SEARCH

AI 지식 검색

전체 #AI Agent #AI Assistant #AI Lab #AI Research #AI Safety #Alignment #Anthropic #Attention #BERT #Benchmark #CUDA #ChatGPT #Claude #Context Window #Data Contamination #Data Provenance

RESULTS

#Reinforcement Learning 문서

Reinforcement Learning from Human Feedback (RLHF)

사람의 선호·평가를 보상 신호로 활용해 모델 행동을 조정하는 학습 접근법.

#Alignment#LLM#RLHF#Reinforcement Learning

2026.08.29 06:47