Here you can find a database of hundreds of search related active patents and papers. The patents and papers are tagged by SEO and LLMO/GEO related topics, steps of the information retrieval process and the probabilty they could be used nowadays or in the future.
You can navigate and filter the analysis by the internal search or by the tags. It is possible to combine the tags.
Subscribe to my newsletter or follow my RSS feed to make sure you don’t miss any new patents and papers.
To read the full patent and paper analysis and full usage of the SEO Research Suite including AI research assistants you have to sign up for a monthly or yearly membership.
I am very grateful if you support and motivate me and this project with a paid membership, recommendation, reference …
Your advantages as a SEO Research Suite member:
A patent application does not mean that the methods described there will find its way into practice in the search engines. An indication of whether a methodology/technology is so interesting for Google that it could find its way into practice can be obtained by checking whether the patent is pending only in the US or other countries. The claim for a patent priority for other countries must be made 12 months after the first filing.Regardless of whether a patent finds its way into practice, it makes sense to deal with Google patents, as you get an indication of the topics and challenges that product developers at Google and other search engines are dealing with.
This Google research paper introduces a new framework for evaluating the effectiveness of similarity metrics, specifically text embeddings, in the process of data curation for Language Model (LM) pretraining. The read more
This Microsoft patent describes a system that allows a chat agent, powered by an AI language model, to provide more personalized and relevant responses. It achieves this by analyzing a read more
This document introduces a framework called AGREE (Adaptation for GRounding EnhancEment) designed to make Large Language Models (LLMs) more factual and trustworthy. The method fine-tunes an LLM to generate answers read more
This Microsoft patent describes a system for improving search engine results by using a generative AI model. When a user enters a broad search term, the AI generates a more read more
This Microsoft patent describes a system for improving how search engines match user queries with new or “novel” keywords that have little to no user interaction data. It works by read more
This Google patent describes a method for making neural networks more efficient by adding a special component called an “N-grammer layer.” This layer processes input data, like text, by converting read more
This study investigates the performance of Organic Large Language Model (OLLM) traffic, specifically from ChatGPT referrals, compared to traditional digital channels in e-commerce using one year of first-party data from read more
This Google paper introduces BlockRank, a new method to make Large Language Models (LLMs) more efficient at ranking documents. The standard approach, known as In-context Ranking (ICR), is powerful but read more
This patent describes a Google system for identifying high-quality, in-depth articles and displaying them in a special section on the search results page. The system calculates scores to determine if read more
This Google patent describes a browser-based contextual search tool that helps users find information related to web content without navigating away from the original page. The system extracts core content read more
