<LK Lab, CoT> The CoT Collection: Improving Zero-shot and Few-shot Learning of Language MOdels via Chain-of-Thought Fine-Tuning (2023.10)

<LK Lab, CoT> The CoT Collection: Improving Zero-shot and Few-shot Learning of Language MOdels via Chain-of-Thought Fine-Tuning (2023.10)

관심있는 NLP 논문을 읽어보고 간단히 정리했습니다. (Language & Knowledge Lab의 Retreival 관련) 혹시 부족하거나 잘못된 내용이 있다면 댓글 부탁드립니다 ️ [KAIST] - new instruction-tuning dataset, CoT Collection - 1.84 million rationales across 1,060 tasks - 100B 이하 사이즈의 LM을 CoT Collection으로 instruction tuning하여, 4개의 domain-specific tasks에 대해 강한 few-shot learning 성능을 보임 배경 Chain-of-Thought (CoT) prompting을 이용하여 LM이 여러 태스크에서 훌륭한 성능을 보일 수 있도록 만..

원문링크 : <LK Lab, CoT> The CoT Collection: Improving Zero-shot and Few-shot Learning of Language MOdels via Chain-of-Thought Fine-Tuning (2023.10)

 등록된 다른 글

<DB, Agent> [FunSearch] Mathematical discoveries from program search with largelanguage models (2023.12)

<DB, Agent> [FunSearch] Mathematical discoveries from program search with largelanguage models (2023.12)

<Retrieval> [GenRead] Generate rather than Retrieval: Large Language Models are Strong Context Generators (2023.01)

<Retrieval> [GenRead] Generate rather than Retrieval: Large Language Models are Strong Context Generators (2023.01)

2023(상반기) 삼성전자 DX SCSA 전형 서류 불합격 후기

2023(상반기) 삼성전자 DX SCSA 전형 서류 불합격 후기

Into to linear regression

Into to linear regression

<Instruction> Self-Alignment with Instruction Backtranslation

<Instruction> Self-Alignment with Instruction Backtranslation

Quiz & Assignment

Quiz & Assignment

<LLM> Llama 2: Open Foundation and Fine-Tuned Chat Models

<LLM> Llama 2: Open Foundation and Fine-Tuned Chat Models

Programming Assignments

Programming Assignments