NLP/Prompt & Problem Solving
A Survey of Prompt Engineering Methods in LLMs for Differenct NLP Tasks (2024) 논문 리뷰
아르카눔
2025. 5. 6. 16:05
A Survey of Prompt Engineering Methods in LLMs for Differenct NLP Tasks 논문 (링크)
저자는 Shubham Vatsal와 Harsh Dubey 두 사람이다.
서베이 논문인데 여러가지 prompting 방법들의 목록을 간단하게 정리하고 표를 올리고자 한다.
Prompt Engineering Techniques
- Basic / Standard / Vanilla Prompting
- Chain-of-Thoguht (CoT)
- Self-Consistency
- Ensemble Refinement (ER)
- Automatic Chain-of-Thoguht (Auto CoT)
- Complex CoT
- Program-of-Thoughts (PoT)
- Least-to-Most
- Chain-of-Symbol (CoS)
- Structured Chain-of-Thoguht (SCoT)
- Plan-and-Solve (PS)
- MathPrompter
- Contrastive CoT / Contrastive Self-Consistency
- Federated Same / Different Parameter Self-Consistency / CoT (FED-SP / DP-SC / COT)
- Analogical Reasoning
- Synthetic Prompting
- Tree-of-Thoguhts (ToT)
- Logical Thoughts (LoT)
- Maieutic Prompting
- Verify-and-Edit (VE)
- Reason + Act (ReAct)
- Active-Prompt
- Thread-of-Thought (ThoT)
- Implicit Retrieval Augmented Generation (Implicit RAG)
- System 2 Attention (S2A)
- Instructured Prompting
- Chain-of-Verification (CoVe)
- Chain-of-Knowledge (CoK)
- Chain-of-Code (CoC)
- Program-Aided Language Models (PAL)
- Binder
- Dater
- Chain-of-Table
- Decomposed Prompting (DecomP)
- Three-Hop Reasoning (THOR)
- Metacognitive Prompting (MP)
- Chain-of-Event (CoE)
Taxonomy Diagram of Prompt Engineering Methods Across Different NLP Tasks