Session Details
[2F5-OS-19a]OS-19
Tue. Jun 9, 2026 3:30 PM - 5:00 PM JST
Tue. Jun 9, 2026 6:30 AM - 8:00 AM UTC
Tue. Jun 9, 2026 6:30 AM - 8:00 AM UTC
Room F(Main hall B)
オーガナイザ:中田 百科(株式会社リクルート),山内 敏嗣(Sansan株式会社)
[2F5-OS-19a-01]Context-Aware Page-Level Image Classification of Multi-Page Invoices Using GIT
〇Sinyu Lai1, Yutaro Honda1 (1. Sansan, Inc.)
[2F5-OS-19a-02]Design of a Search-Based Multimodal Architecture for Pharmaceutical Tablet Identification
〇Taisei Takeda1, Tomoyuki Higuchi1 (1. Chuo University)
[2F5-OS-19a-03]Lost in the Files: A Comparison of Long-Context LLMs and RAG for Comprehensive Information Extraction from Multiple Specialized Documents
〇Shota Sato1, Kei Furukawa2, Kazuya Ikoma2, Atom Sonoda1 (1. Lightblue KK, 2. Shimizu Corporation)
[2F5-OS-19a-04]A RAG System for Technical Document Retrieval Based on Figure and Table Understanding
〇Ryoya Shiraiwa1, Hiroki Yamada1, Takayoshi Fujioka2 (1. Hitachi, Ltd., 2. Hitachi Industrial Equipment Systems Co., Ltd.)
[2F5-OS-19a-05]Improving Vision-Language-Model-Based Anomalous Frame Detection Using Text Prompts Toward Codifying Expert Know-How
Kaname Yokoyama1, 〇Ryo Sakai1 (1. Hitachi, Ltd. Research and Development Group)
[2F5-OS-19a-06]Multimodal Data Knowledge Graph Integration to Enhance Reliability of Equipment Maintenance Support
〇Tatsuya Baba1, Takumi Uezono1, Kentarou Yoshimura1 (1. Hitachi Ltd.)
[[online]]
