大模型
arxiv.org ·
引导 LLM 回应趋向挪威 MFQ-30 上的道德基础
该 arXiv 预印本将挪威语道德基础问卷 MFQ-30 施测于六个开源权重 LLM，并与 1,282 名挪威受访者样本对比。结果显示仅一半模型通过注意力检查，其余输出趋中；不含人类分布信息的中性北欧受访者 persona 可使参与模型在 Mahalanobis d² 上靠近挪威均值 44-77%；固定中层单对 ActAdd 会压平而非定向调整道德基础画像，且同一 persona 对至少一个模型诱发了基线不存在的参与行为。
AI推荐理由 arXiv 预印本，方法描述具体：对六个开源权重 LLM 施测挪威语 MFQ-30，与 N=1,282 挪威受访者样本对比，并测试 prompt 级 persona steering 与激活级 ActAdd 两种干预。新信息在于：一半模型通过注意力检查、另一半输出趋中；中性北欧受访者 persona 在 Mahalanobis d² 上使参与模型靠近挪威均值 44-77%；固定中层单对 ActAdd 反而压平道德基础画像；并报告同一 persona 可诱发基线不存在的参与行为，呼应 Peereboom 等（2025）的 cognitive phantoms 警告。局限是仅六个模型、单一问卷与挪威样本，ActAdd 结论限于特定层与单对设置，且为预印本未经同行评审。
72
正文
Computer Science > Computation and Language
arXiv:2609.21636 (cs)
[Submitted on 18 Sep 2026]
Title:Steering LLMs Responses Towards Moral Foundations on the Norwegian MFQ-30
Authors:Hans Andersen, David Dichas
View a PDF of the paper titled Steering LLMs Responses Towards Moral Foundations on the Norwegian MFQ-30, by Hans Andersen and 1 other authors
View PDF
HTML (experimental)
Abstract: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.
Comments:
Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
Cite as:
arXiv:2609.21636 [cs.CL]
(or
arXiv:2609.21636v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.21636
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Hans Andersen [view email]
[v1]
Fri, 18 Sep 2026 11:25:22 UTC (546 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Steering LLMs Responses Towards Moral Foundations on the Norwegian MFQ-30, by Hans Andersen and 1 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-09
Change to browse by:
cs
cs.AI
cs.CY
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)