# US 2025/0094787 A1 · Privacy-protective knowledge sharing using a hierarchical vector store

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0094787 A1
- Document ID: US20250094787A1
- Status: Published application
- Kind: A1
- Application number: 18/808,300
- Filed: 2024-08-19
- Published: 2025-03-20
- Earliest claimed: 2023-09-15
- Assignees: Oracle International Corporation
- Applicants: Oracle International Corporation
- Inventors: Karoon Rashedi Nia; Anatoly Yakovlev; Sandeep R. Agrawal; Ridha Chahed; Sanjay Jinturkar; Nipun Agarwal
- Web page: https://patentgrep.com/patent/US20250094787A1

## CPC

- G06N 3/0475: Generative networks
- G06F 21/6218: {to a system of files or objects, e.g. local or distributed file system or database}
- G06N 3/092: Reinforcement learning
- G06N 20/00: Machine learning
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Disclosed herein are various approaches for sharing knowledge within and between organizations while protecting sensitive data. A machine learning model may be trained using training prompts querying a vector store to prevent unauthorized user disclosure of data derived from the vector store. A prompt may be received and a response to the prompt may be generated using the machine learning model based at least in part on the vector store.

## Claims

### Claim 1

A method comprising:
training a machine learning model using training prompts querying a vector store to prevent unauthorized user disclosure of data derived from the hierarchical store;
receiving a prompt;
generating, using the machine learning model, a response to the prompt based at least in part on the vector store; and
wherein the method is performed by one or more computing devices.

### Claim 2 (depends on 1)

The method of claim 1, wherein training the machine learning model comprises:
penalizing the machine learning model based on a first response to a first training prompt comprising sensitive information corresponding to an individual one of the plurality of vector stores; and
rewarding the machine learning model based on a second response to a second training prompt comprising aggregate information corresponding to multiple of the plurality of individual vector stores.

### Claim 3 (depends on 1)

The method of claim 1, wherein the vector store stores one or more vector embeddings.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
generating a query vector based at least in part on the prompt;
comparing the query vector with the one or more vector embeddings;
identify one or more vector similar vector embeddings from the vector store;
wherein the one or more similar vector embeddings having a similarity to the query vector that meets or exceeds a predetermined threshold of similarity; and
wherein the response is generated based at least in part on the one or more similar vector embeddings.

### Claim 5 (depends on 4)

The method of claim 4, further comprising determining the similarity of the one or more similar vector embeddings to the query vector based at least in part on a cosine similarity of the one or more vector embeddings to the query vector.

### Claim 6 (depends on 4)

The method of claim 4, further comprising:
decoding the one or more similar vector embeddings to obtain data related to the prompt; and
wherein the response is generated based at least in part on the data related to the prompt.

### Claim 7 (depends on 6)

The method of claim 6, wherein the data related to the prompt comprises a context associated with the one or more similar vector embeddings.

### Claim 8 (depends on 7)

The method of claim 7, wherein the machine learning model is a first machine learning model, the method further comprising determining, using a second machine learning model, whether the response comprises sensitive information corresponding to an individual one of the plurality of vector stores.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
generating a plurality of individualized responses to the prompt, each of the plurality of individualized responses corresponding to one of the plurality of vector stores;
determining a similarity between the response and each of the plurality of individualized responses;
determining to mask the response based at least in part on the response having a similarity to fewer than a predetermined number of the plurality of individualized responses; and
regenerating, using the machine learning model, the response to the prompt based at least in part on the vector store.

### Claim 10 (depends on 1)

The method of claim 1, wherein the machine learning model comprises a large language model.
training a machine learning model using training prompts querying a vector store to prevent unauthorized user disclosure of data derived from the vector store;
receiving a prompt;
generating, using the machine learning model, a response to the prompt based at least in part on the vector store; and
wherein the method is performed by one or more computing devices.

### Claim 11

One or more non-transitory storage media storing instructions which, when executed by one or more computing devices, cause:
training a machine learning model using training prompts querying a vector store to prevent unauthorized user disclosure of data derived from the vector store;
receiving a prompt; and
generating, using the machine learning model, a response to the prompt based at least in part on the vector store.

### Claim 12 (depends on 11)

The one or more non-transitory storage media of claim 11, wherein training the machine learning model comprises:
penalizing the machine learning model based on a first response to a first training prompt comprising sensitive information corresponding to an individual one of the plurality of vector stores; and
rewarding the machine learning model based on a second response to a second training prompt comprising aggregate information corresponding to multiple of the plurality of individual vector stores.

### Claim 13 (depends on 11)

The one or more non-transitory storage media of claim 11, wherein the vector store stores one or more vector embeddings.

### Claim 14 (depends on 13)

The one or more non-transitory storage media of claim 13, further comprising:
generating a query vector based at least in part on the prompt;
comparing the query vector with the one or more vector embeddings;
identify one or more vector similar vector embeddings from the vector store;
wherein the one or more similar vector embeddings having a similarity to the query vector that meets or exceeds a predetermined threshold of similarity; and
wherein the response is generated based at least in part on the one or more similar vector embeddings.

### Claim 15 (depends on 14)

The one or more non-transitory storage media of claim 14, further comprising determining the similarity of the one or more similar vector embeddings to the query vector based at least in part on a cosine similarity of the one or more vector embeddings to the query vector.

### Claim 16 (depends on 14)

The one or more non-transitory storage media of claim 14, further comprising:
decoding the one or more similar vector embeddings to obtain data related to the prompt; and
wherein the response is generated based at least in part on the data related to the prompt.

### Claim 17 (depends on 16)

The one or more non-transitory storage media of claim 16, wherein the data related to the prompt comprises a context associated with the one or more similar vector embeddings.

### Claim 18 (depends on 11)

The one or more non-transitory storage media of claim 11, wherein the machine learning model is a first machine learning model, the method further comprising determining, using a second machine learning model, whether the response comprises sensitive information corresponding to an individual one of the plurality of vector stores.

### Claim 19 (depends on 11)

The one or more non-transitory storage media of claim 11, wherein performing data masking on the response comprises:
generating a plurality of individualized responses to the prompt, each of the plurality of individualized responses corresponding to one of the plurality of vector stores;
determining a similarity between the response and each of the plurality of individualized responses;
determining to mask the response based at least in part on the response having a similarity to a predetermined number of the plurality of individualized responses; and
regenerating, using the machine learning model, the response to the prompt based at least in part on the vector store.

### Claim 20 (depends on 11)

The one or more non-transitory storage media of claim 11, wherein the machine learning model comprises a large language model.
