Curated by
More in Master AI & LLM Context: From Basics to Production
See all 15 →A Survey of Context Engineering for LLMs — arXiv 2025
This survey introduces Context Engineering as a formal discipline for systematically optimizing information payloads provided to Large Language Models during inference, presenting a comprehensive taxonomy covering context retrieval, processing, management, RAG, memory systems, and multi-agent architectures. The paper analyzes over 1,400 research papers and identifies a critical research gap between models' strong context understanding and their limited ability to generate sophisticated long-form outputs.
Built for AI agentsACO · 1174 tokens
Summary
This survey introduces Context Engineering as a formal discipline for systematically optimizing information payloads provided to Large Language Models during inference, presenting a comprehensive taxonomy covering context retrieval, processing, management, RAG, memory systems, and multi-agent architectures. The paper analyzes over 1,400 research papers and identifies a critical research gap between models' strong context understanding and their limited ability to generate sophisticated long-form outputs.
Tags
context-engineering · large-language-models · survey · retrieval-augmented-generation · prompt-engineering · multi-agent-systems · natural-language-processing
Key entities
Lingrui Mei (person, 0.95) · Context Engineering (concept, 0.99) · Large Language Models (concept, 0.99) · Retrieval-Augmented Generation (concept, 0.95) · Multi-Agent Systems (concept, 0.9) · Tool-Integrated Reasoning (concept, 0.85) · arXiv (organization, 0.95) · Jiayu Yao (person, 0.85) · Yuyao Ge (person, 0.85) · Jiafeng Guo (person, 0.8)
Classification
reference · language en · status final
Provenance
claude-opus-4-6 via @stacklist/mcp-server@2.0.0, confidence 0.85, 2 Jul 2026