RLHF
PPO
기존의 LLM 학습 방법론1: The 3 Conventional Feature-Based and Finetuning Approaches 언어모델을 target task에 맞게 adapting하거나 finetuning하는 방법론. Finetuning Large Language Models Feature-Based Approach 사전 학습된 LLM을 로드...
Real-time machine learning: challenges and solutions