S08_Retrieval - trankhoidang/RAG-wiki GitHub Wiki
✏️ Page Contributors: Khoi Tran Dang
🕛 Creation date: 26/06/2024
📥 Last Update: 26/06/2024
In the retrieval stage, our goal is to find information most relevant to the user query. Common approaches include semantic retrieval, which identifies the most similar dense embeddings using cosine similarity, and keyword-based retrieval, which uses traditional keyword search algorithms to assess the relevance of each text to the query. Hybrid search combines both dense (semantic) and sparse (keyword) search methods for more comprehensive results.
For more details on retrieval strategies, please check Advanced RAG.