Advanced search
1 file | 1.13 MB Add to list

Reinforcement learning from human feedback in LLMs : whose culture, whose values, whose perspectives?

Author
Organization
Project
Abstract
We argue for the epistemic and ethical advantages of pluralism in Reinforcement Learning from Human Feedback (RLHF) in the context of Large Language Models (LLMs). Drawing on social epistemology and pluralist philosophy of science, we suggest ways in which RHLF can be made more responsive to human needs and how we can address challenges along the way. The paper concludes with an agenda for change, i.e. concrete, actionable steps to improve LLM development.
Keywords
Artifcial intelligence, Large language models, Social epistemology, Pluralism, Standpoint theory

Downloads

  • s13347-025-00861-0.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 1.13 MB

Citation

Please use this url to cite or link to this publication:

MLA
Gonzalez Barman, Kristian, et al. “Reinforcement Learning from Human Feedback in LLMs : Whose Culture, Whose Values, Whose Perspectives?” PHILOSOPHY & TECHNOLOGY, vol. 38, no. 2, 2025, doi:10.1007/s13347-025-00861-0.
APA
Gonzalez Barman, K., Lohse, S., & de Regt, H. W. (2025). Reinforcement learning from human feedback in LLMs : whose culture, whose values, whose perspectives? PHILOSOPHY & TECHNOLOGY, 38(2). https://doi.org/10.1007/s13347-025-00861-0
Chicago author-date
Gonzalez Barman, Kristian, Simon Lohse, and Henk W. de Regt. 2025. “Reinforcement Learning from Human Feedback in LLMs : Whose Culture, Whose Values, Whose Perspectives?” PHILOSOPHY & TECHNOLOGY 38 (2). https://doi.org/10.1007/s13347-025-00861-0.
Chicago author-date (all authors)
Gonzalez Barman, Kristian, Simon Lohse, and Henk W. de Regt. 2025. “Reinforcement Learning from Human Feedback in LLMs : Whose Culture, Whose Values, Whose Perspectives?” PHILOSOPHY & TECHNOLOGY 38 (2). doi:10.1007/s13347-025-00861-0.
Vancouver
1.
Gonzalez Barman K, Lohse S, de Regt HW. Reinforcement learning from human feedback in LLMs : whose culture, whose values, whose perspectives? PHILOSOPHY & TECHNOLOGY. 2025;38(2).
IEEE
[1]
K. Gonzalez Barman, S. Lohse, and H. W. de Regt, “Reinforcement learning from human feedback in LLMs : whose culture, whose values, whose perspectives?,” PHILOSOPHY & TECHNOLOGY, vol. 38, no. 2, 2025.
@article{01JQ6S84KD962P2TVC5CFR0XTN,
  abstract     = {{We argue for the epistemic and ethical advantages of pluralism in Reinforcement Learning from Human Feedback (RLHF) in the context of Large Language Models (LLMs). Drawing on social epistemology and pluralist philosophy of science, we suggest ways in which RHLF can be made more responsive to human needs and how we can address challenges along the way. The paper concludes with an agenda for change, i.e. concrete, actionable steps to improve LLM development.}},
  articleno    = {{35}},
  author       = {{Gonzalez Barman, Kristian and Lohse, Simon and de Regt, Henk W.}},
  issn         = {{2210-5433}},
  journal      = {{PHILOSOPHY & TECHNOLOGY}},
  keywords     = {{Artifcial intelligence,Large language models,Social epistemology,Pluralism,Standpoint theory}},
  language     = {{eng}},
  number       = {{2}},
  pages        = {{26}},
  title        = {{Reinforcement learning from human feedback in LLMs : whose culture, whose values, whose perspectives?}},
  url          = {{http://doi.org/10.1007/s13347-025-00861-0}},
  volume       = {{38}},
  year         = {{2025}},
}

Altmetric
View in Altmetric