Jang Jaehoon

생성형 인공지능 또는 LLM을 기반으로 RAG(검색 증강 생성: Retrieval-Augmented Generation) 시스템을 구축해도 바라는 성능이 나오지 않고 마땅한 해결책이 없을 때는 어떻게 해야 할까요? 

이 강의에서는 인지 부하(Cognitive Load) 이론을 기반으로 RAG 성능을 개선하는 방안을 제시합니다. 이 강의를 통해서 LLM 컨텍스트 창의 한계를 이해하고, RAG 시스템에서 인지 부하를 효과적으로 관리하는 방법을 알 수 있습니다. 청크(Chunk) 크기와 구조 설계, 고품질 청크 생성 기법, 동적 최적화, 성능 평가 및 실전 기법 등을 다루는 실무 수준 이론 강의입니다.

아리가람

2025 블랙 프라이데이 깜짝 연장 할인

RAG 성능의 한계를 뚫는 인지 부하 관리 기술

유익한 강의가 된 것 같아 기쁩니다.

컨텍스트창관리

chunk전략설계

rag시스템구축

llm성능평가및튜닝

동적최적화

What should be done when building a generative AI or LLM-based RAG (Retrieval-Augmented Generation) system, but the desired performance isn't achieved and there's no suitable solution?

This lecture presents methods to improve RAG performance based on Cognitive Load theory. Through this lecture, you will understand the limitations of LLM context windows and learn how to effectively manage cognitive load in RAG systems. It is a practical-level theoretical lecture covering Chunk size and structure design, high-quality Chunk generation techniques, dynamic optimization, performance evaluation, and practical techniques.

arigaram

입문

초급

중급이상

챗지피티

ChatGPT

대규모 언어 모델

검색증강생성

Generative AI

Cognitive Load Management Technology Breaking Through the Limits of RAG Performance

[Basic 1] Course Introduction and Basic Concepts

[Basic 1] Code Implementation Practice

[Basic 2] Context Window and Token Limitations

[Basic 2] Code Implementation Practice

[Basic 3] Chunking Strategy: Chunk Size and Structure

[Basic 3] Code Implementation Practice

[Basic 4] High-Quality Chunk Generation Techniques

[Basic 4] Code Implementation Practice

[Basic 5] RAG Pipeline Integration Step 1: Preprocessing

[Basic 5] Code Implementation Practice

[Basic 6] RAG Pipeline Integration Stage 2: Retrieval (Similarity Search and Filtering)

[Basic 6] Code Implementation Practice

[Basic 7] RAG Pipeline Integration Step 3: Chunk Reconstruction and Prompt Redesign

[Basic 7] Code Implementation Practice

[Basic 8] RAG Pipeline Integration Step 4: LLM Answer Generation and Post-processing

[Basic 8] Code Implementation Practice

[Basic 9] RAG Pipeline Integration 5 Steps: Hallucination Detection and Re-input Strategy

[Basic 9] Code Implementation Practice

[Basic 10] RAG Pipeline Integration Practice: Building a RAG System that Responds to Simple Questions

[Advanced 1] Dynamic Optimization: Question Complexity Assessment

[Advanced 1] Code Implementation Practice

[Advanced 2] Dynamic Optimization: Dynamic Chunk Size Adjustment Algorithm

[Advanced 2] Code Implementation Practice

[Advanced 3] Dynamic Optimization: Adaptive Summary Parameter Adjustment

[Advanced 3] Code Implementation Practice

[Advanced 4] Dynamic Optimization: Context Accumulation Management in Multi-turn Conversations

[Advanced 4] Code Implementation Practice

[Advanced 5] Dynamic Optimization: System Monitoring and Feedback Loop Design

[Advanced 5] Code Implementation Practice

[Advanced 6] Performance Evaluation: Measuring Recall and Precision

[Advanced 6] Code Implementation Practice

[Advanced 7] Performance Evaluation: Latency Analysis and Cost Analysis

[Advanced 7] Code Implementation Practice

[Advanced 8] Performance Evaluation: Token Usage Efficiency Metrics

[Advanced 8] Code Implementation Practice

[Advanced 9] Performance Evaluation: User Satisfaction Assessment Methodology

[Advanced 10] Performance Evaluation: Strategy Validation through A/B Testing

[Advanced 10] Code Implementation Practice

[Advanced 11] Technology Outlook: Research Tasks for Fully Automated Chunk Optimization

[Advanced 12] Technology Outlook: Long-term Memory Integration

[Advanced 12] Code Implementation Practice

[Advanced 13] Technology Outlook: Large-Scale Document-Based RAG Scenarios

[Advanced 13] Code Implementation Practice

[Advanced 14] Technology Outlook: Multimodal Information Processing

[Advanced 14] Code Implementation Practice

[Advanced 15] RAG Performance Improvement Project Implementation Guide: 1. Project Topic Selection and Proposal Writing

[Advanced 16] RAG Performance Improvement Project Implementation Guide: 2. Building a Pipeline for Data Collection and Preprocessing

[Advanced 16] Code Implementation Practice

[Advanced 17] RAG Performance Improvement Project Implementation Guide: 3. Chunking Strategy Design and Implementation

[Advanced 17] Code Implementation Practice

[Advanced 18] RAG Performance Improvement Project Implementation Guide: 4. RAG System Integration and Tuning

[Advanced 18] Code Implementation Practice

[Advanced 19] RAG Performance Improvement Project Implementation Guide: 5. Performance Evaluation and Evaluation Report Writing

[Advanced 19] Code Implementation Practice

[Advanced 20] RAG Performance Improvement Project Implementation Guide: 6. Final Presentation and Code Review

[Advanced 20] Code Implementation Practice

(Long-term plan: Schedule TBD) [Supplement 1] Supplement to "Chunking Strategy: Size and Structure of Chunks"