81690 research outputs found
Sort by
成果報告書「農業画像データセット Hokkaido Agriculture Image Dataset の公開」
No full text北海道⼤学とソニーグループ株式会社は、産業創出講座 ソーシャル・イノベーション部⾨ for プラネタリーバウンダリーにて、農業画像AI における⾼コストなデータセット構築作業を省⼒化すべく、北海道の様々な農業シーンを含んだ農業画像データセット Hokkaido Agriculture Image Dataset を構築し、公開したため、報告するHokkaido University and Sony Group Corporation have constructed and released Hokkaido Agriculture Image Dataset including various agricultural scenes in Hokkaido to reduce the high-cost data collection process in agricultural image AI. This dataset is the achivement of Joint research, Social Innovation Division for Planetary Boundary.DatasetはHuggingFaceにてCreative Commons Attribution 4.0ライセンスで公
Cross-Cultural Safety Judgments in Child Environments: A Semantic Comparison of Vision-Language Models and Humans
No full textDespite advances in complex reasoning, Vision-Language Models (VLMs) remain inadequately benchmarked for safety-critical applications like childcare. To address this gap, we conduct a multilingual (English, French, Polish, Japanese) comparison of VLMs and human safety assessments using a dataset of original images from child environments in Japan and Poland. Our proposed methodology utilizes semantic clustering to normalize and compare hazard identification and mitigation strategies. While both models and humans identify overt dangers with high semantic agreement (e.g., 0.997 similarity for ‘scissors’), their proposed actions diverge significantly. Humans strongly favor direct physical intervention (‘remove object’: 64.% for Polish vs. 55.0% for VLMs) and context-specific actions (‘move object elsewhere’: 17.8% for Japanese), strategies that models under-represent. Conversely, VLMs consistently over-recommend supervisory actions (such as ‘Supervise children closely’ or ‘Supervise use of scissors’). These quantified discrepancies highlight the critical need to integrate nuanced, human-like contextual judgment for the safe deployment of AI systems
Investigation of multifaceted approach for diabetes treatments [an abstract of entire text]
No full text配架番号：2889この博士論文全文の閲覧方法については、以下のサイトをご参照ください。https://www.lib.hokudai.ac.jp/dissertations/copy-guides
Targeting WEE1 to improve the therapy of KRASG12C mutant non-small cell lung cancer [an abstract of dissertation and a summary of dissertation review]
No full text配架番号：292
ヒメツリガネゴケの極限環境ストレス耐性とその宇宙環境への適応性に関する研究 [全文の要約]
No full textこの博士論文全文の閲覧方法については、以下のサイトをご参照ください。【担当：理学部図書室】https://www.lib.hokudai.ac.jp/dissertations/copy-guides
