Metraipolis › archive › dc_c67ea572fb3f

Steering LLMs Responses Towards Moral Foundations on the Norwegian MFQ-30 — arXiv:2609.21636

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.21636
License (as archived)
CC0-1.0
Archived
2026-10-10T01:12:44.476Z
SHA-256 of the archived text
539e3f3f20adf477011f2c13190e7f79d37db375d58b0010cd775f0dbb9b5e13
Stable URL of this page
https://metraipolis.com/documento/dc_c67ea572fb3f

How to cite

Steering LLMs Responses Towards Moral Foundations on the Norwegian MFQ-30 — arXiv:2609.21636 — arXiv (en), https://arxiv.org/abs/2609.21636 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c67ea572fb3f (Metraipolis archive dc_c67ea572fb3f, sha256 539e3f3f20adf477011f2c13190e7f79d37db375d58b0010cd775f0dbb9b5e13)
BibTeX
@misc{dcc67ea572fb3f,
  title = {Steering LLMs Responses Towards Moral Foundations on the Norwegian MFQ-30 — arXiv:2609.21636},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21636},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c67ea572fb3f; sha256 539e3f3f20adf477011f2c13190e7f79d37db375d58b0010cd775f0dbb9b5e13},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21636. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent work applies human psychometric questionnaires to large language models to elicit moral and value profiles, but it is not clear whether these instruments measure anything stable in models or whether the resulting profiles can be moved toward a target human population. We administer the Norwegian Moral Foundations Questionnaire (MFQ-30) to six open-weight LLMs and compare their foundation profiles to a sample of N = 1,282 Norwegian respondents. We test two steering interventions, prompt-level persona steering and activation-level ActAdd. Half the models engage with the questionnaire under our attention check. The other half default to flat or central-tendency outputs that look near-human on average without tracking item content. A neutral Nordic-respondent persona, written without any distributional information from the human sample, brings the engaging models 44-77% closer to the Norwegian mean in Mahalanobis $d^2$. One-pair ActAdd at a fixed mid-layer flattens the foundation profile rather than steering individual foundations. For at least one model the same persona that shifts the profile also induces engagement that was absent at baseline, a concrete instance of the cognitive phantoms that Peereboom et al. (2025) warn about.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-21 in cs.AI).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Computers and Society (cs.CY).
Pagina del lavoro: https://arxiv.org/abs/2609.21636 — 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 2026-10-10T00:23:42.551Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).