# US 2025/0165718 A1 · Device and method for artificial intelligence reasoning based on language model

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

- Number: US 2025/0165718 A1
- Document ID: US20250165718A1
- Status: Published application
- Kind: A1
- Application number: 18/904,441
- Filed: 2024-10-02
- Published: 2025-05-22
- Earliest claimed: 2023-11-16
- Assignees: Electronics and Telecommunications Research Institute
- Applicants: Electronics and Telecommunications Research Institute
- Inventors: Young-Ae Seo
- Web page: https://patentgrep.com/patent/US20250165718A1

## CPC

- G06F 40/35: Discourse or dialogue representation
- G06F 40/30: Semantic analysis
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Provided are a device and method for artificial intelligence (AI) reasoning based on a language model. The method for AI reasoning based on a language model includes generating a question core summary by inputting a natural language question to a language model, decomposing the natural language question into sub-questions using a reasoning model on the basis of the question core summary and the natural language question, obtaining an answer to the natural language question by solving the sub-questions and synthesizing solutions of the sub-questions.

## Claims

### Claim 1

A method for artificial intelligence (AI) reasoning based on a language model, the method comprising:
inputting a natural language question to a pretrained language model to generate a question core summary;
decomposing the natural language question into sub-questions using a reasoning model on the basis of the question core summary and the natural language question; and
solving the sub-questions and synthesizing solutions of the sub-questions to obtain an answer to the natural language question.

### Claim 2 (depends on 1)

The method of claim 1, wherein the decomposing of the natural language question into the sub-questions comprises decomposing the natural language question into one or more sub-questions, and
each of the sub-questions corresponds to one reasoning operation.

### Claim 3 (depends on 1)

The method of claim 1, wherein the question core summary is a semi-code-style summary.

### Claim 4 (depends on 1)

The method of claim 1, wherein the solutions are in a form of semi-code.

### Claim 5 (depends on 4)

The method of claim 4, further comprising generating a final result by converting the solutions into program code and executing the program code.

### Claim 6 (depends on 5)

The method of claim 5, wherein the generating of the final code further comprises processing unit information.

### Claim 7

A device for artificial intelligence (AI) reasoning based on a language model, the device comprising:
a memory configured to store computer-readable instructions; and
at least one processor configured to execute the instructions,
wherein, when the instructions are executed, the at least one processor generates a question core summary by inputting a natural language question to a pretrained language model, decomposes the natural language question into sub-questions using a reasoning model on the basis of the question core summary and the natural language question, solves the sub-questions, and obtains an answer to the natural language question by synthesizing solutions of the sub-questions.

### Claim 8 (depends on 7)

The device of claim 7, wherein the at least one processor decomposes the natural language question into one or more sub-questions, and
each of the sub-questions corresponds to one reasoning operation.

### Claim 9 (depends on 7)

The device of claim 7, wherein the question core summary is a semi-code-style summary.

### Claim 10 (depends on 7)

The device of claim 7, wherein the solutions are in a form of semi-code.

### Claim 11 (depends on 10)

The device of claim 10, wherein the at least one processor converts the solutions into program code and executes the program code to generate a final result.

### Claim 12 (depends on 11)

The device of claim 11. wherein the at least one processor processes unit information in a process of generating of the final result.