR by 5.0 points over normalized Binoculars. Its human FPR bound stays at most 1% in every section, and no other detector within this bound recalls as much.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.41459 — 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 2026-11-02T04:45:12.893Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).