RAG-Survey

Why we included this project

If you want to understand the retrieval-augmented generation field as a whole before committing to a particular approach, this repository is a solid starting point. It is the companion to the survey 'Retrieval-Augmented Generation for AI-Generated Content: A Survey' and organizes hundreds of papers into a taxonomy of foundations, enhancements, and applications, so you can trace how RAG moved from early retrieval-based language models to today's multimodal, code, audio, and 3D generation systems. Each entry links to its arXiv or proceedings record, which makes the collection work both as a structured reading path and as a bibliography when you are writing a paper or scoping a research direction. Developers will find it just as useful for spotting techniques worth borrowing, such as query reformulation or nearest-neighbor memory, when designing a pipeline. Keep in mind it is a curated reference map maintained alongside the evolving survey, not a runnable application, so treat it as a guide rather than something to deploy.