[논문리뷰] Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent InteractionZhengkang Guo이 arXiv에 게시한 'Controllable Memory Usage: Balancing Anchoring and Innovation in Long-Term Human-Agent Interaction' 논문에 대한 자세한 리뷰입니다.#Review#Long-Term Human-Agent Interaction#Controllable Memory#Memory Anchoring#Large Language Models (LLMs)#Personalization#Reinforcement Learning (RL)#Supervised Fine-Tuning (SFT)#Memory Dependence2026년 1월 12일댓글 수 로딩 중