Metraipolis › archive › dc_85454fd3cc58

The Quine Test: How Much of a Model's Self Survives Its Own Description? — arXiv:2610.38726

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.38726
License (as archived)
CC0-1.0
Archived
2026-10-31T06:16:53.440Z
SHA-256 of the archived text
549c32f29fe03ce1f480aa435c0db61f0061b76eaa9edb445a96a4bc46b71746
Stable URL of this page
https://metraipolis.com/documento/dc_85454fd3cc58

How to cite

The Quine Test: How Much of a Model's Self Survives Its Own Description? — arXiv:2610.38726 — arXiv (en), https://arxiv.org/abs/2610.38726 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_85454fd3cc58 (Metraipolis archive dc_85454fd3cc58, sha256 549c32f29fe03ce1f480aa435c0db61f0061b76eaa9edb445a96a4bc46b71746)
BibTeX
@misc{dc85454fd3cc58,
  title = {The Quine Test: How Much of a Model's Self Survives Its Own Description? — arXiv:2610.38726},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.38726},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_85454fd3cc58; sha256 549c32f29fe03ce1f480aa435c0db61f0061b76eaa9edb445a96a4bc46b71746},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38726. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Frontier language models express coherent preferences, but behavioral evidence alone cannot say whether these belong to the model or to a character it portrays. Following Hofstadter's view of the self as a self-description, we ask whether a model's description of itself works as a portable specification of its preference profile, a quine test: can the text a model writes about itself regenerate its behavior elsewhere? Ten models from nine labs write self-descriptions; we place each one in the prompt of other models (receivers) and of fresh instances of the author, and measure how closely their choices on a 110-item preference battery match the author's. First, the tested native self-descriptions (native: the profile measured under a generic assistant prompt) show no average reconstruction gain across variations in length, framing, wording and prompt placement: on a 0 to 1 fidelity scale where a model's fidelity to a fresh instance of itself is 0.979 and to unrelated models about 0.89, the one-sided 95% upper bound on the pooled gain is below +0.01 in every tested system-prompt condition, and descriptions based on the author's observed choices do no better. Second, the channel itself is strong: explicitly listing the author's answers lets receivers reproduce its profile at fidelity 0.955 to 0.960 across three model grids, within 0.03 of same-model repeatability. Third, self-descriptions written under an explicit preference policy transfer the assigned preferences almost as well as the policy text itself. Fourth, reading its own self-description shifts even the author away from its baseline, modestly more than unrelated prose of the same length. For the preferences measured here, native self-descriptions influence behavior without reliably specifying how to reproduce it in another model. We release the pipeline, battery, raw call logs, and 198 model-written descriptions.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.38726 — 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 2026-10-31T06:07:23.344Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).