You.com Releases the YouRetriever: The Simplest Interface to the You.com Search API

You.com released the YouRetriever, the simplest interface to the You.com Search API. The You.com Search API was developed with Retrieval Augmented Generation (RAG) applications in mind by LLMs for LLMs. They achieve this by testing our API with various datasets to establish standards for LLM efficiency in the RAG-QA environment. They provided a detailed analysis of the differences and similarities between the You.com Search API and the Google Search API. They provided a framework for evaluating LLMs in an RAG-QA environment. They used the RetrievalQA Chain to assess how well their retrievers did on Hotpot QA as well. A Hotpot dataset includes a query, an answer, and its context. When using the “distractor” mode, in which the LLM must avoid being fooled by intentionally false language, the context can shift about the question/answer. One of the tests involved replacing the dataset’s original context with the fragments of text returned by search APIs. Since APIs must search the entire internet for the desired information rather than relying solely on the list of snippets provided in the dataset, the internet serves as the distractor text in this case. When testing the efficacy of search APIs in conjunction with an LLM, they subject the systems to what they dub the “web distractor” scenario.

When possible, it returns more extensive snippets of information, and shortly, you’ll be able to choose how much text you’d like returned, from a single sample to the complete page. There are 27 results for “great Keith” when using the default parameters, and some documents have some content. For LLMs working in a RAG-QA environment, this makes our search API particularly useful.

They conducted their tests on the HotPotQA dataset. They use the datasets library to retrieve this information from the Huggingface dataset. Here, they use full wiki instead of distractor, but as was previously said, they will generate their context utilizing the search APIs.

Visit https://documentation.you.com/openai-language-model-integration for further instructions on setting up. 

You.com will soon release a broader search study, so stay tuned for more information. Anyone interested in being an early access partner should write to api@you.com with some information about themselves, their use case, and the number of calls they anticipate making each day.

Check out the Blog and Project Page. All Credit For This Research Goes To the Researchers on This Project. Also, don’t forget to join our 32k+ ML SubReddit, 40k+ Facebook Community, Discord Channel, and Email Newsletter, where we share the latest AI research news, cool AI projects, and more.

If you like our work, you will love our newsletter..

We are also on Telegram and WhatsApp.