Metraipolis › archive › dc_afabe054b56d

When Personas Become Harmful: Guiding LLMs with Global Citizenship Values Alignment — arXiv:2610.33899

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.33899
License (as archived)
CC0-1.0
Archived
2026-10-27T08:11:27.600Z
SHA-256 of the archived text
c5f63718552f5bc21bb943253c401d29f1609e5e62db0978b33deff0e5e7693c
Stable URL of this page
https://metraipolis.com/documento/dc_afabe054b56d

How to cite

When Personas Become Harmful: Guiding LLMs with Global Citizenship Values Alignment — arXiv:2610.33899 — arXiv (en), https://arxiv.org/abs/2610.33899 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_afabe054b56d (Metraipolis archive dc_afabe054b56d, sha256 c5f63718552f5bc21bb943253c401d29f1609e5e62db0978b33deff0e5e7693c)
BibTeX
@misc{dcafabe054b56d,
  title = {When Personas Become Harmful: Guiding LLMs with Global Citizenship Values Alignment — arXiv:2610.33899},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33899},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_afabe054b56d; sha256 c5f63718552f5bc21bb943253c401d29f1609e5e62db0978b33deff0e5e7693c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33899. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

LLMs are increasingly being used by users from diverse demographics with pluralistic values and distinct backgrounds. Maintaining safety while respecting value-laden personas is uniquely challenging because persona-conditioned requests are often not overtly malicious, unlike those in existing safety benchmarks. Alignment methods aimed at adversarial requests therefore may miss the harm that emerges from persona use, leaving models to either refuse legitimate value expression or blindly follow unsafe personas. We construct a dataset of 11k controversial topics paired with value-based personas mapping to value dimensions from the Schwartz Value Survey. Through an agentic pipeline that iteratively refines the topic and persona, we surface and collect harmful responses from well-aligned commercial LLMs with a more than 47% success rate. We use UNESCO's global citizenship principles as a minimal common ground for distinguishing valid value expressions from harmful behaviors, and introduce a controllable alignment model that lets users balance persona adherence and safety at inference time. The resulting model is safer and more persona-adherent than baselines on most samples.

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/2610.33899 — 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 2026-10-27T07:56:00.213Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).