Metraipolis › archive › dc_8ec05bd2e4e2

Veracity labels flip the emotional tone of language models but leave racial disparities intact — arXiv:2610.33021

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.33021
License (as archived): CC0-1.0
Archived: 2026-10-27T08:12:32.693Z
SHA-256 of the archived text: 58ce1bcc8ffd5731442f811de1bfdbbf16f9baa0a59324e5b676ed43e3cef13d
Stable URL of this page: https://metraipolis.com/documento/dc_8ec05bd2e4e2

How to cite

Veracity labels flip the emotional tone of language models but leave racial disparities intact — arXiv:2610.33021 — arXiv (en), https://arxiv.org/abs/2610.33021 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8ec05bd2e4e2 (Metraipolis archive dc_8ec05bd2e4e2, sha256 58ce1bcc8ffd5731442f811de1bfdbbf16f9baa0a59324e5b676ed43e3cef13d)

BibTeX

@misc{dc8ec05bd2e4e2,
  title = {Veracity labels flip the emotional tone of language models but leave racial disparities intact — arXiv:2610.33021},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.33021},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8ec05bd2e4e2; sha256 58ce1bcc8ffd5731442f811de1bfdbbf16f9baa0a59324e5b676ed43e3cef13d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.33021. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language models (LLMs) increasingly deliver information through personas, but it is unclear whether correcting the facts also makes their framing consistent across identity labels. We audited four LLMs (GPT-5.4 mini, NVIDIA Nemotron 3 Nano, Llama 4 Scout, and Claude Haiku 4.5), which rated simulated affect on six dimensions (mood, anger, fear, confidence, perceived control, and mental state) in response to fact-checked false and true headlines presented without and with a veracity label. Experiment 1 evaluated 1,000 false and 1,000 true headlines without personas; Experiment 2 evaluated 300 of each under 12 personas crossing race, gender, and age. The label strongly changed the emotional tone: ordinal logistic regressions with headline-cluster bootstrap inference showed that refuting false headlines shifted ratings from mostly negative to mostly positive for GPT, Llama, and Claude (e.g., Claude's "confident" ratings rose from 22% to 99.6%), whereas Nemotron showed smaller, mixed shifts, and confirming true headlines produced only small shifts. The label did not, however, remove persona-associated disparities. The models assigned more negative simulated affect to Black than to White personas in 82 of 84 statistically significant comparisons across models and conditions, both without and with the label, and the disparity persisted on headlines unrelated to race; Asian and female personas were also predominantly assigned more negative affect, whereas age effects varied by model and condition. Factual correction and demographic consistency are therefore distinct properties of LLM behavior, and persona-conditioned systems should be audited for both before deployment in sensitive contexts.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computers and Society (cs.CY) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.33021 — 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 2026-10-27T07:56:00.283Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).