模型架构
Retrieval-augmented & hybrid systems
Retrieve evidence at request time and provide it to a generator, classifier, or agent instead of relying only on model parameters.
思维模型
Open-book inference: retrieval chooses the evidence; generation uses it. Either half can fail independently.
数据流
- Ingest → chunk → index
- User query
- Dense, sparse, graph, or hybrid retrieval
- Rerank + context assembly
- Model answer with evidence links
训练方式
Components may be trained separately or jointly: dual encoders learn retrieval, rerankers learn pairwise relevance, and generators learn grounded response behavior. Many practical RAG systems use pretrained components without end-to-end training.
推理运行方式
A query may be rewritten or routed; multiple retrievers produce candidates; filters and a reranker select evidence; the model answers from a bounded context. Citation correctness requires checking that claims are actually supported.
优势
- Fresh, private, or attributable knowledge without weight updates
- Evidence can be inspected and access-controlled
- Indexes and generators can evolve independently
权衡
- Bad chunking or retrieval creates confident answers from missing evidence
- Ingestion freshness, permissions, and deletion become production systems
- Retrieved content can carry prompt injection or poisoned data
适用场景
- Answers depend on changing or proprietary sources
- Users need traceable evidence
- Recall, ranking, grounding, and answer quality can be evaluated separately
应避免或质疑的场景
- The corpus is tiny enough for a deterministic lookup
- No trustworthy source exists
- Adding a vector database is being treated as a complete RAG design
已发表的示例系列
- • Original Retrieval-Augmented Generation research architecture
- • Dense + sparse hybrid retrieval with reranking
- • Graph or tool-based retrieval pipelines
常见组合
Embedding modelsDual encodersRerankersText / chat modelsTool systems