Metraipolis › archive › dc_102feb8146f4

Steerable Pluralistic Policy Alignment from Anonymized Crowd Feedback — arXiv:2610.26914

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.26914
License (as archived)
CC0-1.0
Archived
2026-10-21T07:07:17.837Z
SHA-256 of the archived text
ef37af026443cf713eb862fabd64735f01342e7edb95d8ba27d5f336bb598844
Stable URL of this page
https://metraipolis.com/documento/dc_102feb8146f4

How to cite

Steerable Pluralistic Policy Alignment from Anonymized Crowd Feedback — arXiv:2610.26914 — arXiv (en), https://arxiv.org/abs/2610.26914 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_102feb8146f4 (Metraipolis archive dc_102feb8146f4, sha256 ef37af026443cf713eb862fabd64735f01342e7edb95d8ba27d5f336bb598844)
BibTeX
@misc{dc102feb8146f4,
  title = {Steerable Pluralistic Policy Alignment from Anonymized Crowd Feedback — arXiv:2610.26914},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.26914},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_102feb8146f4; sha256 ef37af026443cf713eb862fabd64735f01342e7edb95d8ba27d5f336bb598844},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.26914. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Pluralistic alignment requires AI systems to learn from diverse populations whose preferences may reflect heterogeneous or even conflicting objectives, and to personalize their behavior to individual users. Existing approaches in pluralistic alignment often assume access to annotator identities and rely on powerful pretrained representations to model individual preference variations. We propose Preference-PCA Pluralistic Learning (P3L), a novel framework for pluralistic alignment that jointly learns trajectory embeddings and the shared basis of preference directions via PCA decomposition from heterogeneous crowd feedback without requiring annotator identities. At test time, our approach enables rapid and accurate personalized reward learning and downstream policy optimization. Experiments across multiple embodied control tasks demonstrate that P3L enables steerable personal alignment that generalizes beyond the offline dataset to out-of-distribution new users and attains competitive RL and test performance, while substantially reducing computational cost and supporting variable adaptation query budgets.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Robotics (cs.RO).
Pagina del lavoro: https://arxiv.org/abs/2610.26914 — 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 2026-10-21T07:05:43.505Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).