Metraipolis › archive › dc_5ddc8cfb2481

Who's Behind the Text? Robust Model Authorship Attribution — arXiv:2610.19583

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19583
License (as archived)
CC0-1.0
Archived
2026-10-15T14:23:28.868Z
SHA-256 of the archived text
63cfd467cadc50785d3890a1ad3572eceeef19a60d53274656fdbd959aba61e3
Stable URL of this page
https://metraipolis.com/documento/dc_5ddc8cfb2481

How to cite

Who's Behind the Text? Robust Model Authorship Attribution — arXiv:2610.19583 — arXiv (en), https://arxiv.org/abs/2610.19583 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5ddc8cfb2481 (Metraipolis archive dc_5ddc8cfb2481, sha256 63cfd467cadc50785d3890a1ad3572eceeef19a60d53274656fdbd959aba61e3)
BibTeX
@misc{dc5ddc8cfb2481,
  title = {Who's Behind the Text? Robust Model Authorship Attribution — arXiv:2610.19583},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19583},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5ddc8cfb2481; sha256 63cfd467cadc50785d3890a1ad3572eceeef19a60d53274656fdbd959aba61e3},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19583. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As large language models (LLMs) are increasingly used to generate text, their misuse has become a growing societal problem, with implications for academic misconduct, model distillation, and cyberattacks. For these reasons, Model Authorship Attribution (MAA), which identifies the source model of a given text, is important for promoting the responsible use of LLMs and improving AI safety. In practice, attribution systems must remain reliable beyond the distributions observed during training. They may encounter texts from unseen domains and writing settings, as well as generations from future versions of model families that were not available when the system was trained. However, existing MAA methods often exhibit substantial performance degradation under such distribution shifts. To address this challenge, we propose Authorship Tracing via Likelihood-based Attribution of Sources (ATLAS), which formulates MAA through source-conditioned text modeling rather than direct source prediction. ATLAS learns how text from each source family is generated and attributes a given text by comparing its likelihood under each learned source condition. We aggregate this likelihood evidence at the token level with clipping so that no single token dominates the decision. We further introduce MAABench, a broad MAA benchmark for evaluating practical OOD settings such as domain and version. Our experiments show that ATLAS consistently outperforms existing baselines across MAABench and OpenTuringBench. In particular, when both the generation domain and the model version change on MAABench, ATLAS achieves 74.27% accuracy, 9.00pp higher than the strongest baseline. Beyond MAA, ATLAS can also help identify distillation source models by measuring which source model a target LLM most closely resembles. Our extended experiments show that ATLAS recovers the known teacher of every distilled model we evaluate.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.19583 — 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 2026-10-15T13:53:35.879Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).