Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
Yuntao Bai; Andy Jones; Kamal Ndousse; Amanda Askell; Anna Chen; Nova DasSarma; Dawn Drain; Stanislav Fort; et al. · 2022 · arXiv (Cornell University) background low priority queued
Main argument
None
Why it matters here
None
Reading notes
Anthropic's HHH/RLHF assistant paper; the operational definition of 'aligned' the corpus critiques. Cite-only. 4 seed-connections.
Yuntao Bai; Andy Jones; Kamal Ndousse; Amanda Askell; Anna Chen; Nova DasSarma; Dawn Drain; Stanislav Fort; et al. (2022). Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2204.05862