Retrieval strategies

[anada10]

Currently we use agents to craft the query, we would either like more standard retrieval strategies that can be added out of the box in our RAG connector (TS, Python, Mongo). Using langchain

Name | Indexing/Retrieval | Index Type | Uses an LLM | When to Use | Description -- | -- | -- | -- | -- | -- Vectorstore | Retrieval | Vectorstore | No | If you are just getting started and looking for something quick and easy. | This is the simplest method and the one that is easiest to get started with. It involves creating embeddings for each piece of text. ParentDocument | Retrieval | Vectorstore + Document Store | No | If your pages have lots of smaller pieces of distinct information that are best indexed by themselves, but best retrieved all together. | This involves indexing multiple chunks for each document. Then you find the chunks that are most similar in embedding space, but you retrieve the whole parent document and return that (rather than individual chunks). Multi Vector | Indexing + Retrieval | Vectorstore + Document Store | Sometimes during indexing | If you are able to extract information from documents that you think is more relevant to index than the text itself. | This involves creating multiple vectors for each document. Each vector could be created in a myriad of ways - examples include summaries of the text and hypothetical questions. Self Query | Retreival | Vectorstore | Yes | If users are asking questions that are better answered by fetching documents based on metadata rather than similarity with the text. | This uses an LLM to transform user input into two things: (1) a string to look up semantically, (2) a metadata filer to go along with it. This is useful because oftentimes questions are about the METADATA of documents (not the content itself). Contextual Compression | Retrieval | Any | Sometimes | If you are finding that your retrieved documents contain too much irrelevant information and are distracting the LLM. | This puts a post-processing step on top of another retriever and extracts only the most relevant information from retrieved documents. This can be done with embeddings or an LLM. Time-Weighted Vectorstore | Retrieval | Vectorstore | No | If you have timestamps associated with your documents, and you want to retrieve the most recent ones | This fetches documents based on a combination of semantic similarity (as in normal vector retrieval) and recency (looking at timestamps of indexed documents) Multi-Query Retriever | Retrieval | Any | Yes | If users are asking questions that are complex and require multiple pieces of distinct information to respond | This uses an LLM to generate multiple queries from the original one. This is useful when the original query needs pieces of information about multiple topics to be properly answered. By generating multiple queries, we can then fetch documents for each of them. Ensemble | Retrieval | Any | No | If you have multiple retrieval methods and want to try combining them. | This fetches documents from multiple retrievers and then combines them. Long-Context Reorder | Retrieval | Any | No | If you are working with a long-context model and noticing that it's not paying attention to information in the middle of retrieved documents. | This fetches documents from an underlying retriever, and then reorders them so that the most similar are near the beginning and end. This is useful because it's been shown that for longer context models they sometimes don't pay attention to information in the middle of the context window. Name Indexing/Retrieval Index Type Uses an LLM When to Use Description Vectorstore Retrieval Vectorstore No If you are just getting started and looking for something quick and easy. This is the simplest method and the one that is easiest to get started with. It involves creating embeddings for each piece of text. ParentDocument Retrieval [Vectorstore](https://python.langchain.com/docs/modules/data_connection/retrievers/vectorstore) + Document Store No If your pages have lots of smaller pieces of distinct information that are best indexed by themselves, but best retrieved all together. This involves indexing multiple chunks for each document. Then you find the chunks that are most similar in embedding space, but you retrieve the whole parent document and return that (rather than individual chunks). Multi Vector Indexing + Retrieval Vectorstore + Document Store Sometimes during indexing If you are able to extract information from documents that you think is more relevant to index than the text itself. This involves creating multiple vectors for each document. Each vector could be created in a myriad of ways - examples include summaries of the text and hypothetical questions. Self Query Retreival Vectorstore Yes If users are asking questions that are better answered by fetching documents based on metadata rather than similarity with the text. This uses an LLM to transform user input into two things: (1) a string to look up semantically, (2) a metadata filer to go along with it. This is useful because oftentimes questions are about the METADATA of documents (not the content itself). Contextual Compression Retrieval Any Sometimes If you are finding that your retrieved documents contain too much irrelevant information and are distracting the LLM. This puts a post-processing step on top of another retriever and extracts only the most relevant information from retrieved documents. This can be done with embeddings or an LLM. Time-Weighted Vectorstore Retrieval Vectorstore No If you have timestamps associated with your documents, and you want to retrieve the most recent ones This fetches documents based on a combination of semantic similarity (as in normal vector retrieval) and recency (looking at timestamps of indexed documents) Multi-Query Retriever Retrieval Any Yes If users are asking questions that are complex and require multiple pieces of distinct information to respond This uses an LLM to generate multiple queries from the original one. This is useful when the original query needs pieces of information about multiple topics to be properly answered. By generating multiple queries, we can then fetch documents for each of them. Ensemble Retrieval Any No If you have multiple retrieval methods and want to try combining them. This fetches documents from multiple retrievers and then combines them. Long-Context Reorder Retrieval Any No If you are working with a long-context model and noticing that it's not paying attention to information in the middle of retrieved documents. This fetches documents from an underlying retriever, and then reorders them so that the most similar are near the beginning and end. This is useful because it's been shown that for longer context models they sometimes don't pay attention to information in the middle of the context window.

[tomlynchRNA]

closed for #151