Metraipolis › archive › dc_f8589c1925a8

On the Indistinguishability of Human v/s AI Generated Text — arXiv:2608.26797

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.26797
License (as archived): CC0-1.0
Archived: 2026-10-17T16:12:22.634Z
SHA-256 of the archived text: 6d264aaa2f5fea5305d8ec7eab95e097d677ace4ab0effbc102a815b1bd74bee
Stable URL of this page: https://metraipolis.com/documento/dc_f8589c1925a8

How to cite

On the Indistinguishability of Human v/s AI Generated Text — arXiv:2608.26797 — arXiv (en), https://arxiv.org/abs/2608.26797 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f8589c1925a8 (Metraipolis archive dc_f8589c1925a8, sha256 6d264aaa2f5fea5305d8ec7eab95e097d677ace4ab0effbc102a815b1bd74bee)

BibTeX

@misc{dcf8589c1925a8,
  title = {On the Indistinguishability of Human v/s AI Generated Text — arXiv:2608.26797},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.26797},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f8589c1925a8; sha256 6d264aaa2f5fea5305d8ec7eab95e097d677ace4ab0effbc102a815b1bd74bee},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.26797. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

The rapid improvement of LLMs has made distinguishing AI-generated text from human writing a pressing problem. This challenge is further amplified by paraphrasing tools designed to make machine-generated text appear more "human". We study how access to human writing samples can be used to strategically paraphrase machine-generated responses toward the human distribution. Under a multi-sample setting with human and machine responses to the same prompts, we show that repeated paraphrasing moves the machine distribution toward the empirical human distribution under simple mixing and stability conditions. Our results derive an explicit convergence rate, extend the analysis to a finite-sample setting, and characterize how the required number of human samples and paraphrasing rounds scale with the desired error.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-28 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.26797 — 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 2026-10-17T15:34:49.927Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-28?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).