![[논문리뷰]: Memory Retrieval and Consolidation in Large Language Models through Function Tokens](https://moonlight-paper-snapshot.s3.ap-northeast-2.amazonaws.com/arxiv/memory-retrieval-and-consolidation-in-large-language-models-through-function-tokens-4.png)
[논문리뷰]: Memory Retrieval and Consolidation in Large Language Models through Function Tokens
논문 링크 Function Token Hypothesis: LLM “기억 검색”의 게이트는 왜 구두점과 개행인가한 줄 요약 (TL;DR)이 논문은 LLM의 기억을 “feature의 재활성화(retrieval) + 학습을 통한 feature 확장 …
51분
Interpretability
feature-retrieval
pretraining-dynamics
steering
token-frequency
![[논문리뷰] Marconi: Prefix Caching for the Era of Hybrid LLMs](https://pbs.twimg.com/media/GdyLXO9W4AADox0.jpg)