Multimodal Assertion Verification Using Textualized Non-Text Content

Solution Overview

Problem

Existing techniques are inefficient in automatically determining the authenticity of assertion details in contents, particularly those involving non-text elements such as images and videos, which are time-consuming and costly when relying on human experts.

Innovation Solution

An information processing apparatus that includes a textualizing unit to convert non-text elements into text, an extraction unit to identify assertion details, and an authenticity determination unit to evaluate the extracted details using a large language model (LLM).

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If human experts perform fact checking on content assertion details, then determination accuracy is improved, but time consumption and costs increase

Engineering Contradiction:

Improveauthenticity determination accuracyVSAvoidtime consumption

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent creates a virtual copy of the expert fact-checking process by training an AI model on expert verification data. The model learns to replicate expert judgment patterns, enabling automated authenticity determination that mirrors human expert accuracy without requiring actual human experts to perform each verification task manually.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical human expert verification process with an automated AI-based system. The manual fact-checking mechanism is substituted with an algorithmic approach that processes content assertion details automatically, eliminating the time-consuming human intervention while maintaining determination accuracy through sophisticated NLP and verification algorithms.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If human experts perform fact checking on content assertion details, then determination accuracy is improved, but costs increase

Engineering Contradiction:

Improveauthenticity determination accuracyVSAvoidcosts

Core Design Contradiction:

Measurement precisionVSLoss of energy

Solution Approach 1:

The patent creates a virtual copy of the expert fact-checking process by training an AI model on expert verification data. The model learns to replicate expert judgment patterns, enabling automated authenticity determination that mirrors human expert accuracy without requiring actual human experts to perform each verification task manually.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical human expert verification process with an automated AI-based system. The manual fact-checking mechanism is substituted with an algorithmic approach that processes content assertion details automatically, eliminating the time-consuming human intervention while maintaining determination accuracy through sophisticated NLP and verification algorithms.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If automated fact checking is implemented, then time consumption and costs are reduced, but ability to handle non-text elements deteriorates

Engineering Contradiction:

Improveprocessing efficiencyVSAvoidhandling of non-text elements

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements a universal verification system that handles multiple content types (text, images, videos, audio) through a single integrated AI platform. The system uses multimodal capabilities to process different data formats, extracting assertion details from various non-text elements and verifying their authenticity against corresponding textual content, thereby achieving both high productivity and broad adaptability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The patent introduces text as an intermediary element that bridges non-text content and verification processes. The system extracts assertion details from images, videos, and audio, then uses textual representation and comparison to verify authenticity. This intermediary approach enables automated processing of diverse content types while maintaining verification accuracy through consistent text-based validation mechanisms.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20260010721A1Information processing apparatus, coverage content verification apparatus, statement details verification apparatus, verification method, and non-transitory computer-readable storage medium

Publication Date: 2026.01.08 NEC CORP

AI summary

An information processing apparatus includes a textualizing unit for textualizing a non-text element included in a content that is a target for determining authenticity of assertion details, an extraction unit for extracting assertion details of the content from the textualized non-text element, and an authenticity determination unit for determining authenticity of the extracted assertion details. According to this information processing apparatus, it is also possible to avoid making an erroneous determination based on content having erroneous details.