{
  "patent_number": "US 12657244",
  "country": "US",
  "title": "AI That Links Information Across Documents for Better Answers",
  "original_title": "Inter-document attention mechanism",
  "summary": "This patent describes an AI system that improves answers to questions by linking and understanding information across several related documents, rather than just one at a time.",
  "what_it_does": "This technology improves how AI systems answer questions by understanding connections between different documents. When you ask a question, the system first finds several related documents. It then feeds these documents into a special AI model called a transformer, which is designed to 'pay attention' to important words and phrases. Crucially, this model propagates attention from words in a first document to words in a second document (claim 1), building a richer understanding. For example, if you ask about a person, and one document mentions their birth city while another mentions their current job, the AI can connect these details by letting the information flow between the documents. This process creates 'contextualized embeddings' for each word, allowing the AI to select the most relevant words to form an accurate answer to your query (claim 1).",
  "what_it_does_not_cover": [
    "Does not cover systems that do not use a transformer-based machine learning model for propagating attention between documents.",
    "Does not cover methods that only propagate attention within a single document, without explicitly linking information between separate documents.",
    "Does not cover systems where the identification of related documents is not based on similarity to the query or explicit links between documents (claim 2).",
    "Does not cover AI models that retrieve answers without first processing tokens from multiple result documents through stacked network layers (claim 1).",
    "Does not cover systems that fail to propagate attention between the query's own words and the words within the retrieved documents (claim 5, 6)."
  ],
  "filed": "2024-05-10",
  "granted": "2026-06-16",
  "expires": "2044-05-10",
  "status": "active",
  "holder": "Microsoft Technology Licensing",
  "holder_url": "https://patentbrief.org/company/microsoft-technology-licensing",
  "inventors": [
    {
      "name": "Paul Nathan Bennett",
      "url": "https://patentbrief.org/inventor/paul-nathan-bennett"
    },
    {
      "name": "Chen Zhao",
      "url": "https://patentbrief.org/inventor/chen-zhao"
    },
    {
      "name": "Xia Song",
      "url": "https://patentbrief.org/inventor/xia-song"
    },
    {
      "name": "Saurabh Kumar Tiwary",
      "url": "https://patentbrief.org/inventor/saurabh-kumar-tiwary"
    },
    {
      "name": "Corbin Louis ROSSET",
      "url": "https://patentbrief.org/inventor/corbin-louis-rosset"
    },
    {
      "name": "Chenyan XIONG",
      "url": "https://patentbrief.org/inventor/chenyan-xiong"
    }
  ],
  "times_cited": 0,
  "tags": [
    "software",
    "ai_ml",
    "telecommunications",
    "consumer_electronics"
  ],
  "abstract": "This document relates to natural language processing using a framework such as a neural network. One example method involves obtaining a first document and a second document and propagating attention from the first document to the second document. The example method also involves producing contextualized semantic representations of individual words in the second document based at least on the propagating. The contextualized semantic representations can provide a basis for performing one or more natural language processing operations.",
  "url": "https://patentbrief.org/patent/us/12657244/inter-document-attention-mechanism-12657244",
  "markdown_url": "https://patentbrief.org/patent/us/12657244/inter-document-attention-mechanism-12657244/md",
  "google_patents_url": "https://patents.google.com/patent/US12657244",
  "relatedPatents": [
    {
      "patentNumber": "20240338414",
      "countryCode": "US",
      "title": "How AI Understands Answers Across Multiple Documents",
      "url": "https://patentbrief.org/patent/us/20240338414/inter-document-attention-mechanism"
    },
    {
      "patentNumber": "10664744",
      "countryCode": "US",
      "title": "How Computers Use Memory Networks to Answer Questions",
      "url": "https://patentbrief.org/patent/us/10664744/watson-question-answering-system-deepqa"
    },
    {
      "patentNumber": "10452978",
      "countryCode": "US",
      "title": "How AI Models Understand Language Using 'Attention'",
      "url": "https://patentbrief.org/patent/us/10452978/transformer-attention-mechanism"
    },
    {
      "patentNumber": "11798529",
      "countryCode": "US",
      "title": "Improving AI Language Understanding with Aligned Knowledge Graphs",
      "url": "https://patentbrief.org/patent/us/11798529/generation-of-optimized-knowledge-based-language-model-through-knowledge-graph-m"
    },
    {
      "patentNumber": "11062228",
      "countryCode": "US",
      "title": "How AI Learns New Tasks Using Old Data Labels",
      "url": "https://patentbrief.org/patent/us/11062228/gpt-3-few-shot-learning"
    }
  ]
}