<LK Lab, Evaluation> Prometheus: Inducing Fine-Grained Evaluation Capability in Language Models (2023.10)

<LK Lab, Evaluation> Prometheus: Inducing Fine-Grained Evaluation Capability in Language Models (2023.10)

관심있는 NLP 논문을 읽어보고 간단히 정리했습니다. (Language & Knowledge Lab의 Retreival 관련) 혹시 부족하거나 잘못된 내용이 있다면 댓글 부탁드립니다 ️ [KAIST, Naver AI, University of Washington, MIT] - GPT-4의 평가 수준에 준하는 완전한 오픈소스 LLM, Prometheus - customized score rubric 기반으로 long-form text를 평가할 수 있는 13B 사이즈의 평가자 LLM, Prometheus - 1K score rubrics, 20K instructions, 100K responses로 구성된 데이터셋, Feedback Collection 배경 최근 LLM을 evaluator로 사용하고자 ..

원문링크 : <LK Lab, Evaluation> Prometheus: Inducing Fine-Grained Evaluation Capability in Language Models (2023.10)

 등록된 다른 글

Building AI Projects(3)

Building AI Projects(3)

<Self> SELF: Self-Evolution with Language Feedback (2023.10)

<Self> SELF: Self-Evolution with Language Feedback (2023.10)

Optimization Algorithms(1) - Mini-batch

Optimization Algorithms(1) - Mini-batch

Matrices as objects that map one vector onto another; all the types of matrices

Matrices as objects that map one vector onto another; all the types of matrices

<Multi-modal> PointLLM: Empowering Large Language Models to Understand Point Clouds

<Multi-modal> PointLLM: Empowering Large Language Models to Understand Point Clouds

Making Multiple Mappings, deciding if these are reversible

Making Multiple Mappings, deciding if these are reversible

<Agent> Agents: An Open-source Framework for Autonomous Language Agents

<Agent> Agents: An Open-source Framework for Autonomous Language Agents

[BOJ] 14500 : 테트로미노 [브루트포스](Python)

[BOJ] 14500 : 테트로미노 [브루트포스](Python)