Wedding Park's Joint Research Paper with CyberAgent Group's AI Organization Accepted at JSAI 2026 National Conference
A joint research paper by Wedding Park and CyberAgent's AI labs on a new technological approach for securely handling personal information in document images has been accepted. This is the company's first paper acceptance and a pioneering case in the wedding industry.
📋 Article Processing Timeline
- 📰 Published: April 8, 2026 at 20:00
- 🔍 Collected: April 8, 2026 at 11:31
- 🤖 AI Analyzed: April 20, 2026 at 18:24 (294h 52m after Collected)
Wedding Park Co., Ltd. (Headquarters: Minato-ku, Tokyo; President: Seigo Hishiki) is pleased to announce that its joint research paper on AI utilization has been accepted at the 2026 Annual Conference of the Japanese Society for Artificial Intelligence (JSAI2026).
This research was conducted jointly with 'AI Lab'*¹, which handles AI research within the CyberAgent group, and 'MIU AI Strategy Headquarters'*², presenting new findings on a technological approach for securely handling personal information within document images.
This also marks the first time a paper from our company's research activities has been accepted. Furthermore, technological initiatives in the image analysis field are still scarce in the wedding industry, positioning this as one of the pioneering examples.
'AI Lab'*¹: An AI research organization established in 2016, involved in all aspects of the digital marketing domain.
'MIU AI Strategy Headquarters'*²: Established in 2025, an organization promoting function development and operational efficiency using generative AI in the Media & IP business.
**■ Our AI Utilization and the Accepted Paper**
Our company actively promotes AI utilization in both internal operations and the services we provide.
In addition to our AI R&D efforts, we have been strengthening our organizational structure, including the use of generative AI, by establishing the 'Generative AI Promotion Office,' a first* in the wedding industry.
As part of these efforts, we have also strengthened collaboration with the CyberAgent group's AI-specialized organizations, working to advance AI utilization rooted in practical applications.
Through this collaboration, we gained new academic insights, leading to the achievement of having a paper accepted at a domestic conference.
This research focused on advancing the series of data processing steps (masking personal information, transcription, analysis of expense items, etc.) for cost statement images posted on our review information site.
Previously, this was a process where achieving both accuracy and speed was difficult, as it required manual checking and processing of vast amounts of documents and text information. We have promoted automation using OCR (Optical Character Recognition) technology and natural language models. Within this, we experimentally demonstrated the effectiveness of 'table-aware text segmentation' as a preprocessing method that contributes to improving the accuracy of masking target determination.
This achievement is expected to deliver both security and efficiency.
**Paper Title: 'Table-Aware Text Segmentation for Masking Sensitive Information in Document Images'**
*First in the domestic wedding industry as a specialized organization for 'Generative AI' (based on our own research as of July 2025).
Going forward, we will continue to strengthen our collaboration with the CyberAgent group, give back the knowledge gained from our research activities to society and the industry, and strive to create value in the wedding x AI domain.
This research was conducted jointly with 'AI Lab'*¹, which handles AI research within the CyberAgent group, and 'MIU AI Strategy Headquarters'*², presenting new findings on a technological approach for securely handling personal information within document images.
This also marks the first time a paper from our company's research activities has been accepted. Furthermore, technological initiatives in the image analysis field are still scarce in the wedding industry, positioning this as one of the pioneering examples.
'AI Lab'*¹: An AI research organization established in 2016, involved in all aspects of the digital marketing domain.
'MIU AI Strategy Headquarters'*²: Established in 2025, an organization promoting function development and operational efficiency using generative AI in the Media & IP business.
**■ Our AI Utilization and the Accepted Paper**
Our company actively promotes AI utilization in both internal operations and the services we provide.
In addition to our AI R&D efforts, we have been strengthening our organizational structure, including the use of generative AI, by establishing the 'Generative AI Promotion Office,' a first* in the wedding industry.
As part of these efforts, we have also strengthened collaboration with the CyberAgent group's AI-specialized organizations, working to advance AI utilization rooted in practical applications.
Through this collaboration, we gained new academic insights, leading to the achievement of having a paper accepted at a domestic conference.
This research focused on advancing the series of data processing steps (masking personal information, transcription, analysis of expense items, etc.) for cost statement images posted on our review information site.
Previously, this was a process where achieving both accuracy and speed was difficult, as it required manual checking and processing of vast amounts of documents and text information. We have promoted automation using OCR (Optical Character Recognition) technology and natural language models. Within this, we experimentally demonstrated the effectiveness of 'table-aware text segmentation' as a preprocessing method that contributes to improving the accuracy of masking target determination.
This achievement is expected to deliver both security and efficiency.
**Paper Title: 'Table-Aware Text Segmentation for Masking Sensitive Information in Document Images'**
*First in the domestic wedding industry as a specialized organization for 'Generative AI' (based on our own research as of July 2025).
Going forward, we will continue to strengthen our collaboration with the CyberAgent group, give back the knowledge gained from our research activities to society and the industry, and strive to create value in the wedding x AI domain.