Towards Reinforcement Learning with AI Feedback (RLAIF). What open-sourced foundation models, instruction tuning, and other recent events mean for the future of AI
amatriain.net
Towards Reinforcement Learning with AI Feedback (RLAIF). What open-sourced foundation models, instruction tuning, and other recent events mean for the future of AI
Developing our position on AI - Blog - Recurse Center
recurse.com
The trend is clear. Bet your money on small open-source models, distillation and fine-tuning, serving, and data collection. One or two frontier closed-source models will still be valuable. But not all. https://t.co/dx4W8wG97S