Training a helpful and harmless assistant with reinforcement learning from human feedback

The authors propose a novel approach that incorporates parameter-efficient tuning to better optimize control tokens, thus benefitting controllable generation.

Data and Resources

Cite this as

Yuntao Bai, Andy Jones, Kamal Ndousse, Amanda Askell, Anna Chen, Nova DasSarma, Dawn Drain, Stanislav Fort, Deep Ganguli, Tom Henighan (2024). Dataset: Training a helpful and harmless assistant with reinforcement learning from human feedback. https://doi.org/10.57702/ueb4xymx

DOI retrieved: December 2, 2024

Additional Info

Field Value
Created December 2, 2024
Last update December 2, 2024
Defined In https://doi.org/10.48550/arXiv.2312.09244
Citation
  • https://doi.org/10.48550/arXiv.2403.16649
  • https://doi.org/10.48550/arXiv.2310.00819
  • https://doi.org/10.48550/arXiv.2307.01139
  • https://doi.org/10.48550/arXiv.2406.15568
Author Yuntao Bai
More Authors
Andy Jones
Kamal Ndousse
Amanda Askell
Anna Chen
Nova DasSarma
Dawn Drain
Stanislav Fort
Deep Ganguli
Tom Henighan
Homepage https://arxiv.org/abs/2204.05862