Metraipolis › archive › dc_65fdfae1633a
Using LLMs to Detect LLM-Generated Texts: A Cross-Generation Analysis — arXiv:2609.34691
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.34691
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:02:50.629Z
- SHA-256 of the archived text
660fdd70069e7c2af07837b736068fbac0b224a2e4eb691148273375c860243e- Stable URL of this page
https://metraipolis.com/documento/dc_65fdfae1633a
How to cite
Using LLMs to Detect LLM-Generated Texts: A Cross-Generation Analysis — arXiv:2609.34691 — arXiv (en), https://arxiv.org/abs/2609.34691 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_65fdfae1633a (Metraipolis archive dc_65fdfae1633a, sha256 660fdd70069e7c2af07837b736068fbac0b224a2e4eb691148273375c860243e)
BibTeX
@misc{dc65fdfae1633a,
title = {Using LLMs to Detect LLM-Generated Texts: A Cross-Generation Analysis — arXiv:2609.34691},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.34691},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_65fdfae1633a; sha256 660fdd70069e7c2af07837b736068fbac0b224a2e4eb691148273375c860243e},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.34691. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Automated detection of LLM-generated texts (LGTs) is critical, yet dedicated detectors often struggle to generalize across domains and models. While general-purpose LLMs offer flexible zero-shot authorship classification with explanatory rationale, their detection behavior, especially regarding self-detection versus cross-detection across model generations, remains poorly understood. We systematically evaluate 15 LLMs spanning three model generations as both generators and detectors. Using a benchmark of 1,000 human-written texts and 15,000 LGTs (1,000 per model), we collected over 233,000 binary classifications alongside natural-language explanations. Our results reveal that detection efficacy is primarily driven by detector capability rather than generator provenance, although outputs from newer generators remain notably harder to detect. Crucially, statistical comparisons show no systematic advantage or disadvantage for self-detection across models. Error analysis further exposes generational bias shifts: first-generation detectors under-detect LGTs (high false-negative rates), second-generation detectors over-flag human texts (high false-positive rates), and the latest models achieve balanced trade-offs. Finally, we highlight significant inconsistencies in how different LLMs apply textual cues to justify their decisions. Code: this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.34691 — 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 2026-10-07T20:29:03.132Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).