Time-Weighted Retriever

A Time-Weighted Retriever is a retriever that takes into account recency in addition to similarity. The scoring algorithm is:

let score = (1.0 - this.decayRate) ** hoursPassed + vectorRelevance;

Notably, hoursPassed above refers to the time since the object in the retriever was last accessed, not since it was created. This means that frequently accessed objects remain "fresh" and score higher.

this.decayRate is a configurable decimal number between 0 and 1. A lower number means that documents will be "remembered" for longer, while a higher number strongly weights more recently accessed documents.

Note that setting a decay rate of exactly 0 or 1 makes hoursPassed irrelevant and makes this retriever equivalent to a standard vector lookup.

Usage

This example shows how to intialize a TimeWeightedVectorStoreRetriever with a vector store. It is important to note that due to required metadata, all documents must be added to the backing vector store using the addDocuments method on the retriever, not the vector store itself.

tip

See this section for general instructions on installing integration packages.

npm

npm install @langchain/openai

Yarn

yarn add @langchain/openai

pnpm

pnpm add @langchain/openai

import { TimeWeightedVectorStoreRetriever } from "langchain/retrievers/time_weighted";
import { MemoryVectorStore } from "langchain/vectorstores/memory";
import { OpenAIEmbeddings } from "@langchain/openai";

const vectorStore = new MemoryVectorStore(new OpenAIEmbeddings());

const retriever = new TimeWeightedVectorStoreRetriever({
  vectorStore,
  memoryStream: [],
  searchKwargs: 2,
});

const documents = [
  "My name is John.",
  "My name is Bob.",
  "My favourite food is pizza.",
  "My favourite food is pasta.",
  "My favourite food is sushi.",
].map((pageContent) => ({ pageContent, metadata: {} }));

// All documents must be added using this method on the retriever (not the vector store!)
// so that the correct access history metadata is populated
await retriever.addDocuments(documents);

const results1 = await retriever.invoke("What is my favourite food?");

console.log(results1);

/*
[
  Document { pageContent: 'My favourite food is pasta.', metadata: {} }
]
 */

const results2 = await retriever.invoke("What is my favourite food?");

console.log(results2);

/*
[
  Document { pageContent: 'My favourite food is pasta.', metadata: {} }
]
 */

API Reference:

• TimeWeightedVectorStoreRetriever from langchain/retrievers/time_weighted
• MemoryVectorStore from langchain/vectorstores/memory
• OpenAIEmbeddings from @langchain/openai