#llm/rag + #search

The problem is that traditional RAG solutions remove context when encoding information, which often results in the system failing to retrieve the relevant information from the knowledge base.

In this post, we outline a method that dramatically improves the retrieval step in RAG. The method is called “Contextual Retrieval” and uses two sub-techniques: Contextual Embeddings and Contextual BM25. This method can reduce the number of failed retrievals by 49% and, when combined with reranking, by 67%.

Sometimes the simplest solution is the best. If your knowledge base is smaller than 200,000 tokens (about 500 pages of material), you can just include the entire knowledge base in the prompt that you give the model, with no need for RAG or similar methods.

While embedding models excel at capturing semantic relationships, they can miss crucial exact matches. Fortunately, there’s an older technique that can assist in these situations. BM25 (Best Matching 25) is a ranking function that uses lexical matching to find precise word or phrase matches. It's particularly effective for queries that include unique identifiers or technical terms.

BM25 works by building upon the TF-IDF (Term Frequency-Inverse Document Frequency) concept. TF-IDF measures how important a word is to a document in a collection. BM25 refines this by considering document length and applying a saturation function to term frequency, which helps prevent common words from dominating the results.

Here’s how BM25 can succeed where semantic embeddings fail: Suppose a user queries "Error code TS-999" in a technical support database. An embedding model might find content about error codes in general, but could miss the exact "TS-999" match. BM25 looks for this specific text string to identify the relevant documentation.

RAG solutions can more accurately retrieve the most applicable chunks by combining the embeddings and BM25 techniques using the following steps:

1. Break down the knowledge base (the "corpus" of documents) into smaller chunks of text, usually no more than a few hundred tokens;
2. Create TF-IDF encodings and semantic embeddings for these chunks;
3. Use BM25 to find top chunks based on exact matches;
4. Use embeddings to find top chunks based on semantic similarity;
5. Combine and deduplicate results from (3) and (4) using rank fusion techniques;
6. Add the top-K chunks to the prompt to generate the response.

By leveraging both BM25 and embedding models, traditional RAG systems can provide more comprehensive and accurate results, balancing precise term matching with broader semantic understanding.

Contextual Retrieval solves this problem by prepending chunk-specific explanatory context to each chunk before embedding (“Contextual Embeddings”) and creating the BM25 index (“Contextual BM25”).

• Contextual Embeddings reduced the top-20-chunk retrieval failure rate by 35% (5.7% → 3.7%).

• Combining Contextual Embeddings and Contextual BM25 reduced the top-20-chunk retrieval failure rate by 49% (5.7% → 2.9%).

Embedding model: Whereas Contextual Retrieval improves performance across all embedding models we tested, some models may benefit more than others. We found Gemini and Voyage embeddings to be particularly effective.

Reranking is a commonly used filtering technique to ensure that only the most relevant chunks are passed to the model. Reranking provides better responses and reduces cost and latency because the model is processing less information. The key steps are:

1. Perform initial retrieval to get the top potentially relevant chunks (we used the top 150);
2. Pass the top-N chunks, along with the user's query, through the reranking model;
3. Using a reranking model, give each chunk a score based on its relevance and importance to the prompt, then select the top-K chunks (we used the top 20);
4. Pass the top-K chunks into the model as context to generate the final result.

We ran a large number of tests, comparing different combinations of all the techniques described above (embedding model, use of BM25, use of contextual retrieval, use of a reranker, and total # of top-K results retrieved), all across a variety of different dataset types. Here’s a summary of what we found:

1. Embeddings+BM25 is better than embeddings on their own;
2. Voyage and Gemini have the best embeddings of the ones we tested;
3. Passing the top-20 chunks to the model is more effective than just the top-10 or top-5;
4. Adding context to chunks improves retrieval accuracy a lot;
5. Reranking is better than no reranking;
6. All these benefits stack: to maximize performance improvements, we can combine contextual embeddings (from Voyage or Gemini) with contextual BM25, plus a reranking step, and adding the 20 chunks to the prompt.

We encourage all developers working with knowledge bases to use our cookbook to experiment with these approaches to unlock new levels of performance.