WO2025238621 - ACCURACY AND RELIABLE IMPROVEMENTS IN FILE ANALYSIS USING LARGE LANGUAGE MODELS
National phase entry is expected:
Publication Number
WO/2025/238621
Publication Date
20.11.2025
International Application No.
PCT/IB2025/055153
International Filing Date
16.05.2025
Title **
[English]
ACCURACY AND RELIABLE IMPROVEMENTS IN FILE ANALYSIS USING LARGE LANGUAGE MODELS
[French]
AMÉLIORATIONS DE PRÉCISION ET DE FIABILITÉ DANS UNE ANALYSE DE FICHIERS À L'AIDE DE GRANDS MODÈLES DE LANGAGE
Applicants **
ID BUSINESS SOLUTIONS, LTD.
Inventors
GERING, David
SATULOVSKY, Javier
FORMAN, Ken
Priority Data
63/649,142
17.05.2024
US
63/653,080
29.05.2024
US
Application details
| Total Number of Claims/PCT | * |
| Number of Independent Claims | * |
| Number of Priorities | * |
| Number of Multi-Dependent Claims | * |
| Number of Drawings | * |
| Pages for Publication | * |
| Number of Pages with Drawings | * |
| Pages of Specification | * |
| * | |
| Number of Office Actions | * |
| * | |
International Searching Authority |
EPO
* |
| Recordal of a Change of the Applicant's Name/Address |
Change of Applicant's Name and Address
* |
| Type of Assignment |
The Standard Agent's Assignment
* |
| Applicant's Legal Status |
Legal Entity
* |
| * | |
| * | |
| * | |
| * | |
| * | |
| Entry into National Phase under |
Chapter I
* |
| Patent Delivery |
Send the Letters Patent by Courier
* |
| 译文 |
|
* The data is based on automatic recognition. Please verify and amend if necessary.
** IP-Coster compiles data from publicly available sources. If this data includes your personal information, you can contact us to request its removal.
Quotation for National Phase entry
| Country | Stages | Total | |
|---|---|---|---|
| China | Filing, Examination, Granting | 2286 | |
| EPO | Filing, Examination, Granting | 10470 | |
| Japan | Filing, Examination, Granting | 2279 | |
| South Korea | Filing, Examination, Granting | 2356 | |
| USA | Filing, Examination, Granting | 4740 |
Total:
22,131
Contact Us
Abstract[English]
Computer systems and methods for operating a taming protocol are described. Queries are generated based on an initial task, which are then sent to one or more large language models (LLMs). The raw or untamed responses received from the LLMs are compared to identify consistent untamed responses. Once a desired level of consistency is obtained, a tamed response is created based on the compiled contents of the identified consistent responses from the LLMs. The tamed response provides accurate and reliable outputs, enhancing data processing and decision-making. The invention improves efficiency, reduces errors, and can help to ensure compliance in various GxP operations, including file ingestion and searching.[French]
L'invention concerne des systèmes informatiques et des procédés de fonctionnement d'un protocole de domestication. Des interrogations sont générées sur la base d'une tâche initiale, et sont ensuite envoyées à un ou plusieurs grands modèles de langage (LLM). Les réponses brutes ou non domestiquées reçues des LLM sont comparées pour identifier des réponses non domestiquées cohérentes. Une fois qu'un niveau souhaité de cohérence est obtenu, une réponse domestiquée est créée sur la base du contenu compilé des réponses cohérentes identifiées provenant des LLM. La réponse domestiquée fournit des sorties précises et fiables, améliorant le traitement de données et la prise de décision. L'invention améliore l'efficacité, réduit les erreurs et peut aider à assurer la conformité dans diverses opérations GxP, y compris l'ingestion et la recherche de fichiers.