Metraipolis › archive › dc_a262371301ff

Same Judgment, Different Yes: Measuring Multilingual Judgment Profiles in Language Models — arXiv:2610.28066

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.28066
License (as archived)
CC0-1.0
Archived
2026-10-21T07:08:12.667Z
SHA-256 of the archived text
0474583a64e0d45afb4d79c36198b1e81583162ee66f370852d648c1f2c1d07c
Stable URL of this page
https://metraipolis.com/documento/dc_a262371301ff

How to cite

Same Judgment, Different Yes: Measuring Multilingual Judgment Profiles in Language Models — arXiv:2610.28066 — arXiv (en), https://arxiv.org/abs/2610.28066 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a262371301ff (Metraipolis archive dc_a262371301ff, sha256 0474583a64e0d45afb4d79c36198b1e81583162ee66f370852d648c1f2c1d07c)
BibTeX
@misc{dca262371301ff,
  title = {Same Judgment, Different Yes: Measuring Multilingual Judgment Profiles in Language Models — arXiv:2610.28066},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.28066},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a262371301ff; sha256 0474583a64e0d45afb4d79c36198b1e81583162ee66f370852d648c1f2c1d07c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.28066. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multilingual evaluations often score a model's opening yes or no, a token that also carries model and language answering conventions. We ask when different ways of eliciting and reading an answer recover the same judgment, and when they measure an answer habit. We collect 1.1 million replies from eight models in six languages on up to 105 moral vignettes, varying the question's form, the answer constraint, and the reading of the reply, token or sentence, in both wordings (the FAR-6 battery). Story-level judgment profiles converge across the tested readouts and question forms, except for the yes/no token after a negative interrogative read under a preset mapping: on the same replies, token and sentence profiles agree at a median r = 0.67 across cells, while sentence profiles agree with free-sentence profiles at 0.94, a known linguistic variability. With each language's story consensus removed, a model's profile is nearest to its own in another language for 39 of 48 model-language cells under proposition labels and 15 under free sentences, against 5 expected by chance. Particle, position, and named-term preferences follow their own patterns. A model can keep its judgment and change its yes: scoring that token requires establishing which proposition it affirms.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.28066 — 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 2026-10-21T07:05:43.616Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).