<Attention> [CALM] LLM Augmented LLMs: Expanding Capabilities through Composition (2024.01)

<Attention> [CALM] LLM Augmented LLMs: Expanding Capabilities through Composition (2024.01)

관심있는 NLP 논문을 읽어보고 간단히 정리했습니다. 혹시 부족하거나 잘못된 내용이 있다면 댓글 부탁드립니다 ️ usechatgpt init success [Google Research, Google DeepMind] - 다른 모델 간의 cross-attention을 통해 새로운 capabilities를 획득하게 하는 기법, CALM - Composition to Augment Language Models - 기존 LLM은 're-using'하면서 새로운 few additional parameters와 data를 사용 - 다양한 도메인과 환경에 적용 가능하다는 특징(장점)을 보유 1. Introduction LLM은 여러 태스크 중에서도 이전과 달리 commonsense 또는 factual reas..

원문링크 : <Attention> [CALM] LLM Augmented LLMs: Expanding Capabilities through Composition (2024.01)

 등록된 다른 글

<CoT, Agent> ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent (2023.12)

<CoT, Agent> ReST meets ReAct: Self-Improvement for Multi-Step Reasoning LLM Agent (2023.12)

Analysis of Algorithms(1) : Analysis of Algorithms Introduction

Analysis of Algorithms(1) : Analysis of Algorithms Introduction

Object Localization

Object Localization

PCA derivation(2) - 복습必

PCA derivation(2) - 복습必

Mean values

Mean values

<Retrieval, In-Context Learning> RAVEN: In-Context Learning with Retrieval Augmented Encoder-Decoder Language Models

<Retrieval, In-Context Learning> RAVEN: In-Context Learning with Retrieval Augmented Encoder-Decoder Language Models

[BOJ] 9019 : DSLR [DFS/BFS](Python)

[BOJ] 9019 : DSLR [DFS/BFS](Python)

Quiz & Assignment

Quiz & Assignment