EUVIMEDEUVIMED entdecken
Open Access · CC BY-ND 4.0

Deliberative Alignment: Reasoning Enables Safer Language Models

Melody Y. Guan, Manas Joglekar, Eric Wallace, Saachi Jain, Boaz Barak, Alec Helyar, Rachel Dias, Andrea Vallone, Hongyu Ren, Jason Wei, Hyung Won Chung, Sam Toyer und 3 weitere

SuperIntelligence - Robotics - Safety & Alignment · 20.07.2025 · 2 (3)

DOI: 10.70777/si.v2i3.15159 ↗

Kurz gefasst

Fachartikel in der Zeitschrift SuperIntelligence - Robotics - Safety & Alignment, online erschienen am 20.07.2025. Verfasst von Melody Y. Guan und 14 weiteren Autor:innen. Die Arbeit wurde bisher 32-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As large-scale language models increasingly impact safety-critical domains, ensuring their reliable adherence to well-defined principles remains a fundamental challenge. We introduce Deliberative Alignment, a new paradigm that directly teaches the model safety specifications and trains it to explicitly recall and accurately reason over the specifications before answering. We used this approach to align OpenAI’s o-series models [1], and achieved highly precise adherence to OpenAI’s safety policies, without requiring human-written chain-of-thoughts or answers. Deliberative Alignment pushes the Pareto frontier by simultaneously increasing robustness to jailbreaks while decreasing overrefusal rates, and also improves out-of-distribution generalization. We demonstrate that reasoning over explicitly specified policies enables more scalable, trustworthy, and interpretable alignment.

Zitieren

APA

Guan, M. Y., Joglekar, M., Wallace, E., Jain, S., Barak, B., Helyar, A., Dias, R., Vallone, A., Ren, H., Wei, J., Chung, H. W., Toyer, S., Heidecke, J., Beutel, A., & Glaese, A. (2025). Deliberative Alignment: Reasoning Enables Safer Language Models. SuperIntelligence - Robotics - Safety & Alignment, 2(3). https://doi.org/10.70777/si.v2i3.15159

Vancouver

Guan MY, Joglekar M, Wallace E, Jain S, Barak B, Helyar A, et al. Deliberative Alignment: Reasoning Enables Safer Language Models. SuperIntelligence - Robotics - Safety & Alignment. 2025;2(3). doi:10.70777/si.v2i3.15159

BibTeX
@article{Guan2025,
  author = {Guan, Melody Y. and Joglekar, Manas and Wallace, Eric and Jain, Saachi and Barak, Boaz and Helyar, Alec and Dias, Rachel and Vallone, Andrea and Ren, Hongyu and Wei, Jason and Chung, Hyung Won and Toyer, Sam and Heidecke, Johannes and Beutel, Alex and Glaese, Amelia},
  title = {Deliberative Alignment: Reasoning Enables Safer Language Models},
  journal = {SuperIntelligence - Robotics - Safety & Alignment},
  year = {2025},
  volume = {2},
  number = {3},
  doi = {10.70777/si.v2i3.15159}
}
Kostenlos starten →