How AI Understands Answers Across Multiple Documents
Microsoft's 2024 patent on using AI to find answers by understanding how information in one document relates to information in another.
Original patent title: “Inter-document attention mechanism”
Microsoft's 2024 patent on using AI to find answers by understanding how information in one document relates to information in another. Owned by Microsoft Technology Licensing with 25 claims and 1 forward citation, and it is expected to expire in 2044.
Coverage
What does this patent actually cover?
This patent describes a method for a computer to find answers to a query. It starts by getting a query and then finding several documents that might contain the answer. These documents are broken down into 'tokens' (like words or parts of words). The system then uses a special AI model, called a transformer, to process these tokens. The key part is that the AI learns to pay 'attention' to how tokens in one document relate to tokens in another document. For example, it can see how a word in a first document helps explain a word in a second document. This helps the AI create a better understanding of the information and then pick out the specific tokens that form the answer to the original query, which is then presented to the user. ClaimclaimA numbered sentence at the end of a patent that legally defines what the inventor owns. The most important section.Read more → 21 specifically mentions propagating attention from first tokens of a first result document to second tokens of a second result document.
The gap
What does this patent NOT cover?
- Methods that do not use a transformer-based machine learning model.
- Systems that only process a single document to find an answer.
- Methods that do not involve propagating attention between tokens from different documents.
- Systems that do not output an answer in response to a query.
- AI models that do not generate contextualized semantic representations of words.
These exclusions are unique to PatentBrief — derived from the actual claim language, not patent-office boilerplate.
Key facts
What made this novel
The innovation lies in teaching an AI to understand how information in one document 'attends' to or relates to information in another document. This allows the AI to build a richer, interconnected understanding of the data, leading to more precise answer extraction.
The Patent Drawing
Schematic visualization of the patent's claim structure. Hand-drawn diagrams in progress for each landmark patent.
Where you've seen this
Real-world examples
AI-powered search engines
Virtual assistants like Cortana
Document analysis tools
Information retrieval systems
Why it matters
The bigger picture
This technology is crucial for modern search engines and AI assistants that need to synthesize information from multiple sources. It enables systems to go beyond simply finding documents and instead understand the relationships between pieces of information across those documents to provide more direct and accurate answers.
Filed
May 10, 2024
Market context
Who's building on this
Companies in this space
Microsoft, the assigneeassigneeThe entity that owns the patent — usually the inventor's employer or a company.Read more →, is actively developing AI technologies that leverage advanced natural language processing. Companies like Google and OpenAI are also heavily invested in similar inter-document attention mechanisms for their search and large language models.
Market impact
This patent is part of a broader trend in AI development focused on improving information synthesis. It contributes to the evolution of search engines and AI assistants from simple document retrievers to sophisticated answer engines capable of understanding complex relationships within data.
Claim 1 — Plain English
What this patent covers
This patent describes a method for a computer to find answers to a query. It starts by getting a query and then finding several documents that might contain the answer. These documents are broken down into 'tokens' (like words or parts of words). The system then uses a special AI model, called a transformer, to process these tokens. The key part is that the AI learns to pay 'attention' to how tokens in one document relate to tokens in another document. For example, it can see how a word in a first document helps explain a word in a second document. This helps the AI create a better understanding of the information and then pick out the specific tokens that form the answer to the original query, which is then presented to the user. Claim 21 specifically mentions propagating attention from first tokens of a first result document to second tokens of a second result document.
The clever bit
The innovation lies in teaching an AI to understand how information in one document 'attends' to or relates to information in another document. This allows the AI to build a richer, interconnected understanding of the data, leading to more precise answer extraction.
What it does not cover
- Methods that do not use a transformer-based machine learning model.
- Systems that only process a single document to find an answer.
- Methods that do not involve propagating attention between tokens from different documents.
- Systems that do not output an answer in response to a query.
- AI models that do not generate contextualized semantic representations of words.
Patent timeline
Application submitted to the patent office
Patent enters public domain
PatentBrief Score
Impact Score
Moderate
Citation count
6/40
Early citations
Claim breadth
17/20
Very broad protection
Recency
0/20
Older than 20 years
Assignee scale
20/20
Major company or institution
PatentBrief Impact Score — based on citation count, claim breadth, recency, and assignee scale. Not a legal assessment.
Heuristic Value Estimate
What this patent might be worth
$62K – $200K
Midpoint $125K · 17.7 yr remaining · industry ×1.6
Heuristic only — blends forward/backward citation counts, claim scope, time remaining, litigation history, and CPC-derived industry baseline. Real valuations need a professional appraisal.
Claim text not yet imported for this patent
The original legal language
Original claims
25 claims as filed with the patent office.
Concepts involved
Citations
Patent lineage
Cite this patent
Bennett, P. N., Zhao, C., Song, X., Tiwary, S. K., ROSSET, C. L., & XIONG, C. How AI Understands Answers Across Multiple Documents (U.S. Patent No. 20,240,338,414). U.S. Patent and Trademark Office. https://patentbrief.org/patent/us/20240338414/inter-document-attention-mechanism
Auto-generated from the patent record. Double-check author order and the issue date against the official USPTO document before submitting.
Embed
Add this patent to your site
Drop this plain-English patent card into any blog post or article — free, no signup. It always links back to the full breakdown here.
<div data-patentlens-widget data-patent-number="US20240338414"></div> <script src="https://patentbrief.org/embed.js" async></script>
Stay in the loop
Get a weekly digest of new patents.
One email per week. No spam. Unsubscribe anytime.
Keep exploring
Related patents you should know
US 4683195 · 1987
How to Make Billions of Copies of a DNA Segment
This patent describes the Polymerase Chain Reaction (PCR), a method to rapidly create many copies of a specific piece of DNA or RNA, enabling its detection and analysis.
Cetus Corp
US 8697359 · 2014
How to Edit Genes in Human Cells Using an Engineered CRISPR System
This patent describes an engineered CRISPR-Cas9 system for precisely cutting DNA in eukaryotic cells to change how genes work, opening the door for gene editing in complex organisms.
Massachusetts Institute of Technology
US 7657849 · 2010
How the iPhone's Slide-to-Unlock Gesture Works
Apple's 2010 patent describes unlocking a device by dragging a specific graphical image across the touchscreen along a predefined path, a gesture that became iconic with the original iPhone.
Apple Inc
US 4733665 · 1988
How Doctors Implant a Permanent Stent Using a Balloon
This patent describes the method for placing a permanent, expandable wire mesh tube inside a blood vessel or other body tube using a balloon-tipped catheter to widen it and keep it open.
Expandable Grafts Partnership
US 4965188 · 1990
How to Make Many Copies of a DNA Piece with Heat
This patent describes the Polymerase Chain Reaction (PCR) method, a technique to make millions of copies of a specific DNA segment using a heat-resistant enzyme and repeated temperature changes.
Cetus Corp
US 4235871 · 1980
How to Encapsulate Active Materials in Lipid Bubbles Efficiently
This patent describes a method for trapping biologically active substances inside tiny, multi-layered fat bubbles called liposomes, using a specific water-in-oil emulsion and gel-forming process to improve how much material gets captured.
Individual
Semantically similar
You might also find these interesting
US 10664744 · 2020 · Facebook Inc
How Computers Use Memory Networks to Answer Questions
US 10452978 · 2019 · Google LLC
How AI Models Understand Language Using 'Attention'
US 6523026 · 2003 · Huntsman International LLC
How Computers Find Hidden Connections Between Different Fields of Knowledge
US 11062228 · 2021 · Microsoft Technology Licensing LLC
How AI Learns New Tasks Using Old Data Labels
More to explore
More in Software & Internet
US 4405829 · 1983 · Massachusetts Institute of Technology
How RSA Public-Key Encryption Keeps Digital Messages Secret
US 6285999 · 2001 · Leland Stanford Junior University
How Websites Get Ranked by Importance
US 5960411 · 1999 · Amazon com Inc
How Amazon's One-Click Ordering Works for Online Purchases
US 7669123 · 2010 · Facebook Inc
Displaying Friends' Activities in a Social Network Feed
New to patents?
Common Questions
Frequently Asked Questions
What does How AI Understands Answers Across Multiple Documents cover?
Microsoft's 2024 patent on using AI to find answers by understanding how information in one document relates to information in another.
Who owns patent US 20240338414?
This patent is owned by Microsoft Technology Licensing.
When does this patent expire?
This patent is expected to expire on May 10, 2044, when the invention enters the public domain.
What is patent US 20240338414 cited by?
This patent has been cited by 1 later patents that build on its ideas.
What problem does this patent solve?
This technology is crucial for modern search engines and AI assistants that need to synthesize information from multiple sources. It enables systems to go beyond simply finding documents and instead understand the relationships between pieces of information across those documents to provide more direct and accurate answers.
What does this patent NOT cover?
Methods that do not use a transformer-based machine learning model.
Same assignee
More from Microsoft Technology Licensing
Patent monitoring