<LK Lab, Evaluation> [FLASK] Fine-Grained Language Model Evaluation Based on Alignment Skill Sets (2023.10)

<LK Lab, Evaluation> [FLASK] Fine-Grained Language Model Evaluation Based on Alignment Skill Sets (2023.10)

관심있는 NLP 논문을 읽어보고 간단히 정리했습니다. (Language & Knowledge Lab의 Retreival 관련) 혹시 부족하거나 잘못된 내용이 있다면 댓글 부탁드립니다 ️ [KAIST] - 각 instruciton에 대해 coarse-level scoring을 skill set-level scoring로 분해 - human & mode based 평가에 대한 fine-grained evaluation protocol, FLASK - fine-graininess of evaluation은 holistic view를 획득하는데 중요 배경 기존 LLM 평가 방식은 single metric이라서 LLM의 능력을 평가하기에 불충분 또한 surface form에 대해 sensitive하기 때문에..

원문링크 : <LK Lab, Evaluation> [FLASK] Fine-Grained Language Model Evaluation Based on Alignment Skill Sets (2023.10)

 등록된 다른 글

<Attention> LongNet: Scaling Transformers to 1,000,000,000 Tokens

<Attention> LongNet: Scaling Transformers to 1,000,000,000 Tokens

[Upstage] 업스테이지 AI Solution Reliability Engineer 합격 후기 (정규직 전환형 인턴십) (비전공자)

[Upstage] 업스테이지 AI Solution Reliability Engineer 합격 후기 (정규직 전환형 인턴십) (비전공자)

Making Multiple Mappings, deciding if these are reversible

Making Multiple Mappings, deciding if these are reversible

Matrices transform into the new basis vector set

Matrices transform into the new basis vector set

<Retrieval, Knowledge Injection> Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs (2023.12)

<Retrieval, Knowledge Injection> Fine-Tuning or Retrieval? Comparing Knowledge Injection in LLMs (2023.12)

YOLO Algorithm

YOLO Algorithm

[BOJ] 15657 : N과 M (8) [백트랙킹](Python)

[BOJ] 15657 : N과 M (8) [백트랙킹](Python)

Time saving rules

Time saving rules