Whose Norms? New preprint disentangles cultural vs personal alignment in LLMs
radamihalcea · x · 2026-10-22
A new preprint, "Whose Norms? Disentangling Cultural and Personal Alignment in Large Language Models," asks whose standards LLMs follow when cultural norms conflict with personal preferences. The team released the paper, dataset, code and website, and will present at the Copenhagen NLP Symposium and EMNLP. The work focuses on cultural alignment and socially grounded LLM evaluation.
More from Research
- Nature's anti-portfolio: Fermi, Krebs and the first MRI paper were all rejected — luismbat · 2026-10-22
- Carbon-X training mix revealed: vertebrates dominate, human genomes favored in post-training — LoubnaBenAllal1 · 2026-10-22
- Larochelle amplifies debate: should papers be required to submit their code? — hugo_larochelle · 2026-10-22
- New psycholinguistics preprint dissects the weird emerging 'dialect' of frontier AI models — tallinzen · 2026-10-22
- New paper: AI agents skip backups 85% of the time — a fix to keep agent actions reversible — shi_weiyan · 2026-10-22
- Carbon-X author admits the DNA model is best kept to already-sequenced organisms — ATinyGreenCell · 2026-10-22